text
stringlengths 0
330k
| metadata
stringlengths 459
10.9k
|
|---|---|
Pilot study evaluating the efficacy of a topical formulation containing emodepside and praziquantel in the treatment of natural feline troglostrongylosis
Background Troglostrongylus brevior, a lungworm usually affecting wild felids, has been recently recorded in a number of cases in domestic cats, mainly in Mediterranean areas. Although feline troglostrongylosis is a severe and life-threatening disease, especially in young cats, treatment options are very limited. The present study evaluated the efficacy and safety of a spot-on formulation containing emodepside 2.1% and praziquantel 8.6% (Profender®, Bayer), which is licensed for treatment of the more common cat lungworm Aelurostrongylus abstrusus, for the treatment of natural troglostrongylosis. Methods Sixteen cats enrolled in the study were 1:1 allocated to two groups, i.e. Group T, treated with Profender® spot-on on days 0 and 14 (± 2) at the recommended clinical dose, and Group C which remained untreated. After study completion, the control cats received two rescue treatments with Profender® on days 28 (± 2) and 42 (± 2). The primary efficacy criterion was the absence of T. brevior L1 following treatment. Other efficacy parameters were the quantitative comparison of L1 presence before (baseline) and after treatment in both groups, and the comparison of clinical signs pre- and post-treatment. Results In terms of stopping larval shedding, Profender® showed an efficacy of 97% and 97.5% (arithmetic and geometric means, respectively) for group T, 97.1% and 98.5% for group C after one administration, and 100% for both groups after two doses. Overall, 12 cats showed clinical signs related to T. brevior. Specifically, 9 were clinically affected before treatment while clinical signs appeared after the first treatment in 3 cats. At the end of the study, all symptomatic cats fully recovered with the exception of 3 cats that showed clinical signs similar to those observed at the pre-treatment examination at the end of the study. Conclusions This study shows that Profender® is effective against T. brevior.
Background
Feline troglostrongylosis is an emerging gastropodborne disease caused by the metastrongyloid lungworm Troglostrongylus brevior [1]. This parasite usually affects wild felids but recently several cases of infection in domestic cats have been described [2]. Adult stages live in bronchi and bronchioles and, after mating, females produce eggs that hatch releasing first-stage larvae (L1), which migrate to the pharynx, where they are swallowed and then shed in feces [1,[3][4][5]. The life-cycle of T. brevior is similar to that of the globally distributed and wellknown cat lungworm Aelurostrongylus abstrusus and cats become infected by ingesting third-stage larvae (L3) in intermediate hosts, i.e. slugs and snails, or paratenic hosts, i.e. rodents, amphibians, birds and reptiles [3,4]. Additionally, a vertical route of transmission has been described [6][7][8][9]. Although it is not definitively proven how T. brevior is vertically transmitted, it is likely that the infection takes place in the first days after birth, likely via the colostrum or milk [8].
Despite the major pathogenic role of T. brevior, very few therapeutic options are available. Thus far, the only licensed product is a spot-on formulation containing eprinomectin 0.4% in combination with fipronil 8.3%, (S)-methoprene 10% and praziquantel 8.3% (Broadline ® , Merial-Boehringer Ingelheim). In field conditions this product has shown an efficacy of up to 100% [10,14], but it should be taken into account that, when specified by the SPC (summary of product characteristics) it is exclusively indicated when cestodes, nematodes and ectoparasites are to be targeted at the same time.
The spot-on formulation containing emodepside 2.1% and praziquantel 8.6% (Profender ® , Bayer) is efficacious for treating natural [15] and experimental [16] infections caused by the cat lungworm A. abstrusus, and it has recently been labelled for this purpose in the European Union. Furthermore, its efficacy against T. brevior showed to be promising in individual clinical cases, even in mixed infections caused by other respiratory nematodes [8,12,13]. The present pilot study has evaluated for the first time the efficacy of Profender ® in the treatment of feline troglostrongylosis in a case series of naturally infected animals.
Methods
The study was a blinded, randomized, negative-controlled field trial carried out at three sites located in endemic areas of the Umbria (Site A) and Abruzzo (Sites B and C) regions of central Italy.
Pre-inclusion screening
Privately owned cats were enrolled upon informed consent signed by the owner. Individual fecal samples from 165 cats, i.e. 85, 54 and 26 from sites A, B and C, respectively, were collected and tested using the Baermann migration method for the presence of T. brevior L1 on days -30/-7. Larvae were identified as T. brevior ( Fig. 1) according to morphometric and morphological features [4,5] and their identity was genetically confirmed using species-specific PCR [7]. A total of 16 (9.7%) positive cats, 9 (10.6%), 5 (9.3%) and 2 (7.7%) from sites A, B and C, respectively, were included in the study according to inclusion/exclusion criteria.
Inclusion and exclusion criteria
Cats were enrolled in the study according to the inclusion/exclusion criteria as follows. Inclusion criteria: (i) cats in good general health for which the owner signed the owner consent form; (ii) copromicroscopic detection of L1 of T. brevior in at least one Baermann examination performed between days -30 and -7, whose identity has been confirmed by PCR. Exclusion criteria: (i) cats treated with a macrocyclic lactone or other anthelmintics with a systemic biodistribution within 2 months before the study; (ii) cats affected by concomitant parasitic respiratory infections; (iii) cats less than 8 weeks-old; (iv) cats weighing less than 0.5 kg; (v) pregnant or lactating All 16 cats fulfilled the inclusion criteria and on days -7/0 were clinically and copromicroscopically examined with a quantitative Baermann test to assess values of T. brevior larvae per gram of feces (LPG).
Treatment and post-treatment evaluation
On day 0 each cat underwent a clinical examination, was weighed and randomly assigned to the treatment (T) or to the control (C) Group. Cats of Group T (n = 8 cats) were treated with Profender ® spot-on on days 0 and 14 (± 2) at the recommended clinical dose, while cats of Group C remained untreated and, after study completion, received two rescue treatments on days 28 (± 2) and 42 (± 2).
Larval counts were again performed post-treatment on days 14 (± 2) and 28 (± 2) for Group T and on days 42 (± 2) and 56 (± 2) for Group C. On day 0 and the same day of each copromicroscopic test all cats underwent a physical examination to evaluate the presence of clinical signs associated with troglostrongylosis. The cats were also observed daily by their owners for the entire period of the study.
Clinical examination
Clinical procedures were performed by the veterinarian in charge for each of the study sites. The presence of respiratory distress and other signs (listed in Table 2) was documented on a clinical examination form using an evaluation grid based on scores for each change in order to provide parameters that were as objective as possible. A total clinical score was calculated for each cat based on the sum of the different scores on days 0, 14 (± 2) and 28 (± 2) (Group T) and 28, 42 (± 2) and 56 (± 2) (Group C).
Efficacy evaluation
The primary efficacy criterion was the presence/absence of T. brevior after treatment. The presence of the nematode was defined on day 28 (± 2) (post-treatment) according to the detection of L1 in Baermann tests of Group T.
The following efficacy criteria were also evaluated: (i) Statistical comparison of the LPG values between pre-treatment (baseline) and post-treatment copromicroscopic evaluations within Group T, and between the pre-treatment and post-treatment copromicroscopic evaluations within Groups T and C, according to the formula: where Mean LPG was calculated as arithmetic and geometric means. (ii) Comparison of qualitative and quantitative presence of L1 before (baseline) and after the rescue treatment in Group C, according to the above formula. (iii) Post-treatment clinical evaluation of clinically affected cats within Groups T and C, compared with pre-treatment clinical evaluations.
Results
All cats included in the efficacy evaluation of Profender ® completed the study according to protocol and none of them showed any adverse event.
Qualitative efficacy
Overall, 7 out of 8 cats from both Groups T and C (87.5%) were negative in the Baermann examination on days 14 (± 2) and 42 (± 2) after a single administration of Profender ® , respectively. The second administration of the drugs guaranteed cessation of larval excretion in the remaining two cats (100%), i.e. 1 from group T on day 28 (± 2) and 1 from group C on day 56 (± 2).
Clinical outcome
On day 0, 6 out of 8 cats (nos. 3, 6, 7, 12, 13 and 16) of Group T showed clinical signs associated with troglostrongylosis (Table 3). Among them, 4 (nos. 3, 6, 7 and 13) completely recovered after the first treatment. Two cats (nos 12 and 16) showed a temporary worsening in the clinical status on day 14 (± 2) while their clinical score was similar to the pre-treatment evaluation after the second administration of Profender ® . Specifically, cat no. 12 showed bronchovesicular sounds at day 0, bronchovesicular sounds, cough, tachypnoea, lethargy, and nasal and ocular discharge at day 14 (± 2), and cough and ocular and nasal discharge at day 28 (± 2). Cat no. 16 showed tachypnoea, dyspnea and bronchovesicular sounds at day 0 while at day 14 (± 2) pale mucous membranes were also evident and, at day 28 (± 2), the clinical presentation was similar to the pre-treatment evaluation. One cat in group T (no. 2) was apparently healthy at day 0, then its health started to worsen on day 14 (± 2) (i.e. tachypnoea was observed) but fully recovered on day 28 (± 2). One cat (no. 8) was apparently healthy throughout the study. With regard to control cats, before treatment (day 28 ± 2), 3 (nos 4, 10 and 11) had clinical signs related to T. brevior (Table 3). In particular, one (no. 11) completely recovered after the first administration of Profender ® while another (no. 4) completely recovered after the second treatment. The clinical signs in the third (no. 10), characterized by ocular and nasal discharge before treatment, worsened after the first administration (i.e. pale mucous membranes, hyperthermia, tachypnoea and dyspnea were recorded at day 42 ± 2), while the clinical score was similar to the pre-treatment evaluation from the second administration until the study end.
Two cats in Group C (nos 1 and 9), that were apparently healthy before treatment, showed a worsened health status on day 42 (± 2) (i.e. tachypnoea, dyspnea and nasal and ocular discharge were observed in cat no. 1 and pallor of mucosae and hyperthermia were recorded in cat no. 9) and recovered completely after the second rescue treatment. Three cats in group C (nos 5, 14 and 15) were apparently healthy throughout the study.
Discussion
Nematodes of the genus Troglostrongylus have, for a long time, been considered as only affiliated to wild felids [4,5] but, in the last decade, reports have documented a possible spread of T. brevior in domestic cats of Mediterranean and eastern regions, i.e. Italy, Greece, Cyprus, Spain and Bulgaria [10,[17][18][19][20].
Feline troglostrongylosis poses important challenges in feline medicine practice. Clinical diagnosis is impossible due to the overlapping clinical signs with aelurostrongylosis and other common diseases of cats [12,21]. Furthermore, the Baermann migration test may have some shortcomings (e.g. false negative results are possible during prepatency and/or due to intermittent larval shedding, repeated examinations are recommended, L1 identification requires a skillful operator) [2,4]. Once a definitive diagnosis is obtained, efficacious and timely treatment is crucial to save the life of the infected cat, especially in the case of young animals [2,3,7,9].
The present study showed that one or two administrations of Profender ® (2 weeks apart) are highly effective and safe in treating troglostrongylosis, as no adverse effects occurred after administration. Indeed, the worsening of the health status of some of the cats in this study is probably due to death of the nematodes with a subsequent inflammatory host response leading to acute signs. This suggests that a concomitant administration of antiinflammatory drugs could be of benefit, especially in severely infected cats. Further clinical studies are warranted to elucidate this aspect.
The results of this study fit with those of published clinical cases where Profender ® has been used for treating single cats with troglostrongylosis and also in mixed infections with other respiratory nematodes [12,13]. Thus, Profender ® can be considered potentially effective for treating feline troglostrongylosis in feline clinical practice, although further studies are necessary to confirm its efficacy in larger cohorts of animals. The Profender ® spot-on label (as also Broadline ® ) claims treatment for A. abstrusus, thus it could be successfully used in treating mixed infections by both A. abstrusus and T. brevior, which are fairly common [7,12,13]. Profender ® treats various round-and tapeworms and the use is not restricted to cases where the cats are at risk of co-infections with ectoparasites, thus allowing a broader use when helminth treatment alone is indicated.
On the other hand, spot-on eprinomectin is 100% efficacious against L4 and adult T. brevior after a single application [10,14,22], while the activity of spot-on emodepside has been thus far shown only against adult stages. Although this product should be given twice, 2 weeks apart to achieve a 100% efficacy, a very high reduction of LPG is already achieved after one administration against A. abstrusus [16] and T. brevior ([12], present results).
The use of spot-on emodepside is safe in kittens aged ≥ 8 weeks, which are at frequent risk of T. brevior infection [2,10,13,20]. No studies have, however, investigated the safety of Profender ® in younger kittens, but in a recent study two administrations of the product administered off label by a veterinary practitioner were efficacious and safe in treating troglostrongylosis parasitologically and clinically in a kitten aged ≤ 8 weeks [8]. In another study the product was applied to kittens at 4 weeks of age, without any adverse effects being reported [23]. If one considers the frequent vertical transmission of T. brevior, this product could be potentially used to prevent potential lactogenic infection in kitten litters [8], because the formulation is safe in pregnant and lactating queens and effective in the prevention of the vertical transmission of Toxocara cati [23,24].
Conclusions
For its safety and efficacy, Profender ® can be considered a suitable choice for the treatment of natural feline troglostrongylosis.\===
Domain: Biology Medicine. The above document has 2 sentences that start with 'The presence of', 2 sentences that start with '3, 6, 7', 2 sentences that end with 'clinical signs related to T', 2 sentences that end with 'day 28 (± 2)', 2 sentences that end with '(± 2) (i.e', 2 sentences that end with 'in cat no', 2 paragraphs that start with 'On day 0', 2 paragraphs that end with 'treatment of natural feline troglostrongylosis'. It has approximately 2403 words, 125 sentences, and 36 paragraph(s).
|
[{"added": "2019-03-14T16:35:38.875Z", "created": "2019-03-12T00:00:00.000", "id": "76659966", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://parasitesandvectors.biomedcentral.com/track/pdf/10.1186/s13071-019-3361-7", "pdf_hash": "ff4bf6275d18fb682f3fc1cfa7221e485dbdc3c9", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39791", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "ff4bf6275d18fb682f3fc1cfa7221e485dbdc3c9", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Isolation and co-culture of rat parenchymal and non-parenchymal liver cells to evaluate cellular interactions and response
The liver is a central organ in the human body, and first line of defense between host and external environment. Liver response to any external perturbation is a collective reaction of resident liver cells. Most of the current in vitro liver models focus on hepatocytes, the primary metabolic component, omitting interactions and cues from surrounding environment and non-parenchymal cells (NPCs). Recent studies suggest that contributions of NPCs are vital, particularly in disease conditions, and outcomes of drugs and their metabolites. Along with hepatocytes, NPCs–Kupffer (KC), sinusoidal endothelial (LSEC) and stellate cells (SC) are major cellular components of the liver. Incorporation of primary cells in in vitro liver platforms is essential to emulate the functions of the liver, and its overall response. Herein, we isolate individual NPC cell fractions from rat livers and co-culture them in a transwell format incorporating primary rat hepatocytes with LSECs, SCs, and KCs. Our results indicate that the presence and contributions of multiple cells within the co-culture capture the interactions between hepatocytes and NPC, and modulates the responses to inflammatory stimulus such as LPS. The isolation and co-culture methods could provide a stable platform for creating in vitro liver models that provide defined functionality beyond hepatocytes alone.
Preparation of co-cultures and functional characterization.
To create a stratified culture condition using hepatocytes, LSECs, KCs, and SCs, we have extended collagen sandwich culture of hepatocytes to incorporate SCs, with LSECs and KCs cultured in a fibronectin-coated transwell ( Fig. 2A). Briefly, hepatocytes were seeded on collagen-coated plates and cultured for 24 hours. Freshly isolated LSECs and KCs (from the same liver for hepatocyte isolation) were seeded onto a fibronectin-coated transwell and cultured for 24 hours. SCs (from a previous isolation and in culture for 7-10 days) were trypsinized, added to a collagen gel pre-mix, and were overlaid on top of the hepatocytes, allowing the gel to form. The transwell containing LSEC and KCs was added on top of the hepatocyte+SC culture in the well. To evaluate the interactions and effect of hepatocytes and NPC in co-culture, control cultures were prepared with A) hepatocyte and B) stellate + (LSEC + Kupffer) cells and were maintained in hepatocyte maintenance media (see methods). In the prepared co-cultures, 0.5 M Hepatocytes, 50,000 SCs, 1 M LSEC and 0.15 M KCs were added with a hepatocyte:SC:LSEC:KC ratio at 10:1:20:3 cells. To assess the function of hepatocytes within the co-cultures, albumin, urea, and lactate production were monitored in the cell culture media. Time-course secretion from various culture configurations on Day 1, 3 and 6 days of co-culture are shown (Fig. 2B-D). Albumin and urea production was similar in hepatocyte and hepatocyte-NPC cultures up to 7 days, while a higher lactate in the cell culture media was observed in hepatocyte-NPC cultures.
Live/dead microscopy and cellular characterization of co-cultures. Cell viability (calcein AM and ethidium homodimer staining) and morphology of the co-cultures was assessed using fluorescence and phase imaging ( Fig. 3A-E). Hepatocyte monocultures within collagen sandwich (Fig. 3A), SCs and LSEC + KCs within NPC cultures (Fig. 3B,C), and hepatocyte + SCs and LSEC + KCs within hepatocyte-NPC cultures (Fig. 3D,E) showed high viability as visualized by calcein staining. To characterize LSECs and KCs within the co-cultures, LSECs and KCs were stained for SE-1 and CD 163 respectively. LSEC + KCs within hepatocyte-NPC showed high expression of SE-1 and CD 163 on Day 2 of co-culture, while LSECs lost significant expression of SE-1 marker by Day 7 (Figs 3F,G and S3). Similarly, LSEC + KCs within NPC cultures showed retention of CD 163 marker on Day 7 of culture while LSECs lost significant expression of SE-1 marker by Day 7 (Figs 3H and S3).
Inflammatory response of co-cultures to Lipopolysaccharide (LPS) stimulation. To evaluate the inflammatory response, we exposed the co-cultures to an inflammatory stimulus (LPS) and measured their response. Co-cultures maintained up to Day 6 were exposed to 0, 1, 10 and 100 μ g/mL LPS and their cytokine response and secretion into the media was measured (Fig. 4A,B). In addition, the effect on hepatocyte function in various co-culture formats was evaluated (Fig. 4C,D). Within the hepatocyte-NPC co-culture system, incorporation of multiple cell types leads to a modulated inflammatory response. While hepatocytes monocultures and KCs within NPC show an increase in TNF-α release into the media, the response in hepatocyte-NPC is reduced in comparison with NPC cultures (Fig. 4A). In contrast, KC monocultures have lost their functional response within Day 3 of culture ( Figure S2). While the TNF-α response showed a dose and culture-dependent response, IL-10 secretion in both NPC and hepatocyte-NPC cultures was similar (Fig. 4B). Hepatocyte response to LPS exposure, and the effect of secreted inflammatory factors by NPC in hepatocyte-NPC cultures was evaluated using CYP1A1/2 activity and albumin response. We observed a decrease of CYP activity and albumin secretion within hepatocyte-NPC cultures, while no significant response was observed in hepatocyte monocultures (Fig. 4C,D).
Discussion
Incorporation of NPCs with hepatocytes in in vitro models is essential for developing platforms that have applications in drug metabolism and disease models. Herein, we demonstrate the successful isolation of individual non-parenchymal cells from rat livers and the assembly of a liver co-culture model comprising of rat hepatocytes Rat livers were extracted from rats and digested using collagenase and cells obtained as a suspension. Hepatocytes were pelleted and enriched using a percoll cushion (45%), and pelleted. The non-parenchymal fraction is obtained as the supernatant from hepatocyte isolation. Stellate cells were isolated using centrifugation (pelleted at 200 × g) and cultured for 7-10 days with 1 passage before use. Sinusoidal Endothelial cells were isolated using a percoll gradient (interface between 50% and 25%). Kupffer cells were isolated using centrifugal elutriation (at 600 × g, 45 mL/min flow rate) and obtained as a pellet. Isolated cells were stained for markers hepatocytes (albumin), stellate (actin), sinusoidal endothelial cells (SE-1) and Kupffer cells (ED-2/CD 163) respectively. Scale bar = 100 μ m. and non-parenchymal cells to mimic the organization within the liver. Our co-culture model (Fig. 2) is based on standard monolayer of hepatocytes cultured in a collagen sandwich, a model which is frequently used in pharmacological industry for drug toxicity screening and mechanistic studies, and which allows for long term culture 6,7,21,35 . Within the rat liver, hepatocytes comprise 60-65%, SCs-8%, LSECs-16%, and KCs-12% cells 14,36,37 . These cell proportions within the liver indicate a hepatocyte:SC:LSEC:KC ratio of 10:1.3:2.7:2. The hepatocyte-NPC co-culture system in this work incorporates the cells with a ratio 10:1:20:3 such that the relative proportion of hepatocytes, SC, and KC is similar to the ratio in vivo. However, our system incorporates greater number of LSECs due to the difficulty associated with maintaining sub-confluent cultures of LSECs. Analysis of cellular secretions indicates that hepatocytes within these co-cultures were stable and retained their functions (Fig. 2). In addition, SCs and KCs showed high viability and retained phenotype in co-culture, while LSECs showed a decrease in SE-1 expression. Culture conditions that promote long term maintenance of LSECs remain challenging. Co-culture with hepatocytes has enabled long term maintenance of LSECs, albeit under specialized condition such as collagen overlay 21 or short range interaction mediated by hepatocyte/fibroblast 38 . The latter study further demonstrated that culture configuration where LSECs were separated by transwell from hepatocytes or hepatocytes/fibroblast failed to promote maintenance of LSECs. We observed the similar lack of maintenance of LSECs in our model where hepatocytes were separated from LSECs by transwell ( Figure S3). Since dedifferentiation of LSECs has been observed under inflammatory microenvironment created by Kupffer cells 39 and KC secrete TNF-α on isolation ( Figure S2), the presence of KC alongside LSECs (in our model) is likely not conducive to maintenance of LSECs.
While KC monocultures have lost the response to LPS exposure, their functionality is retained to an extent within the co-culture models (NPC and hepatocyte-NPC) ( Fig. 3F-H). An interesting observation of the hepatocyte-NPC co-cultures is the mitigated response to LPS, possibly due to the presence of hepatocytes and the interactions with KCs 40,41 . While Kupffer cell monocultures maintain their inflammatory response for a relatively short period of time 24-48 hours, their long-term culture is not stable to elicit any responses to inflammatory stimulus 42 . In the co-cultures, we captured a response to LPS exposure on day 6 of culture in which TNF-α secretion by KC is less in hepatocyte-NPC co-culture than in NPC co-culture (Fig. 4A). This could be due to the potential uptake of TNF-α by hepatocytes 43 through surface receptors resulting in a decrease of TNF-α concentration in hepatocyte-NPC co-culture systems 44 or hepatocyte mediated reduction in TNF-α secretion by KC. In comparison to primary hepatocyte monocultures that have been shown to secrete relatively low amounts of cytokines and TNF-α upon inflammatory challenge 36,42 , which is also observed in our experiments (Fig. 4), incorporation of NPCs enables capture of certain key aspects of the inflammatory response of liver mediated by cytokines. In response to LPS, not only there is secretion of cytokines such as TNF-α and IL-10 in hepatocyte-NPC cultures, but there is also modulation of hepatocyte function as measured by the CYP and albumin activity. Both CYP activity and albumin secretion is decreased post LPS exposure likely due to intercellular interaction occurring between KC, hepatocytes, and the effects of TNF-α and other inflammatory cytokine secretions (Fig. 4C,D). Presently, the hepatocyte-NPC model potentially recapitulated some of the regulatory mechanisms seen in vivo within a co-culture system (Fig. 5) 45,46 . Although only the effect of co-culturing hepatocytes with KCs and other NPCs in regards to TNF-α and IL-10 secretion was shown, exploring other potential interactions of hepatocytes and NPCs in the co-culture system will allow for further investigative work.
Conclusion
In summary, our work is a demonstration of the techniques for achieving a multi-cell primary culture of liver cells to mimic cellular arrangement and long-term culture. This manuscript details the isolation of four main liver cell types as well as their assembly; and that the majority of cell populations within the co-culture retain their functionality and stability. While several in vitro liver models have been proposed, it is becoming increasingly important to incorporate NPCs into in vitro liver models to mimic organ physiology. Such models could lead to increased sensitivity to monitor drug toxicity and function detection with therapeutically relevant concentrations, and in diseased states in particular. For instance, in vitro models can be manipulated to provide varied conditions within the liver, such as fibrotic, inflammatory processes that lead to changes in overall hepatic function, which is lost in a hepatocyte-alone model. Taken together, our results demonstrated that the multi-cellular liver model shown here capture aspects of tissue physiology other in vitro hepatocyte-alone cell models lack. In future, these models can provide information regarding drug metabolism in both healthy and disease states, which is essential in drug screening process. Multi-cellular co-cultures enable understanding the interplay between various cells, however, further questions about the effect of cell-cell communication and effect of cells within same microenvironment is a matter of future investigation.
Materials and Methods
Primary rat hepatocyte isolation. Hepatocytes were obtained from female Lewis rats using two-step collagenase protocol. Two to three month old female Lewis Rats (Charles River Laboratories, Wilmington, MA) weighing 150-175 g were used as a hepatocyte source and were maintained in accordance with National Research Council guidelines. Experimental protocol for cell isolation from rat livers was approved by the subcommittee on research animal care at Massachusetts General Hospital. All procedures were carried out in accordance with the approved guidelines. Using a modification on the two-step collagenase perfusion method 47,48 , which involves purification of cell suspension by means of centrifugation over percoll, we routinely isolated approximately 200-300 million (M) hepatocytes per rat liver with 85-90% viability as evaluated by trypan blue exclusion.
Briefly, freshly isolated rat hepatocytes in culture were fixed in ice-cold Methanol for 20 min followed by incubation with PBS containing 1% BSA and 0.3% Triton-X100 (blocking buffer). Cells were then exposed to 0.2% FITC-anti albumin antibody and DAPI solution in blocking buffer for 1 hour at room temperature, washed with imaging solution and imaged.
Non-Parenchymal Cell Fraction. The rat non-parenchymal cell (NPC) fraction was obtained as the supernatant from the first centrifugation step performed to pellet primary hepatocytes. Typically, 150-200 M cells were obtained as the NPC fraction per isolation. LSEC and Kupffer cells were obtained from a single isolation, using 60-70% NPC for LSEC and the remainder for Kupffer cell isolation. Stellate cells were obtained from a previous isolation and maintained in culture. Stellate cell separation. Stellate cells were separated from NPC fraction using a two-step centrifugation protocol as previously described 31 . Briefly, the NPC fraction was collected in 50 mL conical tubes and centrifuged at 50 × g for 5 min, without brake. The supernatant after the spin was collected in fresh conical tubes and the pellet (containing LSEC and Kupffer cells) was collected separately. To increase the purity of the stellate fraction, 5mL of PBS in the bottom along with pellet was discarded for each spin and only the top fraction was collected. The procedure was repeated at least 3 times or until no pellet was observed. The supernatant was then centrifuged at 200 × g for 10 min, without brake. The supernatant was removed and the pellet was collected and suspended in DMEM (for separating the other NPCs from the same isolation, pellets and supernatants were pooled and centrifuged at 300 × g for 15 min and used). The cells were washed by centrifuging at 200 × g for 10 min with DMEM. Purified stellate cell fraction was plated in a tissue culture treated T-75 flask and incubated at 37 °C, 5% CO 2 . Media was changed every 24 h for the first 3 days after which cells were trypsinized and re-seeded and maintained in DMEM.
Stellate cell staining. Actin. Stellate cells were fixed by incubating with 4% Paraformaldehyde for 10 min at 37 °C followed by incubation with 0.08% Triton X-100 for 10 min at 37 °C. To these cells, a mixture of 1% solution of Alexa 488-Phalloidin and DAPI was added and incubated for 1 hour at 37 °C. Staining solution was replaced with fresh imaging solution and imaged using a fluorescent microscope.
Desmin. Stellate Cells were fixed by incubating with 4% Paraformaldehyde for 10 min at 37 °C. To these cells, a mixture of 1% solution of Alexa 488-anti-desmin antibody (in Imaging solution) and DAPI was added and incubated for 1 hour at 37 °C. Staining solution was replaced with fresh imaging solution and imaged using a fluorescent microscope.
Lipid droplet staining. Lipid droplets within the cells were visualized staining with BODIPY. Briefly, Stellate cells were fixed by incubating with 4% Paraformaldehyde for 10 min at 37 °C. To these cells, a mixture of 1% solution of BODIPY (in Imaging solution) and DAPI was added and incubated for 1 hour at 37 °C. Staining solution was replaced with fresh imaging solution and imaged using a fluorescent microscope.
Kupffer cell enriched fraction. Kupffer cell enriched fractions were separated from the NPC fraction using a modified centrifugal elutriation protocol 34 . Briefly, the NPC fraction (in 50 mL conical tubes) was centrifuged at 300 × g for 15 min. The supernatant was discarded and the pellet was re-suspended in ice cold PBS. Typically, the NPC fraction from a rat liver was suspended in 20-25 mL PBS (~3-4 × 10 6 cells/mL). The cells were then passed through a cell strainer (40 μ m pore size) to remove any debris, collected in a syringe, and stored on ice.
Cell sample was introduced into the elutriator using a syringe connected to the 3-way valve between the buffer reservoir and pump. Briefly, 10-15 mL of the NPC fraction was loaded into a syringe, avoiding any bubble formation. The syringe end of the 3-way valve was sterilized with ethanol and filled with sterile HBSS. The syringe containing the NPC fraction was attached by liquid-liquid contact, avoiding any bubble formation. Once attached, the cells were introduced into the elutriator by switching the source to the syringe, and shutting down the buffer reservoir. 10 mL of the NPC fraction was introduced into the elutriator (at 10 mL/min) and the source was immediately switched to the buffer reservoir. Cells loaded into the elutriator were washed for 10 min to remove any cell debris, while maintaining the rotor at 600 × g, and flow at 10 mL/min. Contaminating cells (LSECs and stellate) were washed out at 600 × g, 22.5 mL/min. Kupffer cells were eluted at 45 mL/min, 600 × g and 100 mL of the cell suspension was collected and stored on ice. The cells were pelleted at 500 × g for 7 min (no brake), re-suspended in fresh hepatocyte maintenance media and used as an enriched fraction.
Kupffer cell staining. Kupffer Cells were identified using CD 163 (ED-2) Antibody. Briefly, Kupffer cells were incubated with a PBS containing 0.2% Alexa-488 CD 163 antibody and DAPI solution for 1 hour at 37 °C. Staining solution was replaced with fresh imaging solution and imaged using a fluorescent microscope.
Inflammatory response of Kupffer Cells. Kupffer cell enriched fraction was seeded on a fibronectin coated (50 μ g/mL) 24 well plate at 0.15 M cells/well in 500 μ L hepatocyte maintenance media and allowed to attach overnight. To the cells, 1 μ g/mL Lipopolysaccharide (LPS) was added and maintained for 24 hours. At the end of incubation, media was collected and the TNF-α content in the media was measured.
Liver Sinusoidal Endothelial Cell Separation. Liver Sinusoidal Endothelial Cells (LSECs) were separated from NPCs using density separation in a percoll gradient 21,33 . Briefly, NPC fraction in 50 mL conical tubes was centrifuged at 300 × g for 15 min. The supernatant was discarded and the pellet was suspended in ice cold PBS. Typically, the NPC fraction from a rat was suspended in 20-30 mL ice cold PBS.
Percoll Gradient. A percoll gradient was prepared in 50 mL conical tubes with 15 mL of 50% percoll on the bottom and 20 mL of 25% percoll layered on top of the first percoll layer. 10 mL of the NPC fraction was carefully placed on the 25% percoll layer and centrifuged at 900 × g for 25 min, without brake. At the end of the Scientific RepoRts | 6:25329 | DOI: 10.1038/srep25329 centrifugation, the layer between 10-17.5 mL of the percoll gradient is collected and diluted in ice cold PBS (to 50 mL) and centrifuged at 900 × g for 25 min without brake to pellet the cells. Any remaining percoll was aspirated and the cells were suspended in fresh hepatocyte culture media (10 mL). To remove any contaminating cells (Kupffer), the cell fraction was incubated on a 10 cm diameter tissue culture dish for 1-2 min and the non-adherent cells were collected. The tissue culture plate was discarded.
Endothelial cell staining. Endothelial cells were identified by staining for SE-1. Briefly, cells were washed with fresh media and incubated with a solution of 1:1000 dilution of Hepatic Sinusoidal Endothelial Cells Antibody (SE-1) [DyLight 550] ( [URL]) and DAPI stain for 1h at 37 °C. Samples were washed in imaging Solution and imaged using a fluorescent microscope.
Hepatocyte culture media. Hepatocyte culture media was prepared with high glucose (4. Collagen. Type I Collagen was prepared by extracting acid-soluble collagen from Lewis rat-tail tendons as previously reported 49 . Collagen coating was prepared by mixing collagen with PBS (0.1125 mg/mL). Collagen gel was prepared by mixing collagen (1.125 mg/mL) with 10× DMEM resulting in a solution of ~1 mg/mL Collagen. Freshly prepared gel solution is stored on ice and used immediately.
Multiple-cell co-culture preparation. Hepatocyte only culture. Freshly isolated rat hepatocytes were seeded in collagen coated 12 well plates at 0.5 M cells/well in hepatocyte culture media and incubated overnight at 37 °C, 10% CO 2 . Hepatocytes (after overnight incubation) were washed with PBS and overlaid with 200 μ L of collagen gel and incubated for 1 hour at 37 °C, 10% CO 2 . At the end of incubation, 500 μ L of hepatocyte maintenance media was added to the well and an empty transwell with 300 μ L of media was added to the well. Albumin Assay. Albumin concentration in the media was evaluated using an in-house ELISA protocol.
Briefly, 96-well high-binding ELISA plates were coated with 5 μ g/well rat albumin in 100 μ L PBS overnight at 4 °C. The plates were washed with PBS-Tween (0.05%) at least 4 times and 50 μ L of the media or standards were added to the plate. Each plate has a set of standards. Albumin antibody ( [URL], Cat No. 55776) was diluted 1:10000 in PBS-Tween and 50 μ L was added to each well and incubated overnight at 4 °C or for 2 h at 37 °C. At the end of incubation, the plates were washed with PBS-Tween (0.05%) at least 4 times. A substrate solution of o-Phenylenediamine dihydro-chloride (OPD, 400 μ g/mL) and 4 μ M H 2 O 2 solution was prepared in a citric acid buffer. 50 μ L of the solution was added to each well and incubated for 5 min. Reaction was stopped by addition of 50 μ L 8N H 2 SO 4 solution and the absorbance was read at 490 nm. CYP1A1/2 (EROD) Assay. CYP450 1A/2 activity of the co-cultures was evaluated using 7-ethoxyresorufin.
Briefly, cell culture samples were rinsed with Earle's Balanced Salt Solution (EBSS) 3 times at least with 5 min incubations to remove any phenol red from the media and collagen gel. To each of the wells, 500 μ L of substrate (10 μ M 7-ethoxyresorufin + 80 μ M Dicumarol) was added and incubated at 37 °C. 100 μ L of the reagent was withdrawn at 5, 10, 15 and 25 min intervals. Fluorescence from the collected sample was measured at λ ex = 525 ± 10 nm and λ em = 580 ± 10 nm. Rate of resorufin production was calculated by diluting resorufin standard in EBSS.
TNF-α Assay. TNF-α in the cell media was measured using a BD Biosciences (Cat No. 558535) kit as per the manufacturer's protocol. Briefly, high-binding ELISA plates were coated with capture antibody (in coating buffer) overnight at 4 °C. The plates were washed with a washing buffer (0.05% PBS-Tween) 3 times and an assay diluent (PBS with 10% FBS) was added to the wells and incubated for 1 hour at RT. At the end of incubation, the plates were washed with PBS-Tween and 100 μ L of samples or standards were added to the wells and incubated for 2 hours at RT. The plates were then washed with PBS-Tween followed by addition of detection antibody (in assay diluent) and incubated for 1 hour at RT. At the end of incubation, the plates were washed with PBS-Tween and SAv-HRP (in assay diluent) was added and incubated for 1 hour at RT. The plate was then washed thoroughly with PBS-Tween. To each well, 100 μ L TMB substrate was added and incubated for 30 min in dark at RT. The reaction was stopped by addition of 50 μ L ELISA stop solution and the absorbance was measured at 450-570 nm. IL-10 Assay. IL-10 in the cell media was measured using a BD Biosciences (Cat No. 555134) kit as per the manufacturer's protocol. Briefly, high-binding ELISA plates were coated with capture antibody (in coating buffer) overnight at 4 °C. The plates were washed with a washing buffer (0.05% PBS-Tween) 3 times and an assay diluent (PBS with 10% FBS) was added to the wells and incubated for 1 hour at RT. Cell culture media samples were diluted 1:1 in assay diluent to ensure the concentration of samples was in range. At the end of incubation, the plates were washed with PBS-Tween and 100 μ L of samples or standards were added to the wells and incubated for 2 hours at RT. The plates were then washed with PBS-Tween followed by addition of detection antibody + SAv-HRP (in assay diluent) and incubated for 1 hour at RT. The plate was then washed thoroughly with PBS-Tween. To each well, 100 μ L TMB substrate was added and incubated for 30 min protected from light at RT. The reaction was stopped by addition of 50 μ L ELISA stop solution and the absorbance was measured at 450-570 nm.
Lactate Assay. Lactate concentration in media was measured using lactate kit (Trinity Biotech, Cat No. 735-10) using the protocol provided by manufacturer. Briefly, lactate assay reagent was prepared by adding 10 mL distilled water to 1 vial and mixed well. Cell culture media samples were diluted 1:20 (10 μ L sample + 190 μ L distilled water). 10 μ L of the diluted sample media was mixed with 190 μ L of the lactate reagent mix and incubated for 10 min in the dark at room temperature. Absorbance of the samples was read at 540 nm. Statistical analysis. Data was obtained from n ≥ 3 experiments with n = 2 or 3 samples per condition and averaged. Standard error is plotted for all the conditions.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T02:32:03.387Z", "created": "2016-05-04T00:00:00.000", "id": "205695861", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/srep25329.pdf", "pdf_hash": "c973a705ed5b9361664951e7361c24a2d4b124bd", "pdf_src": "SpringerNature", "provenance": "peS2o-0075.json.gz:38015", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "0021408fdba7255f2e589c9c424762158c3bdfd7", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Molecular network, pathway, and functional analysis of time-dependent gene changes associated with pancreatic cancer susceptibility to oncolytic vaccinia virotherapy
Background: Pancreatic cancer is a fatal disease associated with resistance to conventional therapies. This study aimed to determine changes in gene expression patterns associated with infection and susceptibility of pancreatic cancer cells to an oncolyticvaccinia virus, GLV-1h153, carrying the human sodium iodide symporter for deep tissue imaging of virotherapy. Methods: Replication and susceptibility of pancreatic adenocarcinoma PANC-1 cells to GLV-1h153 was confirmed with replication and cytotoxicity assays. PANC-1 cells were then infected with GLV-1h153 and near-synchronous infection confirmed via flow cytometry of viral-induced green fluorescent protein (GFP) expression. Six and 24 hours after infection, three samples of each time point were harvested, and gene expression patterns assessed using HG-U133A cDNA microarray chips as compared to uninfected control. Differentially expressed genes were identified using Bioconductor LIMMA statistical analysis package. A fold change of 2.0 or above was used as a cutoff, with a P value of 0.01. The gene list was then analyzed using Ingenuity Pathways Analysis software. Results: Differential gene analysis revealed a total of 12,412 up- and 11,065 downregulated genes at 6 and 24 hours postinfection with GLV-1h153 as compared to control. At 6 hours postinfection. A total of 139 genes were either up or downregulated >twofold (false discovery rate < 0.05), of which 124 were mapped by Ingenuity Pathway Analysis (IPA). By 24 hours postinfection, a total of 5,698 genes were identified and 5,563 mapped by IPA. Microarray revealed gene expression changes, with gene networks demonstrating downregulation of processes such as cell death, cell cycle, and DNA repair, and upregulation of infection mechanisms (P < 0.01). Six hours after infection, gene changes involved pathways such as HMGB-1, interleukin (IL)-2, IL-6, IL-8, janus kinase/signal tranducer and activator of transcription (JAK/STAT), interferon, and ERK 5 signaling (P < 0.01). By 24 hours, prominent pathways included P53- and Myc-induced apoptotic processes, pancreatic adenocarcinoma signaling, and phosphoinositide 3-kinase/v-akt murine thymoma vial oncogene homolog 1 (PI3/AKT) pathways. Conclusions: Our study reveals the ability to assess time-dependent changes in gene expression patterns in pancreatic cancer cells associated with infection and susceptibility to vaccinia viruses. This suggests that molecular assays may be useful to develop safer and more efficacious oncolyticvirotherapies and support the idea that these treatments may target pathways implicated in pancreatic cancer resistance to conventional therapies.
GLV-1h153 facilitated enhanced uptake in tumors which was readily detected by positron emission tomography.
In this study, we conducted gene expression analysis using cDNA-GeneChip microarray Human Genome U133A (Affymetrix, Santa Clara, CA) to determine changes in gene expression patterns over time associated with infection and susceptibility of pancreatic cancer cells to GLV-1h153. Understanding into the molecular mechanisms associated with sensitivity to GLV-1h153 may enable identification of cancers resistant to viral therapy, thus avoiding undesirable side effects associated with the need for higher doses of viral treatment. Furthermore, knowledge of these mechanisms may be useful to develop safer and more efficacious oncolytic virotherapies.
ReSUltS
GLV-1h153 replication was assessed via flow cytometric detection of GFP GFP expression in cells infected with GLV-1h153 was quantified using flow analysis and was shown to be both time and multiplicity of infection (MOI) dependent. Almost 70% of live cells expressed GFP at an MOI of 5.0 at 24 hours postinfection ( Figure 1a). Viral infection, replication, and cell viability were successfully visualized by assessing GFP expression and were time dependent. Phase overlay pictures shows GFP expression as early as 6 hours postinfection with an MOI of 5, with maximal GFP expression after by 24 hours, and cell death and decline of GFP expression by day 2 (Figure 1b). Based on flow cytometry and visualization of GFP expression, we harvested cells after infection with an MOI of 5 at 0, 6, and 24 hours postinfection. A near-synchronous infection rate was achieved without cell death and lysis occurring too early for harvest.
Identification of time-dependent gene-fold changes After infecting and harvesting our samples, microarray analysis was performed. mRNA from cells were extracted, and using Affymetrix HG-U133A cDNA microarray chips, differentially expressed genes were identified using Bioconductor LIMMA statistical analysis package. A fold change of 2.0 or above was used as a cutoff, with a P value of 0.01. At 6 hours postinfection, a total of 129 genes were either up or downregulated greater than twofold (false discovery rate < 0.05), of which 124 were mapped by Ingenuity Pathway Analysis (IPA). By 24 hours postinfection, a total of 5,698 genes were identified and 5,563 mapped by IPA (Figure 2a; complete microarray data is available in the public Gene Expression Omnibus repository under the accession number GSE48121). The top five genes up-or downregulated at each time point are listed (Table 1). The top five significant molecular and cellular function groups (according to P value) with which common genes were involved entailed roles in cell morphology (11 genes, P = 7.06 × 10 5 ), cellular development (11 genes, P = 2.34 × 10 4 ), cellular movement (13 genes, P = 6.09 × 10 0 ), cellular growth and proliferation (16 genes, P = 6.37 × 10 4 ), and cell-to-cell signaling and interaction (9 genes, P = 6.49 × 10 4 ) (Supplementary Files S1 and S2). Utilizing hierarchical clustering heat map, time-dependent gene changes are visually illustrated, with genes gradually becoming less downregulated by 6 hours to upregulated by 24 hours, and vice versa when compared to 0 hours postinfection (Figure 2b).
Network analysis The IPA software system enables systemic analysis of microarray and other data in a biologic context. Our up-or downregulated genes at each time point were overlaid onto a global molecular network developed from information contained in the Ingenuity Pathways Knowledge Base. Networks of these focus genes were then algorithmically generated based on their interrelationships. Nine major networks were identified by 6 hours, and by 24 hours, more than 25 networks were identified utilizing involved genes at each time point (Figure 3a, Supplementary Files S1). The top network at 6 hours postinfection included genes with functions related to Cell Death and Cellular Development and involved mostly downregulated genes such as Il8, hmox1, bcl3, BIRC3, cxcl2, IRF1, cx3cl1, cdkna (26 genes). At 24 hours post infection, the top network functions involved Gene Expression, Infection Mechanism, and Tumor Morphology and involved up and downregulated genes such as TP53, GSR, TRIO, HSPA1L, PLK2, ABL2 (35 genes) (Figure 3b).
Gene function analysis
We then investigated our overall gene list more closely and analyzed these genes in terms of some important gene and cellular functions. These graphs shows some of the functions deemed associated to our genes. All bars above the line on the graph has a P value less than 0.05. Statistical significance was based on the ratio of up or downregulated genes in our data set to all genes involved in the pathway. At 6 hours, important and statistically significant gene (Figure 4a). Looking at cell death and apoptosis more closely, statistically significant genes at 6 hours involved mainly antiapoptotic pathways, with downregulation of antiapoptotic genes such as BIRC3, and TNF1IP3. By 24 hours, there is shift to modulation of genes involved in proapoptotic mechanisms, such as underexpression of BID, BAX, casp 3, and BAD creating an overall antiapoptotic state ( Figure 4b, Supplementary Files S1 and S2).
Pathway analysis
Canonical pathways were then identified and analyzed from the IPA libraries that were most significant to our common gene data set. The significance of the association between the data set and the canonical pathway was measured in two ways: (i) a ratio of the number of genes from the data set that map to the pathway divided by the total number of genes that map to the canonical pathway is displayed and (ii) Fischer's exact test was used to calculate a P value determining the probability that the association between the genes in the dataset and the canonical pathway is explained by chance alone. IPA identified 52 genes eligible for pathways analysis. Top statistically significant canonical pathways included HMGB-1, interferon, IL-6, and JAK/STAT, signaling at 6 hours postinfection, and by 24 hours prominent pathways included PI3/AKT, epidermal growth factor receptor (EGFR), and extracellular signal-regulated kinases/mitogen-activated protein kinases (ERK/MAPK), signaling pathways (P < 0.05; Figure 5b). Some pathways were activated early and became more active by 24 hours, such as the EGFR pathway, an implicated pathway in pancreatic cancer resistance to conventional therapy ( Figure 5b)
DiSCUSSiON
Vaccinia is a broad-spectrum virus known to infect a wide range of cell types. [5][6][7] Conditionally replicating viruses have been gaining increasing attention for their ability to kill tumor cells by oncolysis and apoptosis, and hence, this attenuated strain has shown promise as a selective anticancer agent. 5 We constructed a new virus which had three nonessential viral genes deleted and replaced with GFP, β-galactosidase, and the hNIS genes, creating GLV-1h153. hNIS is a symporter which facilitated uptake of radioiodine into GLV-1h153-infected cells, therefore enabling imaging of viral replication in vivo utilizing deep tissue imaging modalities such as positron emission tomography and single photon emission computed tomography. 2,8,9 The virus also killed pancreatic cells both in cell culture and in animal models.
However, there remains a need to understand molecular mechanisms related to oncolytic viral infection and sensitivity to treatment. Therefore, this study investigated time-dependent changes in gene expression patterns associated with infection and susceptibility of pancreatic cancer cells to GLV-1h153. Identification and targeting these gene expression changes are needed to avoid side effects associated with higher doses of virus treatment, help develop new strategies to overcome resistance, and identify candidates for clinical trials. Mechanisms of oncotherapy are still poorly understood. 10 The proposed theory of anticancer effects appears to be mainly through oncolysis, although other cellular mechanisms such as apoptosis, necrosis, and autophagy, as well as modification to the tumor vasculature and microenvironment, may also play a role. A study by Weibel et al. investigated the contribution of the tumor vasculature and host immune response after therapy of breast cancer xenografts with GLV-1h68, the parent virus of GLV-1h53, and found that VACV-mediated oncolysis was the primary mechanism of tumor shrinkage in the late regression phase with neither the destruction of the tumor vasculature nor the massive VACV-mediated intratumoral inflammation being a prerequisite for tumor regression. 11 Response of several cancer cell lines to GLV-1h68 was also investigated by Ascierto et al. and showed highly heterogeneous permissivity to VACV infection among the cell lines, but no clear transcriptional pattern could be identified as predictor, suggesting multifactorial basis for permissivity to viral infection. 12 Time and MOI also appear to be important in the effectiveness of viral infection and treatment in a study by Oberg et al. 13 utilizing smallpox and peripheral blood cells from human subjects. They determined that the optimal time to assess the largest amounts of gene changes was at an MOI of 0.5 PFU/cell, and at 18 hours postinfection. It has been recently shown that GFP-marker gene expression correlated with viral copy number is several cell lines, and we thus harvested infected and control pancreatic cells at a maximum of 24 hours postinfection, as determined by maximal GFP expression. 11 It is well known that VACV infection causes significant alterations in cell function and metabolism and interferes with host cell DNA and RNA expression. 14 Our differential gene analysis revealed a total of 12,412 up-and 11,065 downregulated genes at 6 and 24 hours postinfection with GLV-1h153 as compared to control. The top network at 6 hours postinfection included genes with functions related to cell death and cellular development and involved mostly downregulation of genes such as IL-8, hmox1, bcl3, BIRC3, cxcl2, IRF1, cx3cl1, cdkna. At 24 hours postinfection, the top network functions involved gene expression, infection mechanism, and tumor morphology and involved up-and downregulated genes such as TP53, GSR, TRIO, HSPA1L, PLK2, ABL2. Looking at cell death and apoptosis more closely, it was interesting that at 6 hours, the focus is on genes involved in antiapoptotic pathways, with downregulation of antiapoptotic genes There is limited data on specific genes found to be associated with VACV infection, and none looking at pancreatic adenocarcinoma specifically. 13,[17][18][19][20] For example, Guerra et al. identified upregulation of genes in two clusters containing 20 immune response genes at 2, 6, and 16 hours postinfection in Human HeLa cells in response to modified vaccinia virus Ankara. 19 Some of the key immune response genes belonging to these two clusters, including IL-1A, IL-6, IL-8, and components of signal transduction pathways, such as NFKB2, were identified in our study. Agrawal et al. found that vaccinia infection Official journal of the American Society of Gene & Cell Therapy of dendritic cells induced the secretion of IL-6 and TNF-α, which in turn stimulated IFN-γ secretion from T cells, also upregulated in our study. 21 We then investigated the involvement of these genes in known cell signaling pathways. Our analysis showed several statistically significant pathways including HMGB-1, IFN, IL-6, and JAK/STAT signaling at 6 hours postinfection, and by 24 hours, prominent pathways included PI3/AKT, EGFR, and ERK/MAPK, signaling pathways. Overexpression of EGFR has been implicated in pancreatic carcinoma aggressiveness and resistance to chemo-and radiation therapy. 22,23 In a recent study by Morgan et al. 24 , EGFR inhibitors cetuximab and erlotinib was used in combination with gemcitabine for enhanced efficacy against pancreatic cancer. This pathway was significantly modulated by 24 hours after infection with GLV-1h153 in our study, with downregulation of the EGFR receptor, potentially mimicking effects of EGFR inhibitors.
These pathways were also involved in several studies investigating the parent virus of GLV-1h153, GLV-1h68 mostly in tumor xenografts. Worschech et al. demonstrated several pathways also common in our study including interferon, IL-6, JAK/STAT, PTEN, and ERK/MAPK signaling. 25 Only purine metabolism was a common pathway with Reinboth et al. 's study. 26 Like Worschech et al., most of our postinfection cellular gene changes reflected a downregulation and shut down of cellular metabolism. 25 It is difficult to directly compare changes in gene expression due to RNA isolation from tissue rather than cells, adding the variable of the host immune response and gene expression changes which cannot be assessed in cell culture.
Moreover, attenuation of our virus, GLV-1h153, also likely has an important impact on results demonstrated. GLV-1h153 is a derivative of GLV-1h68, which in turn was derived from wild-type LIVP virus. 27 Like GLV-1h153, GLV-1h68 carries three separate insertions in the F14.5L, J2R (TK), and A56R (hemagglutinin) loci of LIVP genome. Zhang et al. demonstrated that gene insertions in GLV-1h68 greatly reduced the replication of GLV-1h68 in normal mouse cells, whereas the replication of GLV-1h68 in tumor cells was not detrimentally affected. Furthermore, i.v. injection of GLV-1h68 into nude mice with human breast tumor xenografts showed enhanced preference for colonization of tumors when compared with wt LIVP and WR strains, leading to restricted distribution of GLV-1h68 mostly to tumors but not to other organs and therefore resulted in less toxicity and extended survival of tumor-bearing nude mice.
In addition, another paper demonstrated via experiments using wild-type single, double, and triple mutant LIVP viruses that infection and replication of LIVP mutant viruses were not very different from those of wt LIVP, suggesting that the single, double, or triple insertions within the LIVP genome did not detrimentally affect the entry and replication of the virus in tumor cells. 28 In normal cells, however, the replication capacity of LIVP mutant viruses was greatly reduced or diminished compared with its wt strain. In addition, virus replication efficiency increased with removal of each of the expression cassettes F14.5L, J2R (TK), and A56R. The increase in virus replication efficiency was also show to be proportionate to the strength of removed VACV promoters linked to foreign genes, and weeks post viral treatment with GLV-1h68 that genes denoting infiltration and activation of immune cells were strongly detected, which are expressed on activated T cells, natural killer cells, macrophages, granulocytes, and dendritic cells and associated with leukocyte activation and natural killer cytolytic function, suggesting a mouse-related immune response is part of the process leading to breast tumor regression. Preferential activation of proinflammatory transcripts, such as chemokine ligands, IL, and chemokine receptors, as well as a panel of IFN-stimulated genes was also seen. This highlights the importance of applying molecular methodologies in vivo in a tumor microenvironment that is as close to clinical application as possible, as it is likely that the innate and adaptive immune responses will also play a role in the way cells behave in response to infection to GLV-1h153. However, prior to conducting such experiments, it would be important to start with a more simplistic approach in cell culture, in order to be in a position to better understand more complex and intertwined gene expression changes and interactions.
Several databases investigating genes associated with poxvirus infection have been established, such as the virus pathogen database and analysis resource (ViPR), 5 the poxvirus bioinformatics resource center, 6 the poxvirus proteomics database, 7 as well as literature addressing detection of and identification of different strains of orthopoxvirus. 1,[29][30][31][32] Although these databases were aimed mainly at the risk of bioterrorism or possible virus pandemic, information from these databases may be crucial to better understand virus behavior for oncotherapy.
In summary, this study reveals the ability to assess time-dependent changes of gene expression patterns in pancreatic cancer cells associated with infection and susceptibility to vaccinia viruses. Our study suggests that molecular assays may be useful to develop safer and more efficacious oncolytic viral therapies and that oncolytic viral treatments may target pathways implicated in pancreatic cancer resistance to conventional therapies.
However, we do recognize limitations of our study. Further work is needed to characterize the role of individual genes and pathways in viral therapy susceptibility and possible resistance, and to confirm these genes on a protein and translational level. Moreover, these experiments were conducted in cellular culture, and further work is necessary to establish how pancreatic cancer cells may behave in a biological model, possibly with an intact innate and adaptive immunity in place. This investigation provides a list of genes and pathways for further detailed studies and provides a framework for the observation of possible cellular events, in addition to potential biologic and molecular targets to overcome oncolytic viral resistance.
MAteRiAlS AND MetHODS
Virus and cell culture African green monkey kidney fibroblast CV-1 cells and human pancreatic ductal carcinoma PANC-1 cells were purchased from American Type Culture Collection (Manassas, VA) and were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 1% antibiotic-antimycotic solution (Mediatech, Herndon, VA) and 10% fetal bovine serum (Mediatech) at 37 °C under 5% CO 2 . GLV-1h68 was derived from VACV LIVP, as described previously. 27 GLV-1h153 was derived from GLV-1h68, as also previously described. 27 Flow cytometry Cells were seeded on six-well plates at 5 × 10 5 Preparation of RNA for microarray Total mRNA preparation was performed in six-well culture plates. Cells were plated at 5 × 10 5 cells per well and infected with GLV-1h153 at an MOI of 5.0. Zero, 6, and 24 hours postinfection, three samples of at each time point were harvested and lysis performed directly using RNeasy mini kit protocol (Qiagen, Valencia, CA). The mRNA samples were measured by spectrophotometer for proof of purity and hybridized to HG-U133A cDNA microarray chips (AffymetrixInc, Santa Clara, CA) by the genomic core laboratory at Memorial Sloan-Kettering Cancer Center.
Microarray analysis
The HG-U133A cDNA microarray chip images were scanned and processed to CEL files using the standard GCOS analysis suite (AffymetrixInc). CEL files were then normalized and processed to signal intensities using the gcRMA algorithm from the Bioconductor library for the R statistical programming system. All subsequent analysis was done on the log (base 2) transformed data. To find differentially expressed genes a moderated t-test was used as implemented in the Bioconductor LIMMA package. To control for multiple testing the false discovery rate method was used with a cutoff of 0.05.
==
Domain: Biology Medicine
|
[{"added": "2016-05-12T22:15:10.714Z", "created": "2016-03-16T00:00:00.000", "id": "10818742", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.cell.com/article/S237277051630047X/pdf", "pdf_hash": "7f446deec6f54dfdc2444aa234eebfb37a358b03", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:37176", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "7f446deec6f54dfdc2444aa234eebfb37a358b03", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The Importance of LDL and Cholesterol Metabolism for Prostate Epithelial Cell Growth
Cholesterol-lowering treatment has been suggested to delay progression of prostate cancer by decreasing serum LDL. We studied in vitro the effect of extracellular LDL-cholesterol on the number of prostate epithelial cells and on the expression of key regulators of cholesterol metabolism. Two normal prostatic epithelial cell lines (P96E, P97E), two in vitro immortalized epithelial cell lines (PWR-1E, RWPE-1) and two cancer cell lines (LNCaP and VCaP) were grown in cholesterol-deficient conditions. Cells were treated with 1–50 µg/ml LDL-cholesterol and/or 100 nM simvastatin for seven days. Cell number relative to control was measured with crystal violet staining. Changes in mRNA and protein expression of key effectors in cholesterol metabolism (HMGCR, LDLR, SREBP2 and ABCA1) were measured with RT-PCR and immunoblotting, respectively. LDL increased the relative cell number of prostate cancer cell lines, but reduced the number of normal epithelial cells at high concentrations. Treatment with cholesterol-lowering simvastatin induced up to 90% reduction in relative cell number of normal cell lines but a 15–20% reduction in relative number of cancer cells, an effect accompanied by sharp upregulation of HMGCR and LDLR. These effects were prevented by LDL. Compared to the normal cells, prostate cancer cells showed high expression of cholesterol-producing HMGCR but failed to express the major cholesterol exporter ABCA1. LDL increased relative cell number of cancer cell lines, and these cells were less vulnerable than normal cells to cholesterol-lowering simvastatin treatment. Our study supports the importance of LDL for prostate cancer cells, and suggests that cholesterol metabolism in prostate cancer has been reprogrammed to increased production in order to support rapid cell growth.
Introduction
Current literature suggests that cholesterol may play an important role in the development and progression of prostate cancer. Several epidemiologic studies have reported a significant positive correlation between hypercholesterolemia or dyslipidemia and prostate cancer incidence [1][2][3][4][5][6][7]. Experimental studies support these findings, as elevation of circulating cholesterol promotes tumor growth and tumor cholesterol content in a mouse LNCaP xenograft model [8,9], while reduction in cholesterol levels retards prostate cancer growth, possibly by inhibition of tumor angiogenesis [10]. Recently, epidemiological and laboratory studies have suggested that cholesterol-lowering statin drugs might lower the risk of advanced prostate cancer [11].
In vitro studies have proposed that the elevated cholesterol levels in prostate tumor cells could be due to dysregulation of the key regulators of cholesterol homeostasis [12,13], which could have significance in the progression of prostate cancer into androgenindependent state [14,15]. Very little is currently known, however, about cholesterol metabolism in normal prostatic epithelial cells and its differences compared to cancer cells.
In the present study we evaluated the effect of cholesterol on growth of both primary and in vitro immortalized prostate epithelial cells, and on the growth of androgen-dependent cancer cells. Additionally, we studied the effects of cholesterol and statin treatment on the expression of key participants in cholesterol metabolism: 3-hydroxy-3-methylglutaryl-Coa-reductase (HMGCR), a rate-limiting enzyme in cholesterol-producing mevalonate pathway; Low density lipoprotein receptor (LDLR), required for LDL uptake; Sterol-regulatory element binding protein 2 (SREBP2), regulator of intracellular cholesterol content [16] and the ATP-binding cassette, subfamily A, member 1 (ABCA1), which mediates the efflux of cellular cholesterol [17].
Cell Lines and Culture Conditions
Generation and authentication of P96E and P97E primary prostatic normal epithelial cell lines has been described previously [19]. RWPE-1 and PWR-1E cells (immortalized prostate epithelial cell lines) were a gift from VTT Technical Research Centre, Turku, Finland. P96E, P97E, PWR-1E and RWPE-1 cells were cultured in K-SFM supplemented with 50 mg/ml BPE, 5 ng/ml rEGF and 1% A/A. LNCaP prostate cancer cells were from American Type Culture Collection (Rockville, MD, USA). VCaP prostate cancer cells were a gift from Professor T. Visakorpi, IBT institute, University of Tampere, Finland. LNCaP and VCaP cells were cultured in RPMI 1640 supplemented with 10% FCS, 1% Lglutamine and 1% A/A.
For studies on cell number relative to control, 4610 4 (PWR-1E), 5610 4 (RWPE-1), 6610 4 (P96E, P97E and LNCaP) or 3610 5 (VCaP) cells per well were seeded on 6-well plates and allowed to attach for 48 hours. LNCaP and VCaP cells were grown on CorningH CellbindH 6-well plates, whereas normal cell lines were grown on 6-well plates from Nalge Nunc International. After attachment, LNCaP and VCaP cells were grown in lipid deficient medium (RPMI 1640 supplemented with 10% LPDS, 1% Lglutamine and 1% A/A). The normal prostate epithelial cells were routinely grown in Keratinocyte-SFM which is serum free and essentially lipid deficient.
The cells were treated with LDL-cholesterol or vehicle (DMSO) for seven days. LDL-cholesterol was used in 1-50 mg/ml concentrations to test the dose-dependence of effect. This is the concentration range in standard cell culture conditions when 10% fetal calf serum is being used [20]. This range also allows proper functioning of the LDL-receptor [21]. The highest concentration (50 mg/ml) is in the range of that found in human plasma (from ,100 mg/dl to .250 mg/dl) assuming relation 10:1 between concentration in plasma to that of interstitial tissue.
Growth medium and drugs were renewed every other day. After treatments, the cells were fixed, stained and their number was assessed with modified crystal violet staining method [22]. Absorbances were measured at day 0 and day 7with a Victor 1420 Multilabel Counter (Wallac, Turku, Finland), and the value at day 0 was subtracted from the values at day 7.
For the RNA and protein studies, the cells were seeded to 75 cm 2 flasks and allowed to attach for 48 hours. After attachment, the cells were grown in lipid deficient medium as described above and treated with vehicle (DMSO), 100 nM simvastatin, 50 mg/ml LDL-cholesterol or in their combination for 48 hours and then subjected to Trizol (Invitrogen, Carlsbad, CA, USA) reagent for RNA extraction or M-PERH (PIERCE, Rockford, IL, USA) reagent modified with protease inhibitors (Complete Mini Protease inhibitor cocktail tablets (Roche Diagnostics GmbH, Indianapolis, IN, USA)) for protein extraction according to the manufacturer's instructions.
SDS-PAGE and Western Blot
Total protein concentrations were measured using BCA Protein Assay Kit (Pierce) according to the manufacturer's instructions. 50 mg of total protein was mixed (1:1) with 2X Laemmli sample buffer (Sigma, ST. Louis, MO, USA), boiled for 5 min and analyzed by electrophoresis in 12% polyacrylamide gel (PAGE). An exception to this, protein samples for HMGCR were not boiled to avoid protein aggregation upon heating. Precision Plus Protein Standards were used (Bio-Rad Laboratories, Hercules, CA, USA). Proteins separated by PAGE were transferred (1 hour) to the Immobilon-P polyvinylidene fluoride transfer membrane (0.45 mm pore size) (Millipore, Billerica, MA USA) at room temperature (RT) using NuPage transfer buffer (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Membranes were then incubated for 1 hour at RT in Tris buffer containing salt and Tween (TBST) (50 mM Tris-HCL, 150 mM NaCL, 0,05% Tween 20, pH 8.0) and 5% non-fat dry milk powder (5% milk-TBST) to saturate the non-specific protein binding sites. Membranes were incubated with the primary antibodies in 5% milk-TBST overnight at 4uC with mild agitation. The membranes were washed 3 times for 5 min with TBST and incubated for 1 hour with horse radish peroxidase -conjugated secondary antibody in 5% milk-TBST with mild agitation at RT. The membranes were washed 3 times for 5 min with TBST and subjected to enhanced chemiluminescence reagents (ECL Western Blotting Detection Reagents, GE Healthcare, Buckinghamshire, UK) according to the manufacturer's instructions and exposed to X-ray film.
Real-Time RT-PCR
The RNA samples were reverse transcribed to cDNA with High Capacity Archive Kit (Applied Biosystems, CA, USA) following the instructions of the manufacturer. The real-time RT-PCR was performed by using SYBR Green PCR Master Mix Kit (Applied Biosystems) in ABI PRISM 7000 Detection System (Applied Biosystems) according to the manufacturer's instructions. The data were analyzed by ABI PRISM 7000 SDS Software (Applied Biosystems). The final results, expressed as N-fold relative differences (ratio) in gene expression between the studied samples and the control (i.e. calibrator) sample, were calculated according to the following equation [15]:
Statistical Analysis
All experiments were repeated separately three times. The median, the highest and the lowest values are reported for each treatment. The non-parametric Mann-Whitney U-test was used to analyze the statistical significance of differences in the outcome measurements between treatments. All p-values are two-sided.
LDL, Simvastatin and Cell Number
The response in cell number relative to control to increasing concentrations of LDL-cholesterol differed between cancer cells and primary or transformed prostate epithelial cells ( Fig.1). High concentrations (30 and 50 mg/ml) of LDL-cholesterol clearly reduced the number of primary cells. However, only slight reduction in the relative cell number was observed in PWR-1E when the highest concentration of LDL-cholesterol (50 mg/ml) was used. On the other hand, relative cell number of both cancer cell lines was slightly stimulated by LDL-cholesterol at the highest concentrations (Fig.1).
Both 100 nM simvastatin and 50 mg/ml LDL-cholesterol reduced the number of normal epithelial cells, with the exception of PWR-1E (Fig. 2). LDL-cholesterol attenuated the relative cell number reduction caused by simvastatin in P96E and P97E cells (p,0.05 for difference between combination of simvastatin and LDL treatment as compared to simvastatin alone), although the reduction relative to control cells remained significant. Compared to the control, addition of LDL to simvastatin removed the significant relative cell number decreasing effect of simvastatin alone in RWPE-1 cells, although the difference between the two treatments remained non-significant. In cancer cell lines simvastatin caused only modest reduction in relative cell number, an effect fully compensated by LDL-cholesterol (Fig 2). Simvastatin slightly reduced the relative cell number increasing effect of LDLcholesterol on cancer cells.
Expression of Cholesterol Metabolizing Factors at Baseline
The basal protein expression levels of important regulators of cellular cholesterol metabolism in a standard amount of protein were compared between normal epithelial cells and cancer cell lines after the cells had been grown in cholesterol-deficient medium for seven days. All cell lines expressed SREBP2 at protein level, cancer cell lines more strongly than normal cell lines (Fig. 3a, suppl. Fig S1), although the mRNA expression did not differ greatly between the cell lines (Fig 3b). The exception was RWPE-1, where mRNA expression of SREBP2 was low compared to any other cell line.
Under these circumstances the cancer cell lines exhibited upregulation of HMGCR at protein level, suggesting increased cholesterol production, whereas normal primary cells showed upregulation of LDLR (Fig 3a, suppl. Fig S1). Again, baseline mRNA expression differed from protein expression as both HMGCR and LDLR expression were markedly higher in normal primary cells P96E and P97E as compared to cancer cell lines (Fig 3b). In PWR-1e and RWPE-1 the mRNA expressions were similar to cancer cell lines.
Even under depletion of extracellular cholesterol, the normal epithelial cells (with the exception of PWR-1E) expressed cholesterol transporter ABCA1 at protein level, whereas cancer cell lines did not (Fig. 3a, suppl. Fig S1). For ABCA1 The mRNA expression was similar to protein expression: high expression in normal cell lines P96E, P97E and RWPE-1, but almost no expression in cancer cell lines and PWR-1e (Fig 3b).
Effect of LDL and Simvastatin on the Expression of Key Cholesterol-metabolizing Factors
Inhibition of de novo cholesterol synthesis with simvastatin sharply upregulated the mRNA (Fig. 4 a and b) and protein expressions of HMGCR and LDLR ( Fig. 4c and d) in all cell lines. Simvastatin also upregulated mRNA expression of SREBP2 (Fig. 5b). At protein level simvastatin treatment did not increase SREBP2 expression, but rather caused cleavage of the protein into 125 kDa and 60 kDa bands (Fig. 5d). In normal cells the expression of ABCA1 was clearly down-regulated by simvastatin ( Fig. 5a and c). In cancer cells simvastatin did not markedly affect ABCA1 expression (Fig. 5a and c).
Compared to control, LDL-cholesterol downregulated HMGCR mRNA expression significantly only in P97E, while downregulation of protein expression was most clearly observed in cancer cell lines LNCaP and VCaP (Fig 4 a and c). LDL also downregulated LDLR mRNA expression in P97E and LNCaP (Fig 4b), but protein expression was downregulated in all cell lines except P96E (Fig. 4d). The response in ABCA1 differed between cancer cells and normal cells: availability of extracellular LDL upregulated ABCA1 in the normal cell lines, but the cancer cells did not express this transporter at detectable protein level even after LDL-cholesterol treatment, although slight changes were observed in mRNA expression (Fig. 5a and c). LDL decreased the mRNA expression of SREBP2 in P97E, but in other cell lines the expression was comparable to vehicle-treated cells (Fig. 5b). At protein level LDL prevented the effect of simvastatin on cleavage of SREBP2 into two bands in P96E, P97E and RWPE-1 (Fig 5d).
LDL prevented most of the effects of simvastatin on the expression of cholesterol metabolizing factors (Fig. 4a-d and Fig 5ad). An exception was HMGCR in the LNCaP, where simvastatin caused upregulation of the enzyme expression even in the presence of LDL (Fig. 4a and c).
Discussion
Our observations support the importance of cholesterol for the growth of prostate cancer cell lines: 1) increase in cell number relative to control after treatment with increasing concentrations of LDL; 2) decreased relative cell number after inhibition of intracellular cholesterol synthesis with simvastatin, which could be prevented by addition of LDL; 3) enhanced expression of HMG-CoA reductase, the rate-limiting enzyme of cholesterol biosynhesis at baseline in cancer cell lines and 4) no evidence of ABCA1 expression in cancer cells under any circumstances, even after LDL treatment.
Cholesterol is important for cell membrane integrity and cellular metabolism, as well as for signalling pathways essential for cellular proliferation, such as PI3K/Akt [23]. Combined, our results suggest that LDL is needed for growth of prostate cancer cells. Increased expression of the biosynthetic machinery along with no expression of the major participant in cholesterol efflux from the cells suggests reprogramming of cholesterol metabolism in cancer cells. Although remaining responsive to changes in extracellular conditions such as treatment with simvastatin or LDL, the metabolism has been geared towards providing the cells with maximal supply of cholesterol to enable rapid cell growth under any conditions. Even in cholesterol-free conditions inhibition of intracellular cholesterol synthesis with simvastatin reduced the number of cancer cells 15-20%, but up to 90% of normal epithelial cells; presumably higher baseline cholesterol synthesis protects cancer cells against the effects of simvastatin. However, we did not directly measure intracellular cholesterol synthesis.
Besides cholesterol, mevalonate pathway produces also isoprenoids farnesylpyrophosphate (FPP) and geranylgeranylpyrophosphate (GGPP), which in turn have important cell growth regulatory functions [24]. Inhibition of these end-products of mevalonate pathway and resulting cellular changes are termed pleiotropic effects of statins. The differing ability of LDL to restore the relative cell number reduction caused by simvastatin between the cell lines could have been due to differing role of pleiotropic effects. In future the relationship between inhibition of isoprenoid production and cholesterol production when studying statins' effects on cell growth should be further studied.
The relative cell number of normal epithelial cell lines was not induced by LDL, but conversely high concentrations caused reduction. Normal cells also require cholesterol for cell growth as treatment with simvastatin caused a powerful growth inhibition, again restored by addition of LDL. Normal cells responded to simvastatin treatment by increasing HMGCR and LDLR expressions, but unlike the cancer cell lines, normal cells also increased expression of cholesterol exporting transporter ABCA1 as a result of treatment with LDL. This suggests that normal cells need equilibrium in cholesterol homeostasis for undisturbed cell growth. The changes in normal cells reflect attempts to adapt to changing extracellular conditions by adjusting intracellular cholesterol metabolism to maintain the equilibrium. Very high LDL concentrations, however, likely exceed this adaptive potential, causing toxic growth inhibition. Such was not observed in cancer cell lines, however. These differences between normal prostatic epithelial cells and cancer cell lines reflect the changes in cholesterol metabolism occurring during carcinogenesis in the prostate. Likely reprogramming of cholesterol metabolism is a crucial part of the rearrangement of energy metabolism in cancer cells supporting constant proliferation [25], a trait that has been recognized as one of the hallmarks of cancer [26].
Also in previous studies cholesterol has increased the growth of prostate cancer cell lines PC3 and DU-145 [27,28]. Unlike in our study LDL treatment has not been previously found to induce growth of LNCaP cells [29], despite similar downregulation of LDLR expression. The discrepancy in the results is possibly explained by the shorter duration of LDL treatment in the previous study (48 h) compared to ours (seven days). In this paper we focused on effects of LDL to further explore the association with prostate cancer risk reported in epidemiological studies [7,30] and observed in our previous studies [31]. Nevertheless, also highdensity lipoprotein (HDL) has been reported to induce prostate cancer cell growth [32], suggesting that cancer cells can probably use various types of lipoproteins as a source of cholesterol.
The importance of cholesterol for prostate cancer growth is further supported by experimental studies, where elevation of circulating cholesterol has been reported to increase tumor growth and intra-tumoral cholesterol accumulation in a mouse LNCaP xenograft model [8,9], PC-3 xenograft [27] and DU-145 xenograft [33]. A hypercholesterolemic diet changes prostate morphology in male Wistar rats [34]. On the other hand, reducing cholesterol levels retards prostate cancer growth possibly by inhibition of tumor angiogenesis in a prostate cancer xenograft model [10].
Changes in the expression levels of the key regulators of cholesterol homeostasis, namely sterol regulatory element binding transcription factors (SREBPs), HMGCR LDLR, acetyl-CoA acetyltransferase 1 (ACAT1) and scavenger receptor class B member 1 (SR-B1) have been shown to occur during the progression of prostate cancer from androgen-independent to castration-resistant cancer in an LNCaP xenograft model [14,15]. Cholesterol influx by SR-B1 is essential for viability of prostate cancer cell lines such as LNCaP [35]. We have demonstrated that marked differences in expression of key regulators of cholesterol metabolism are observed already between normal epithelial cell lines and androgen responsive LNCaP and VCaP cancer cell lines. Nevertheless, cholesterol metabolism remains responsive to extracellular stimuli; our results are in concordance with a previous study by Krycer et al [13] reporting feedback regulation of SREBP2, HMGCR and LDLR mRNA expression in cancer cells and normal epithelial cells by extracellular cholesterol. We further show that this regulation occurs at protein level, and also in primary normal prostate epithelial cells which have been isolated directly from prostatic tissues. The differences observed between mRNA and protein expressions of HMGCR, LDLR and SREBP2 suggests that mRNA of these enzymes may undergo posttranslational modifications before transcription into protein level. Further research will be needed.
In vivo evidence for the importance of cholesterol in prostate cancer progression comes from epidemiological studies reporting increased risk of advanced prostate cancer among hypercholesterolemic men [1,2]. Serum cholesterol decreases spontaneously within nine years before a cancer diagnosis [3], which might indicate that a developing tumor consumes cholesterol from the circulation to enable cell growth; a notion supported by some in vitro studies [36]. Improved recurrence-free survival after radical treatment of prostate cancer has been reported among men using cholesterol-lowering statin drugs, an association possibly related to serum LDL decrease during statin therapy [37].
We could not test the responses of early-stage prostate cancer cells to LDL and statin treatments as these are not currently commercially available. However, it could be reasonably presumed that the responses of well-differentiated prostate cancer cells at the early stages of carcinogenesis resemble those of normal epithelial cells. In our study the cells were grown in monolayer cultures, whereas in vivo prostate epithelial cells are in close contact with the surrounding stroma, which has important functions in carcinogenesis [26] and could modify epithelial cells' responses to LDL and simvastatin. Thus in vivo studies will be needed to confirm our findings.
We have shown that increasing doses of LDL induce number of prostate cancer cells, but not normal epithelial cells. Both normal and cancer cells increase the production of effectors that ensure the synthesis and uptake of cholesterol under depletion, but cancer cells do not express the major exporter of cholesterol, ABCA1 even in the abundance of LDL. Cholesterol availability is likely an important prerequisite for prostate cancer growth and cholesterol metabolism in prostate cancer cells is reprogrammed to supply the cells with abundance of cholesterol. Cholesterol-lowering might prove to be a good strategy to prevent and delay prostate cancer progression. Hypercholesterolemia as an etiologic factor for prostate cancer deserves further studies. Figure S1 Quantification of relative intensities of immunoblotted bands shown in Fig 3. Relative intensities of bands on Western blots were quantified using ImageJ 1.45 ( [URL]:// imagej.nih.gov/ij/) according to instructions by Luke Miller available in [URL]/2010/11/ analyzing-gels-and-western-blots-with-image-j/with minor modifications. Shortly, band density for a given protein in different cell types was divided with that of P96E cells, to obtain relative densities of bands. The relative densities in P96E cells represent the value 1. Values below 0.1 are denoted ,0.1. Cases in which no band was detected are denoted as n.d. (not detected). (DOC)
==
Domain: Biology Medicine
|
[{"added": "2017-04-12T20:41:37.895Z", "created": "2012-06-27T00:00:00.000", "id": "14752575", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0039445&type=printable", "pdf_hash": "8698fe3a50efbea24a663d4e021689172d55df91", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:35816", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "8698fe3a50efbea24a663d4e021689172d55df91", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Everyday and exotic foodborne parasites
MB Lee. Everyday and exotic foodborne parasites. Can J Infect Dis 2000;11(3):155-158. Everyday foodborne parasites, which are endemic in Canada, include the protozoans Entamoeba histolytica , Giardia lamblia and Cryptosporidium parvum. However, these parasites are most frequently acquired through unfiltered drinking water, homosexual activity or close personal contact such as in daycare centres and occasionally via a food vehicle. It is likely that many foodborne outbreaks from these protozoa go undetected. Transmission of helminth infections, such as tapeworms, is rare in Canada because of effective sewage treatment. However, a common foodborne parasite of significance is Toxoplasma gondii . Although infection can be acquired from accidental ingestion of oocysts from cat feces, infection can also result from consumption of tissue cysts in undercooked meat, such as pork or lamb. Congenital transmission poses an immense financial burden, costing Canada an estimated $240 million annually. Also of concern is toxoplasmosis in AIDS patients, which may lead to toxoplasmosis encephalitis, the second most common AIDS-related opportunistic infection of the central nervous system. Exotic parasites (ie, those acquired from abroad or from imported food) are of growing concern because more Canadians are travelling and the number of Canada’s trading partners is increasing. Since 1996, over 3000 cases of Cyclospora infection reported in the United States and Canada were epidemiologically associated with importation of Guatemalan raspberries. Unlike toxoplasmosis, where strategies for control largely rest with individual practices, control of cyclosporiasis rests with government policy, which should prohibit the importation of foods at high risk.
T his paper is a brief survey of the everyday foodborne parasites endemic in Canada and exotic foodborne parasites, which may be acquired by the travelling Canadian or be brought into Canada by an immigrant. Special attention will be paid to Toxoplasma gondii infection because of its severity in terms of human misery and costs, and to Cyclospora cayetanensis because of its recent appearance in North America.
Parasites have been with us since the beginning of humankind, despite the media's new found attention to them. The protozoan parasites, Cryptosporidium parvum, Giardia lamblia and Entamoeba histolytica are commonly acquired through water, homosexual activity or by close personal contact, such as in daycare centres (1), rather than by food. However, in countries where raw human excrement is used as fertilizer, foodborne transmission is undoubtedly higher. Occasionally, reports of protozoan foodborne outbreaks make their way into the literature, but this is uncommon (2). In 1997, in Spokane, Washington, 54 people who had eaten at a banquet became ill with diarrhea, fever, chills and headache (3). C parvum was the culprit, epidemiologically associated with green onions that were not consistently washed during preparation, and were used in the au gratin potatoes, romaine salad and pasta. Because protozoan parasites are sparsely distributed in food, recovery and identification are not routinely possible, so foodborne outbreaks go largely undetected.
The most common of the protozoan parasites in Ontario is G lamblia (Table 1). Many infected individuals are asymptomatic, while others may have gas, abdominal cramping, fatigue and diarrhea. Often physicians will choose to treat symptomatic patients while not treating asymptomatic ones.
Entamoeba dispar is not a pathogen but is morphologically indistinguishable from its close relative E histolytica. It is thought that many laboratory identifications in Canada for E histolytica are really E dispar. However, without knowing, physicians usually choose to treat patients when the report says E histolytica. A rapid test that can distinguish between these two pathogens is awaited.
Endemic helminth infections are rare in Canada. However, Trichinella spiralis in underprocessed wild boar meat caused illness in 24 people in southern Ontario in 1993 (4). The wild boar may have acquired the organism by eating undercooked pig slaughterhouse scraps or foraging for rats. One may question what is meant by the term 'wild' boar if they were feeding on garbage scraps, rather than roaming free to seek their food. Outbreaks of trichinosis have also been reported from the consumption of raw, fresh horsemeat, which is popular in Europe. Because horses are herbivorous, it is a mystery how they become infected. Carnivores may have been ground up in their hay. Outbreaks in France and Italy from 1975 to 1993 resulted in over 2600 human cases of trichinosis (5). Imported carcasses, some from the United States and Canada, that were refrigerated, but not frozen, were linked to these outbreaks. Freezing is a reliable method for inactivating trichinae when done at the appropriate temperatures for the required time.
The adult tapeworms Taenia solium and Taenia saginata are acquired by eating undercooked pork or beef, respectively. T solium is the nastier of the two. A person harbouring an adult T solium worm will release worm eggs in their stool. If the person's handwashing habits after toileting are poor, they may unwittingly transmit eggs to others via food that it handled. This may occur within a family, where the eggs, after being ingested, may develop into cystic lesions in the brain. In one case in Boston in 1990, a 16-month-old girl had a seizure. A computed tomography (CT) scan revealed several ring-enhancing lesions. Stool specimens from her father contained Taenia species eggs (6). He had undoubtedly eaten undercooked pork sometime previously, likely in the Cape Verde Islands where he had lived before his daughter's birth. Fortunately, pork tapeworm is rare in Canada. T saginata is more common, due to the more common habit of eating undercooked beef than pork. Laboratories cannot distinguish the egg of T saginata from that of T solium,so only the genus is reported. A detailed food history may point to a particular species.
Raw fish may also be a source of parasites. Anisakid worms are occasionally found in raw sea fish and, when consumed, may burrow into the stomach or intestine, causing acute pain. With the increasing popularity of sushi in Canada (seaweed wrapped rice topped with raw seafood), one might think that anisakids would be an increasing concern. However, documented cases of anisakiasis in Canada are few (7), unlike Japan where hundreds of cases are reported annually (8). In Spain, allergic reactions to the dead anisakids have been reported which may be a more significant consequence than having the live parasite (9). One fish parasite that is reported, albeit infrequently, in Ontario is Diphyllobothrium latum, which is acquired by eating undercooked freshwater fish, such as pike. Twenty isolations were made by private and public health laboratories in Ontario in 1997 (Table 1).
T gondii is the most significant foodborne parasite because it is common, its acquisition can be devastating and it places an immense financial burden on society. The most likely route of exposure is through consumption of undercooked meat; pork or lamb are more likely to be infected than beef (Table 2) (10). Cattle appear to develop a more effective immune response to toxoplasma than sheep. Additionally, T gondii may be acquired from accidental ingestion of oocysts in cat excrement, by transplant or transplacentally, a potentially devastating consequence. Tizard et al (14) tested 7060 human serum specimens from a cross section of Ontario residents for antibody to T gondii and found 38% testing positive. As the population ages, the incidence rises due to increased opportunity for exposure. Rates of seropositivity are known to climb from 2.7% in infants, who have had little opportunity to be exposed to T gondii, to 50% in individuals aged 25 years or older, who have had increased opportunities for exposure.
The congenital form of toxoplasmosis may result in children being developmentally delayed, blind or deaf. Economic losses primarily result from institutionalization or special education needs of victims. In 1992, Roberts et al (15) estimated that 4179 cases of congenital toxoplasmosis occur annually in the United States at a cost to society of US$2.4 billion. One could interpolate the American data, using a 1 to 10 ratio, to predict about 400 cases of congenital toxoplasmosis in Canada annually. This is a reasonable estimate because in 1986, Carter and Frank (16) estimated 140 to 1400 cases annually of congenital toxoplasmosis in Canada. Likewise, using 10% of the American cost estimates would place the cost to Canadian society at $240 million annually, no small amount.
Toxoplasmosis in AIDS patients is another concern because the organism may go to the brain, causing toxoplasmic encephalitis (TE). TE is the second most common AIDS-related opportunistic infection of the central nervous system, and occurs in 10% to 50% of patients with AIDS who are positive by antibody serology to toxoplasma and have a low CD4+ T lymphocyte count (17). American costs for infected patients' care are estimated at over US$17,000/patient annually (18). Toxoplasmosis in the AIDS patient is likely a reactivation of latent tissue cysts and not a recent infection because immunoglobulin G antibodies in patients indicate prior infection.
Food control of T gondii includes cooking meats to at least 66°C (19) (better yet cook to 71°C because that will destroy salmonellae too) or freezing to -12°C (20). Gamma irradiation of meat has been shown to be effective at 0.5 kGy in inactivating tissue cysts of toxoplasma. If consumers are willing to accept irradiated food, this would appear to be the best strategy because public health efforts are maximized when control is centralized rather than relying on individual behaviour.
Potential solutions for controlling toxoplasmosis in live-stock include a live vaccine for sheep, available in New Zealand and the United Kingdom (21). The vaccine is intended to prevent abortion in sheep. A live vaccine is being developed in the United States for cats to reduce oocyst shedding. In this discussion, 'exotic parasites' refers to parasites entering in imported foods or brought back by Canadians travelling abroad. C cayetanensis has been in the North American news since 1996, when it was associated with an outbreak arising from importation of Guatemalan raspberries (22)(23)(24). Washing raspberries will not remove all the parasites which may be hiding in nooks and crannies. From 1996 to 1998, there were over 3000 cases of cyclosporiasis in North America. Raspberries are not the only food vehicle linked to illness; a Florida outbreak involved mesculun lettuce (23) and another outbreak in the Maryland area involved basil in a basil-pesto salad (25). One of the Maryland victims did not even eat the salad but used the serving spoon to serve himself leftovers. This suggests a small infective dose, as does the fact that this is a coccidian parasite (T gondii and C parvum are others), which are infective in small numbers. Although a team of federal investigators travelled from Canada and the United States to Guatemala in the spring of 1998 to investigate, the source of the C cayetanensis was not found. Human sewage from pits near fields or from workers with poor personal hygiene were suspect. Other documented outbreaks in Nepal and Papua, New Guinea have occurred from contaminated water (26). Although coccidia are not destroyed by chlorination, because of their resistant oocyst wall, filtration in municipal treatment plants can capture the 10 m m oocyst of Cyclospora species.
Potential solutions on farms include the use of potable water for spraying crops, workers thoroughly hand washing after using the toilet and toilet facilities that keep sewage from contaminating the environment. It would also be useful to be able to trace berries back to a particular farm if an outbreak were to occur. Canadians must rely on federal government policy to restrict the sale of imported produce if the produce is 'high risk', eg, produced under unhygienic conditions. Irradiation may be a future alternative if it will not compromise the texture of the berry.
Other exotic parasites include the flatworm species Echinostoma, Hetrophyes, Metagonimus, Clonorchis, and Opisthorchis, which live in intestines or bile ducts. They are usually acquired from eating raw or improperly processed fish, snails or crab in Asian countries (Table 1). Some species of Opisthorchis in Thailand infect over 90% of the inhabitants in some villages (27). Nematodes, roundworms of the intestinal tract, that are transmitted in food include Ascaris lumbricoides and Trichuris trichiura. They are not associated with any specific food but are acquired when human sewage contaminates produce that is not cooked. These parasites may pose a medical problem to the infected but not to the public's health because they are not directly infective, needing to mature in soil for one to several weeks. Because of good sewage control in developed countries, through municipal treatment plants and septic tank systems, the eggs are not disseminated to fertilize produce where they would start the cycle again.
CONCLUSIONS
We are fortunate to live in Canada, where we generally have good public health practices such as cooking foods well, municipal water filtration and sewage disposal, that aid in reducing our exposure to infectious agents. Additionally, we rely on government policy to prohibit importation of foods that may be high risk. Occasionally, however, these systems can fail as this discussion has indicated. We still have room for reducing risk and improving our health.\===
Domain: Biology Medicine. The above document has 2 sentences that start with 'Everyday and exotic foodborne', 2 sentences that end with 'rare in Canada'. It has approximately 2148 words, 109 sentences, and 18 paragraph(s).
|
[{"added": "2018-04-03T00:16:18.881Z", "created": "2000-05-01T00:00:00.000", "id": "10705697", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://downloads.hindawi.com/journals/cjidmm/2000/120498.pdf", "pdf_hash": "5c96004d99eba4ffc18134a3e9f820b0ba849b97", "pdf_src": "ScienceParsePlus", "provenance": "peS2o-0075.json.gz:34937", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "127ab2d1095835ffcda9ab5bd9ae3493949f091e", "year": 2000}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Longitudinal Transcriptome Analysis Reveals a Sustained Differential Gene Expression Signature in Patients Treated for Acute Lyme Disease
ABSTRACT Lyme disease is a tick-borne illness caused by the bacterium Borrelia burgdorferi, and approximately 10 to 20% of patients report persistent symptoms lasting months to years despite appropriate treatment with antibiotics. To gain insights into the molecular basis of acute Lyme disease and the ensuing development of post-treatment symptoms, we conducted a longitudinal transcriptome study of 29 Lyme disease patients (and 13 matched controls) enrolled at the time of diagnosis and followed for up to 6 months. The differential gene expression signature of Lyme disease following the acute phase of infection persisted for at least 3 weeks and had fewer than 44% differentially expressed genes (DEGs) in common with other infectious or noninfectious syndromes. Early Lyme disease prior to antibiotic therapy was characterized by marked upregulation of Toll-like receptor signaling but lack of activation of the inflammatory T-cell apoptotic and B-cell developmental pathways seen in other acute infectious syndromes. Six months after completion of therapy, Lyme disease patients were found to have 31 to 60% of their pathways in common with three different immune-mediated chronic diseases. No differential gene expression signature was observed between Lyme disease patients with resolved illness to those with persistent symptoms at 6 months post-treatment. The identification of a sustained differential gene expression signature in Lyme disease suggests that a panel of selected human host-based biomarkers may address the need for sensitive clinical diagnostics during the “window period” of infection prior to the appearance of a detectable antibody response and may also inform the development of new therapeutic targets.
ABSTRACT Lyme disease is a tick-borne illness caused by the bacterium Borrelia burgdorferi, and approximately 10 to 20% of patients report persistent symptoms lasting months to years despite appropriate treatment with antibiotics. To gain insights into the molecular basis of acute Lyme disease and the ensuing development of post-treatment symptoms, we conducted a longitudinal transcriptome study of 29 Lyme disease patients (and 13 matched controls) enrolled at the time of diagnosis and followed for up to 6 months. The differential gene expression signature of Lyme disease following the acute phase of infection persisted for at least 3 weeks and had fewer than 44% differentially expressed genes (DEGs) in common with other infectious or noninfectious syndromes. Early Lyme disease prior to antibiotic therapy was characterized by marked upregulation of Toll-like receptor signaling but lack of activation of the inflammatory T-cell apoptotic and B-cell developmental pathways seen in other acute infectious syndromes. Six months after completion of therapy, Lyme disease patients were found to have 31 to 60% of their pathways in common with three different immune-mediated chronic diseases. No differential gene expression signature was observed between Lyme disease patients with resolved illness to those with persistent symptoms at 6 months post-treatment. The identification of a sustained differential gene expression signature in Lyme disease suggests that a panel of selected human host-based biomarkers may address the need for sensitive clinical diagnostics during the "window period" of infection prior to the appearance of a detectable antibody response and may also inform the development of new therapeutic targets. IMPORTANCE Lyme disease is the most common tick-borne infection in the United States, and some patients report lingering symptoms lasting months to years despite antibiotic treatment. To better understand the role of the human host response in acute Lyme disease and the development of post-treatment symptoms, we conducted the first longitudinal gene expression (transcriptome) study of patients enrolled at the time of diagnosis and followed up for up to 6 months after treatment. Importantly, we found that the gene expression signature of early Lyme disease is distinct from that of other acute infectious diseases and persists for at least 3 weeks following infection. This study also uncovered multiple previously undescribed pathways and genes that may be useful in the future as human host biomarkers for diagnosis and that constitute potential targets for the development of new therapies.
cytes constitute the first line of defense, engulfing the spirochete and producing Th1-type proinflammatory cytokines. Spirochetal lipoproteins can directly stimulate the B-cell response, and both lipidated and nonlipidated proteins trigger T-cell-dependent humoral responses. Decreased Th1 and increased Th17 responses have also been shown to play a role in the development of posttreatment Lyme disease symptoms during the chronic phase of the illness (8,9). However, with the exception of antibiotic-refractory Lyme arthritis, very few studies have looked at the molecular mechanisms underlying persistent symptomatology in treated Lyme disease patients, and all to date have used targeted approaches assaying specific cytokine levels (8)(9)(10). The overall global and temporal pathways involved in human clinical infection with B. burgdorferi remain to be elucidated.
In this study, we applied next-generation sequencing of peripheral blood mononuclear cells (PBMCs) to investigate the transcriptomes of 29 patients with acute Lyme disease longitudinally from the time of diagnosis to 6 months post-treatment and those of 13 matched controls. We performed network and pathway analyses in order to gain insights into the molecular mechanisms underpinning acute Lyme disease and post-treatment symptoms and to discover potential diagnostic biomarkers.
RESULTS
Patient enrollment, sample collection, and transcriptome analysis. This study included a cohort of 29 patients with acute Lyme disease and 13 matched controls without acute illness. Transcriptome profiling by RNA sequencing (RNA-Seq) and pathway analysis were performed with PBMC samples collected at three time points, V1 (time of acute Lyme disease diagnosis and prior to starting antibiotic therapy), V2 (immediately after the completion of a 3-week course of doxycycline treatment), and V5 (6 months after the completion of therapy) (Fig. 1). Approximately 73 (Ϯ 43 [standard deviation]) million reads were generated per sample, and on average, 64.9% of the genes had nonzero counts (see Fig. S1 in the supplemental material). No significant differences in age, sex, ethnicity, or preexisting comorbidities were noted between Lyme disease patients and controls (Table 1). Two-tiered antibody testing for Lyme disease with whole-cell lysates was positive in 20 (71.4%) of 28 patients tested, with 14/28 (50%) patients testing positive at the pretreatment visit and an additional 6/28 (21.4%) seroconverting during treatment ( Table 1). The 29 Lyme disease patients were enrolled in a single season at the same geographic location, an outpatient clinic in suburban Maryland. At the 6-month follow-up visit (V5), 15 patients had fully recovered from the infection while 13 experienced persistent symptoms post-treatment, defined as new-onset fatigue, widespread musculoskeletal pain involving Ն3 joints, and/or cognitive dysfunction (11); 1 patient was lost to follow-up. Of the 13 patients with persistent symptoms, 4 were diagnosed with PTLDS on the basis of a recently proposed standardized case definition that included a documented functional decline at 6 months as a key criterion (6).
Six (40%) of the 15 patients with resolved illness and 6 (46%) of the 13 with persistent symptoms presented with early disseminated disease consisting of multiple erythema migrans (EM) lesions at the time of diagnosis (see Table S1 in the supplemental material). The average duration of acute illness, defined as the time from onset of EM rash and/or influenza-like symptoms to study enrollment and initiation of doxycycline therapy, was sig-nificantly longer in patients developing persistent symptoms (9.7 days for non-PTLDS and 19.3 days for PTLDS) than in patients with resolved illness (5.2 days) (P Ͻ 0.036) (see Table S1 in the supplemental material). In addition, the number of symptoms was significantly higher at all time points in patients with persistent symptoms than in those with resolved illness (P Ͻ 0.04) (see Table S1 in the supplemental material).
Lyme disease gene expression signature. We initially compared the transcriptomes of 29 Lyme disease patients at the time of diagnosis (V1) with those of 13 matched controls. This analysis revealed a total of 1,235 differentially expressed genes (DEGs) ( Fig. 2A; Table 2). Approximately 69% (n ϭ 847) of the DEGs were upregulated, and 31% (n ϭ 388) were downregulated ( Fig. 2A). Three weeks after diagnosis (V2), at the time of completion of a standard course of antibiotic treatment, 1,060 DEGs were found in both Lyme disease patients and controls, with 63% (n ϭ 670) upregulated and 37% (n ϭ 390) downregulated. Sixtytwo percent of the DEGs occurred at both the V1 and V2 time points (Fig. 2B). At 6 months after treatment completion (V5), the Lyme disease transcriptome did not fully return to the baseline relative to controls, with 686 DEGs, 54% (n ϭ 373) upregulated and 46% (n ϭ 313) downregulated. Partially overlapping clusters were observed for each sample category (V1, V2, V5, and controls) by principal component analysis (PCA) (Fig. 2C).
We then calculated differential gene expression between subjects with single versus multiple disseminated EM lesions and between seropositive and seronegative subjects ( four DEGs were found to be upregulated in seronegative Lyme disease patients relative to those who were seropositive, namely, HLA-DQA1, HLA-DQB1, HLA-DRB5, and NSA2 (see Fig. S3 in the supplemental material).
Pathway analyses of the Lyme disease transcriptome. Pathway analysis of Lyme disease DEGs revealed predicted activation of inflammatory response, immune cell trafficking, and hematologic system pathways at V1, as expected in the setting of the acute phase of an infection such as Lyme disease ( Fig. 2D and 3). However, the same categories also remained activated following the completion of antibiotic treatment and the clinical resolution of symptoms (V2 and V5), with the general pattern of gene expression more inhibitory at V5 (Fig. 3).
Eight, 10, and 4 of the top 10 predicted canonical pathways at V1, V2, and V5, respectively, were directly related to the host immune response (Fig. 2E). The eIF2 signaling pathway, modulating translational initiation and elongation, was found to be sig- Table S2 in the supplemental material). In contrast, TREM1-mediated activation of a Th1 proinflammatory response through upregulation of the factors DAP12, interleukin-6 (IL-6), and IL-12 was prominent at only V1 and V2 ( Fig. 2E; see Fig. S4 in the supplemental material). Multiple Toll-like receptors (TLRs) associated with inflammation and apoptosis were also found to be significantly upregulated at V1 and V2 (TLR1, -2, -4, -7, and -8) (see Fig. S4 in the supplemental material).
The most important upstream regulators in Lyme disease at V1 and V2 were found to be proinflammatory cytokines and markers (CSF2, gamma interferon [IFN-␥], IL-1, IL-6, tumor necrosis factor alpha [TNF-␣]), anti-inflammatory cytokines (IL-6, IL-10), the cell surface marker CD40L, transforming growth factor 1, the signal transduction mediator TICAM, the transcriptional regulator NF-B, and the immunoglobulin complex (Fig. 2F), with TNF-␣ shown to be a master regulator of eIF2 signaling, Comparison of acute Lyme disease with other infections. We compared our V1 RNA-Seq data, derived from patients with acute Lyme disease, to 12 available, previously published transcriptome data sets from cell culture models of in vitro infection or from patients with acute viral and bacterial infections other than Lyme disease (Fig. 4). Unsurprisingly, the greatest overlap in shared DEGs (44%) was observed with in vitro B. burgdorferi infection of human PBMCs (44%), followed by infection of human endothelial cells (29%), human neuroblastoma cells (13%), or primary monkey glial cells (11%) (Fig. 4A). We also compared DEGs from acute Lyme disease patients with those corresponding to human PBMCs stimulated in vitro by lipopolysaccharides (LPS), infected TNF IFNG Ig CSF2 IL6 IL1B TGFB1 IL10 CD40LG IL4 TNF IFNG CD40LG NFkB IL1B TICAM1 Ig CSF2 TNFSF11 RELA MYCN HOXA7 SATB1 HNF4A MYC VDAC1 GGA3 ELAVL1 APC Note that the PC3 axis in the PCA plot accounts for only 8% of the variance in the data set. (D to F) Top 10 disease and functional categories (D), top 10 canonical pathways (E), and top 10 upstream regulators (excluding drug categories) (F) predicted to be involved in Lyme disease at (V1, V2, and V5) with categories, pathways, and genes ranked by the negative log of the P value of the enrichment score. The color scheme is based on Z scores, with activation in orange, inhibition in blue, and undetermined directionality in gray. The red line represents the designated significance threshold (P Ͻ 0.05).
by the fungal pathogen Candida albicans, or infected by two tickborne bacterial agents, i.e., Francisella tularensis (tularemia) and Anaplasma phagocytophilum (anaplasmosis). Interestingly, stimulation of PBMCs by LPS (39%) resulted in a greater overlap of shared DEGs than in vitro infection with C. albicans (27%), F. tularensis (28%), or A. phagocytophilum (15%) (Fig. 4A). Next, we compared the acute Lyme disease transcriptome at V1 to transcriptomes corresponding to other acute infectious syndromes ( Fig. 4A and B). Patients with acute influenza had 35% of their DEGs in common with Lyme disease patients, while they had only 28, 26, and 21% of their DEGs in common with patients with bacteremia due to Staphylococcus aureus, Streptococcus pneumoniae, and Escherichia coli, respectively.
To determine if these infectious diseases have pathways in common, we then compared the top canonical pathways significantly involved in acute Lyme disease at V1 (P Ͻ 0.05) with those corresponding to other clinical infectious diseases or in vitro cell culture models of infection. The greatest number of shared pathways was observed with S. pneumoniae bacteremia (82%) and in vitro F. tularensis (80%)-and C. albicans (78%)-infected cell cultures. Strikingly, downregulation of eIF2 signaling was restricted to Lyme disease and not found in other infectious diseases or in vitro cell culture infection models (Fig. 4B). In contrast, TREM1 and TLR pathways were activated in all five infections, whereas upregulation of IFN signaling pathways was predicted in only Lyme disease and influenza. B-cell development and downregulation of calcium-induced T-cell apoptosis, prominent in other viral and bacterial infections, were not found to be significant pathways in acute Lyme disease (Fig. 4B).
Pathway analysis of Lyme disease patients at 6 months posttreatment and comparison with chronic disease syndromes. We next compared the PBMC transcriptome from all treated Lyme disease patients at V5 (both resolved illness and persistent symptoms) to publicly available transcriptome data sets from patients diagnosed with chronic illnesses that have some symptoms in common with PTLDS, including chronic fatigue syndrome (CFS), systemic lupus erythematosus (SLE), and rheumatoid arthritis (RA). The percentages of shared DEGs and pathways ranged from 9 to 18% and from 31 to 60%, respectively (Fig. 4C). Inhibition of eIF2 signaling was common to Lyme disease, SLE, and (to a lesser extent) RA ( Fig. 4D; see Fig. S5 in the supplemental material). Glutathione-mediated detoxification and IL-6 signaling pathways were found to be specific to Lyme disease patients. No additional differences were revealed by subset analyses of Lyme disease patients manifesting persistent symptoms (non-PTLDS and PTLDS) and patients with these chronic illnesses.
Comparison of differential gene expression between Lyme disease patients with resolved versus persistent disease. No DEGs were significantly expressed at any single time point between patients with resolved Lyme disease (n ϭ 15) and patients with persistent symptoms, including both non-PTLDS and PTLDS patients (n ϭ 13) ( Table 2). A single DEG (GPR15) was found at V1 when patients with resolved Lyme disease (n ϭ 15) were compared with patients with PTLDS (n ϭ 4). When all of the time points were combined, a total of four different DEGs overall were identified in Lyme disease patients with persistent symptoms (non-PTLDS and/or PTLDS) compared to those with resolved disease, i.e., MIAT, CCDC163P, ZNF266, and GPR15.
DISCUSSION
We provide the first transcriptome analysis of B. burgdorferi infection in Lyme disease patients, revealing a gene signature that persisted for at least 3 weeks following the acute phase of infection and had fewer than 44% DEGs in common with other infectious (13) Resolved Lyme disease (15) Control ( or noninfectious syndromes. Notably, no differences in gene expression were observed between Lyme disease patients with resolved illness and those with persistent symptoms at 6 months, although larger cohort studies are needed to confirm this finding. The identification of a distinct and sustained transcriptome signature in early Lyme disease may facilitate the development and validation of human gene expression biomarker panels to improve diagnostic testing in the future, in parallel with other published studies investigating cytokine (12) or metabolic (13) biosignatures.
To define the longitudinal transcriptome profile of patients with acute Lyme disease at 0 weeks, 3 weeks, and Ͼ6 months, unbiased RNA-Seq analysis was employed with the goal of investigating the molecular basis of early and convalescent-phase Lyme disease. Potential advantages of RNA-Seq relative to microarrays include detection of low-abundance transcripts, a broader dynamic range in detecting fold changes in gene expression, unbiased detection of novel isoforms and transcripts, and elimination of hybridization-based limitations such as background noise, saturation, and probe redundancy (14). However, the utility of RNA-Seq data is dependent on a number of factors, including the number and quality of samples, sequencing depth, and designated thresholds for gene expression and differential analyses. In this study, we estimated the statistical power as 98% when analyzing samples at all three time points combined, 78% for samples collected at a single time point, and only 62% when considering a stratification of the Lyme disease cohort according to PTLDS status, serology, or the presence of disseminated lesions (15). The finding of a profound and sustained change in the transcriptome of acute Lyme disease patients refutes the idea that treatment and resolution of the infection result in a prompt return to a transcriptional baseline, as typically seen in the acute phase of other infections (16). In addition, failure to return to a transcriptome baseline cannot be accounted for solely by patients with persistent symptoms, given that no DEGs were found comparing Lyme disease patients with resolved illness to those with persistent symptoms ( Table 2). Persistent transcriptional changes may be characteristic of not only Lyme disease but also a number of other infections. For example, viral clearance in hepatitis C patients did not result in normalization of the baseline transcriptome (17 our knowledge, this is first time that sustained changes in the human host transcriptome have been reported for a bacterial infection after treatment with appropriate antibiotics. Persistence of such a signature for at least 3 weeks following infection suggests that a clinical diagnostic test for acute Lyme disease based on host gene expression is feasible. Such a test would also directly address the current diagnostic gap created by the "window period" between acute Lyme disease infection and the subsequent appearance of detectable antibody. Infection by B. burgdorferi drives a complex immune response with robust inflammation and overt clinical signs and symptoms in early stages of the disease (7). The eIF2 signaling pathway, found to be downregulated here during all stages of Lyme disease, plays a central role in protein synthesis in response to cellular stress (18). Intracellular bacterial pathogens such as Legionella pneumophila encode effectors that actively disrupt and downregulate the eIF2 signaling pathway (19). However, Borrelia spirochetes are not known to enter cells during infection in vivo, nor are they thought to express such effectors (20). Consistent with a previous report (21), the eIF2 pathway in this study was also found to be downregulated in SLE as well as PTLDS patients. Inhibitors of the eIF2 pathway have been reported as potential therapeutic drugs for inflammatory bowel disease, and further studies are needed to assess whether eIF2 inhibitors may constitute potential targets for inflammatory sequelae of Lyme disease (22). Nevertheless, it remains to be determined whether inhibition of the eIF2 pathway in Lyme disease patients is caused directly by Borreliamediated immune dysregulation or is strictly a host response mechanism to limit tissue injury.
The prominent TREM1 signaling in acute Lyme disease observed here is consistent with previously published in vitro gene expression data of B. burgdorferi infection of human neural and primary monkey glial cell lines (23). TREM1 acts as an amplifier of the immune and inflammatory response in vivo (24), and modulation of TREM1 has been shown to impact a number of inflammatory conditions, including septic shock, and acute dengue virus infection (25,26). Our data also showed upregulation of more TLRs (TLR2, TLR4, TLR7, and TLR8) in acute Lyme disease than previously described (27). This broad upregulation is likely to be indirect, reflecting a general increase in TLR regulatory activity rather than direct association of TLRs with B. burgdorferi proteins. In the present study, TNF-␣ was predicted to be a common upstream regulator of the eIF2, TREM1, and TLR signaling pathways. Notably, anti-TNF-␣ therapy has been proposed to reduce inflammation in the Jarisch-Herxheimer response to Borrelia recurrentis infection (28), and treatment was previously reported to be clinically efficacious in 4/4 patients with antibiotic-refractory Lyme arthritis (29).
Comparisons with 15 previously published transcriptome data sets found that the greatest overlap of DEGs (44%) was with the transcriptome of PBMCs stimulated with B. burgdorferi in vitro. Although this observation is to be expected, given the same infectious agent and cell type, the only partial overlap likely reflects differences between in vivo or in vitro B. burgdorferi infections and underscores the critical importance of analyzing "real-life" clinical samples from patients in studies of disease pathogenesis. Given the lymphocytic infiltrates characteristic of Lyme disease, in contrast with the suppurative lesions common to many bacterial infections (1), it is perhaps not surprising that the percentage of DEGs in Lyme disease patients also found in patients with acute influenza was greater than that of DEGs also found in patients with other bacterial infections. Among bacterial infections, infection with S. pneumoniae had the highest number of top canonical pathways in common with acute Lyme disease, consistent with similarities in virulence factors shared by S. pneumoniae and B. burgdorferi, such as lipoproteins, that produce shared IgMmediated immunological responses (30).
Importantly, Lyme disease patients did not show any changes in the calcium-dependent T-cell apoptosis pathway, in contrast to the marked downregulation observed in other bacterial and viral diseases (Fig. 4B). In addition, an absence of significant DEGs linked to B-cell development in Lyme disease relative to other infections was observed. These findings suggest that Lyme disease may be associated with a smaller proportion of B and T cells in peripheral blood than other diseases. Interestingly, suppression of long-lived humoral responses has been observed in a mouse model of Borrelia infection (31). The absence of DEGs corresponding to B-cell maturation may also potentially explain why prior infection with B. burgdorferi is associated with a serological response yet does not appear to confer immunity to reinfection. Certain alleles of HLA genes have been previously reported to be associated with serological responses to Lyme disease infection (32). Here we found that upregulation of certain HLA genes (HLA-DQA1, HLA-DQB1, HLA-DRB5) is associated with seronegativity in Lyme disease and may thus constitute potential diagnostic biomarkers for seronegative patients.
Following the acute phase of infection, recent treatment trials among patients with EM have estimated that approximately 10 to 20% of patients treated for Lyme disease experience lingering symptoms that may progress to PTLDS, although the incidence can be as high at 50% (4). The pathogenetic mechanisms of PTLDS remain unknown, but autoantigens and/or central nervous system sensitization have been postulated to play a role (10,(33)(34)(35). In our study, the relatively large proportion of posttreatment Lyme disease patients with persistent symptoms of fatigue, widespread musculoskeletal pain, and/or cognitive dysfunction (13 [46.4%] of 28) can be potentially accounted for by more stringent enrollment criteria at the time of presentation (requiring the presence of EM and concurrent influenza-like symptoms rather than EM alone). This may have resulted in the selection of patients with more severe disease and thus with an increased likelihood of persistent symptoms (36). Of note, according to the proposed formal case definition for PTLDS, which requires a functional decline in patients in addition to lingering symptoms, only 4 (14.3%) of our 28 patients met all of the criteria, within the range of the 10 to 20% frequency reported in the literature (4).
Notably, Lyme disease at 6 months post-treatment (V5) had 60 and 31% of their predicted pathways overall in common with SLE and RA, respectively. Circulating immune complexes have been identified as features common to all three conditions (37,38). Symptoms of fatigue and cognitive impairment occur in a variety of chronic syndromes, including SLE, CFS, and PTLDS. Although some pathways were common to Lyme disease at V5 and CFS, melatonin signaling, prominent in CFS, was not predicted to be involved in Lyme disease (Fig. 4D). As melatonin is a hormone that regulates the circadian rhythms of the sleep-wake cycle and thus is strongly linked to fatigue, the absence of increased melatonin signaling suggests that the fatigue in Lyme disease patients with persistent symptoms is related to a different mechanism.
Overall, our results, showing only 18% of the DEGs and 34% of the pathways common to CFS and Lyme disease, are consistent with a proteomic study of cerebrospinal fluid that clearly discriminates between the two conditions (39).
Transcriptome analysis of Lyme disease patients with persistent symptoms (non-PTLDS and/or PTLDS) versus those with resolved illness revealed an absence of DEGs at each of the three time points, with the sole exception of a single gene (GPR15), which was upregulated at V1 in PTLDS patients relative to controls. Possible explanations for the overall lack of observed differences include (i) lack of statistical power from low sample numbers, (ii) sampling at designated time points instead of during periods of peak symptomatology, and (iii) that transcriptome profiling of PBMCs in blood is insufficient to discriminate between Lyme disease patients with persistent symptoms and those with resolved illness. Larger studies with increased sampling resolution are needed to establish whether there are indeed any detectable differences in gene expression between these two groups.
Patient information.
Patient enrollment, collection of clinical data and biological samples, and analysis of clinical samples by transcriptome profiling were done under protocols approved by the Institutional Review Boards of Johns Hopkins University and the University of California, San Francisco. Written informed consent was received from all participants prior to inclusion in this study.
All 29 participants with Lyme disease included in this study presented with a physician-documented EM rash of Ն5 cm and concurrent influenza-like symptoms that included at least one of the following; fever, chills, fatigue, headache, and/or new muscle or joint pains. At the time of enrollment, all of the participants with Lyme disease were treatment naive and subsequently underwent 3 weeks of doxycycline therapy between the first and second follow-up visits. All 29 subjects with Lyme disease were enrolled at the same geographic location (an outpatient clinic in Maryland) in a single season, from 1 May to 23 November 2009, with follow-up visits 3 weeks and 6 months after the first visit. Controls were matched by age and gender and enrolled from the same physician practice as case participants and across different seasons to account for seasonal variations in the transcriptome. Two-tier antibody testing for Lyme disease by whole-cell sonicate enzyme immunoassay, followed by IgM/IgG Western immunoblot assays, was performed for all patients and controls by a clinical reference laboratory (Quest Diagnostics). Seropositivity was assessed according to established CDC criteria (40) by the investigators (A. R. and J. A.) who enrolled and provided clinical care for the patients enrolled in this study. All control subjects were required to have a negative Lyme disease antibody test in order to be enrolled in this study. We screened both patients and controls prior to enrollment for a history of chronic fatigue, fibromyalgia, autoimmune, immunodeficiency, neurologic, psychiatric, and malignancy disorders, in which case they were excluded from the study. Prospective case patients and controls were also excluded if they had a prior documented history of Lyme disease and/or if they had previously received the Lyme disease vaccine.
Controls were enrolled primarily during the winter and spring seasons, while most Lyme disease patients were enrolled in summer during the peak season for tick bites, a difference that was statistically significant (P Ͻ 0.03) ( Table 1). Nonetheless, the differences in seasonal sampling did not result in gene expression bias, as shown by the absence of seasonal clustering by PCA of the overall gene expression of the 13 controls (see Fig. S2b in the supplemental material). In addition, an intragroup comparison of eight controls sampled during the winter and five controls sampled during other seasons did not yield any significant DEGs (Table 2).
PBMCs from whole-blood samples at V1 (the acute phase of infection, prior to initiation of antibiotic treatment), V2 (3 weeks later, at the time of treatment completion), and V5 (6 months following treatment comple-tion) were analyzed in this study (Fig. 1A). V2 and V5 were specifically chosen for analysis because fever and rash from acute Lyme disease typically resolve by completion of treatment (V2), while chronic persistent symptoms are clinically apparent after 6 months (V5).
The presence of persistent symptoms in Lyme disease patients at V5 was assessed by using a standardized case definition proposed by the Infectious Diseases Society of America (6,11) that incorporates the presence of at least one of the following: new-onset fatigue, widespread musculoskeletal pain, or cognitive dysfunction. For a diagnosis of PTLDS, patients were also required to have a composite score of Յ45.00 on four subscales of the SF-36, a measurement of health-related quality of life (6) (Fig. 1B). The chi-square test was used to evaluate the statistical significance of differences between independent samples in one or more categorical variables, while Welch's t test was used for continuous variables.
Sample processing. PBMCs were isolated from fresh whole blood with Ficoll (Ficoll-Paque Plus; GE Healthcare), and total RNA was extracted from 10 7 PBMCs with TRIzol reagent (Life Technologies). mRNA was isolated with the Oligotex mRNA minikit (Qiagen). The ScriptSeq RNA-Seq library preparation kit (Epicentre) was used to generate RNA-Seq libraries according to the manufacturer's protocol. Libraries were sequenced as 100-bp paired-end runs on a HiSeq 2500 (Illumina). One hundred samples from the first cohort (29 patients at three time points and 13 control subjects, matched by age, sex, and geography) were mixed and blindly processed in three batches. Three samples, 01-36_V2, 01-42_V2, and 01-51_V1, were not included in the pooled analysis because of insufficient read counts and transcriptome coverage (see Fig. S1 in the supplemental material). No batch effect was observed by PCA of the global expression of all 25,278 genes (see Fig. S6 in the supplemental material).
Next-generation sequencing data analysis. Paired-end reads were mapped to the human genome (hg19), followed by annotation of exons and calculation of FPKM (fragments per kilobase of exon per million fragments mapped) values for all 25,278 expressed genes with version 2 of the TopHat-Cufflinks pipeline (41). Differential expression of genes was calculated by using the variance modeling at the observational level transformation (42), which applies precision weights to the matrix count, followed by linear modeling with the Limma package (43). Genes were considered to be differentially expressed when the change was greater than Ϯ1.5-fold, the P value was Ͻ0.05, and the adjusted P value (or falsediscovery rate, FDR) was Ͻ0.1% (44). Pathway and network analyses of the transcriptome data were performed with Ingenuity Pathway Analysis (IPA) software (Qiagen) (45). The molecule activity predictor tool in the IPA software was used to predict the upstream and/or downstream activation or inhibition of a given pathway. The P value of the enrichment score was used to evaluate the significance of the overlap between observed and predicted gene sets, while the activation Z score was used to assess the match between observed and predicted patterns of upregulation and downregulation. The statistical significance of the difference in gene expression levels was determined with Welch's t test for independent samples by two-group comparisons. The statistical power for the transcriptome study was determined according to the algorithm developed by Hart et al. (15), with the use of a generalized linear model on normalized FPKM data instead of a negative binomial distribution on raw gene count data. The generalized linear model has been reported to be more reliable for differential analysis of data sets with small sample sizes (41,43).
Comparison of RNA-Seq and microarray data. Microarray transcriptome data were downloaded from public servers ( [URL]:// www.ncbi.nlm.nih.gov/geo) and include expression sets GSE12108, GSE2405, GSE42606, GSE8650, GSE6269, GSE6092, GSE14577, and GSE15573 (46)(47)(48)(49)(50)(51)(52). Raw data were extracted and preprocessed by using the Robust Multichip Average algorithm (53). Differential expression was calculated with the Limma package (43), which is applicable for analysis of both RNA-Seq and microarray data (42,43). Genes were considered to be differentially expressed when the change was greater than Ϯ1.5-fold, the P value was Ͻ0.05, and the FDR was Ͻ0.1%, in accordance with conven-tional thresholds (44). Microarray data were not available for one study of in vitro B. burgdorferi infection (23), so tables of DEGs were used as provided instead, incorporating a change of greater than Ϯ1.5-fold as a threshold cutoff for differential expression.
Data availability. All of the transcriptome data obtained in this study have been submitted to the Gene Expression Omnibus data repository under accession number GSE63085.
==
Domain: Biology Medicine
|
[{"added": "2016-09-24T08:40:55.662Z", "created": "2016-02-12T00:00:00.000", "id": "4821818", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://mbio.asm.org/content/mbio/7/1/e00100-16.full.pdf", "pdf_hash": "74362f59d74e61950f256da318f82aecd144142e", "pdf_src": "Highwire", "provenance": "peS2o-0075.json.gz:38979", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "1d14e184b57f8e2a572caa6d53d3da93b8d5e264", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Protective Immunity of the Primary SARS-CoV-2 Infection Reduces Disease Severity Post Re-Infection with Delta Variants in Syrian Hamsters
The Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) Delta variant has evolved to become the dominant SARS-CoV-2 lineage with multiple sub-lineages and there are also reports of re-infections caused by this variant. We studied the disease characteristics induced by the Delta AY.1 variant and compared it with the Delta and B.1 variants in Syrian hamsters. We also assessed the potential of re-infection by these variants in Coronavirus disease 2019 recovered hamsters 3 months after initial infection. The variants produced disease characterized by high viral load in the respiratory tract and interstitial pneumonia. The Delta AY.1 variant produced mild disease in the hamster model and did not show any evidence of neutralization resistance due to the presence of the K417N mutation, as speculated. Re-infection with a high virus dose of the Delta and B.1 variants 3 months after B.1 variant infection resulted in reduced virus shedding, disease severity and increased neutralizing antibody levels in the re-infected hamsters. The reduction in viral load and lung disease after re-infection with the Delta AY.1 variant was not marked. Upper respiratory tract viral RNA loads remained similar after re-infection in all the groups. The present findings show that prior infection could not produce sterilizing immunity but that it can broaden the neutralizing response and reduce disease severity in case of reinfection.
Introduction
The B.1.617.2 lineage of Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) was first detected in India on 22 September 2020 [1]. The variant was later categorized as a variant of concern (VOC) and was named the Delta variant by the World Health Organization on 31 May 2021 [2]. The variant spread at an alarming rate to become the most dominant SARS-CoV-2 lineage circulating globally and spread to 201 countries by 2 December 2021 [2]. The amino acid substitutions in the spike protein of the Delta variant, such as D614G, T478K, P681R and L452R, are known to affect transmissibility and neutralization [3]. The variant was responsible for the rise in Coronavirus disease 2019 (COVID- 19) cases in 2021 in many countries, including India, the United Kingdom, Fiji, South Africa, parts of Asia, the United States, Australia and New Zealand [2]. The Delta variant has been further subdivided into Delta AY.1 and AY.127 according to the Pango lineage designation system. Among these sub-lineages, AY.1 and AY.2 possess the K417N substitution, which is also present in the B.1.351 variant, suggesting that it plays a role in immune evasion. As of 6 December 2021, the AY.1 lineage has been detected in at least 43 countries and AY.2 in 8 countries. AY.1 has amino acid substitutions at T19R, E156G, 157/158 del, W258L, K417N, L452R, T478K, D614G, D950N and P681R [1]. The information on the biological characteristics of these sub-lineages, including transmissibility, disease severity, and immune evasion, are still unknown.
SARS-CoV-2 generates a neutralizing antibody (NAb) response after infection in humans, but the protective immune titre required to prevent subsequent infection is not yet known [4]. In the case of other human coronaviruses (HCoV), waning of immunity is observed in 1 to 3 years and re-infection events have been reported as a common feature of HCoV-NL63, HCoV-229E, HCoV-OC43 and HCoV-HKU1 [5,6]. After natural SARS-CoV-2 human infection, immune response is suspected to persist for about 90 days in most patients [4]. SARS-CoV-2 re-infection cases with varied disease severity have been reported in many countries [7][8][9][10][11]. The speculated reasons for re-infection are infection with a higher virus dose/another virulent strain, antibody-dependent enhancement and waning of immune response [8]. Laboratory studies have shown that the duration of infection-acquired immunity is inconsistent and responses against VOCs also differ [12]. The risk of re-infection also depends on host susceptibility, vaccination status and exposure to COVID-19 patients during the infectious phase [13]. Understanding the potential risk of re-infection is important in improving COVID-19 prevention and control measures. Re-infection studies in the population have not been widely reported and the rate of reinfection is also unclear [14]. There are few reports of aggravated disease severity in the case of Delta variant re-infection [15,16]. The impact of immunity on the threat of re-infection posed by different variants still needs to be understood.
Animal models are important in understanding virus properties, disease pathogenesis, measuring the efficacy of countermeasures, etc. Syrian hamsters have been widely used in studying SARS-CoV-2 disease characteristics. Our previous studies have revealed the pathogenicity and immune-evasive properties of the Delta variant in hamsters [17].
Few animal model studies have demonstrated that the NAb response generated after primary SARS-CoV-2 infection can reduce the viral load and severity of re-infection [18,19]. The degree of protection against re-infection by Delta lineage variants is still unclear. A recent study which evaluated the protection offered by infection-acquired immunity after 15 months in hamsters showed protection against re-infection with the Delta variant and prevention of transmission to naive hamsters [20].
The SARS-CoV-2 B.1 variant possessing the D614G mutation in the spike protein was found to be highly transmissible and became the predominant variant during the early phase of the pandemic [21]. High upper respiratory tract viral load with a lower disease severity has been reported for this variant [21,22]. This ancestral variant with only the D614G mutation in the spike protein has been used as a comparator virus in multiple research studies [22,23]. We have also used the B.1 variant in the present study for the comparison of the disease characteristics of other variants.
Here we have studied the pathogenicity of the Delta AY.1 variant in comparison with the Delta and B.1 variants and also assessed the re-infection potential of these variants in B.1 infection-recovered hamsters 3 months after initial infection. In addition, the neutralization potential of the infected hamster sera was assessed against the B.1, Delta, Delta AY.1, and Beta variants.
Animal Experiments
The experiments were performed in the Containment Facility of ICMR-National Institute of Virology, Pune. Three study groups of 17 female Syrian hamsters (procured from a CPCSEA-authorized breeding facility) 12-14 weeks old were included in the study to assess pathogenicity/virus shedding. A virus dose of 10 5 TCID50 (0.1 mL volume) of the Delta/Delta AY.1/B.1 variants was used intranasally to inoculate the hamsters (Figure 1a). Throat swab, nasal wash and faeces samples (n = 7) were collected on alternate days during the study period. Hamsters were observed for a period of 14 days for body weight loss, and five hamsters/group were sacrificed 3, 7 and 14 days post infection (DPI) to collect organs (lungs, nasal turbinates, heart, liver, kidney, intestine, spleen and brain) and blood samples. For the re-infection study, 12 female hamsters, 16-18 weeks old, that were previously infected with the B.1 variant of SARS-CoV-2 (with an infectious dose of 10 4.5 TCID50) were used 3 months after initial infection (Figure 1b). IgG response and NAb levels were assessed and the animals were randomly divided into three groups (four animals per group). The hamsters were re-infected with the Delta/Delta AY.1/B.1 variants with a virus dose of 10 5 TCID50 (0.1 mL volume intranasally). Throat swab, nasal wash and faeces samples were collected on 2, 4, 6 DPI and body weight change was monitored for 7 days. The hamsters were sacrificed on 7 DPI to collect lungs, nasal turbinates and blood samples.
Viral Load Estimation
Nasal wash, throat swab and organ tissue samples were used for viral load estimation. Organ samples collected during necropsy were weighed and homogenized in sterile media using beads in a tissue lyser machine (Qiagen, Hilden, Germany). The lysate was used for RNA extraction using the MagMAX™ Viral/Pathogen Nucleic Acid Isolation Kit (Thermo Fisher Scientific, Waltham, MA, USA) as per the manufacturer's instructions. Quantitative real-time RT-PCR was performed for the E gene of SARS-CoV-2 using published primers to estimate the genomic viral RNA (gRNA) load and for the N gene of SARS-CoV-2 using published primers to estimate the subgenomic viral RNA (sgRNA) load [24,25]. The lung samples collected 7 days post primary infection and re-infection were used for virus titration in Vero (ATCC ® CCL-81 TM ) cells (ATCC, Manassas, VA, USA). Lung tissue homogenates were centrifuged at 1984× g for 10 min and 0.1 mL of the supernatant was used for the titration. The supernatant was added onto 24-well tissue culture plate cell monolayers and incubated at 37 • C. The cells were washed with phosphate-buffered saline after the incubation period of one hour. Maintenance media containing 2% fetal bovine serum (Sigma Aldrich, St. Louis, MO, USA) was added onto the cells and further incubated in a CO 2 incubator at 37 • C. The cells were examined for cytopathic effects for 4 days. The titres were determined by the Reed and Muench method.
Anti-SARS-CoV-2 IgG Detection
The serum samples were tested for IgG antibodies by an in-house developed qualitative ELISA [26]. Briefly, inactivated SARS-CoV-2 antigen/Vero (ATCC ® CCL-81 TM ) cell lysate-coated microtitre plates were blocked with liquid plate sealer. The diluted hamster sera samples (1:100 to 1:10,000) were added and incubated for 60 min at 37 • C. The plates were washed following incubation and 1:3000 dilution of anti-hamster IgG horseradish peroxidase (Thermo Fisher Scientific, Waltham, MA, USA) was added and incubated for 60 min. The plates were washed and substrate was added to each well for color development. The reaction was terminated with sulfuric acid and the absorbance was measured at 450 nm using an ELISA reader. The assay was performed in duplicate and the assay cutoff was set at an optical density (OD) value of 0.2 and positive/negative ratio of 1.5.
Serum Neutralizing Antibody Level Estimation
A plaque reduction neutralization test (PRNT) was performed against the B.1, Delta, Delta AY.1 and Beta variants, as described previously [27]. Diluted sera were mixed with 50-60 plaque forming units/0.1 mL virus and the virus-sera mixture was incubated for 60 min. The mixture was then added on to a tissue culture plate with a Vero (ATCC ® CCL-81 TM ) monolayer (ATCC, Manassas, VA, USA). After 60 min, the mixture was aspirated and media with 2% carboxymethyl cellulose with 2% fetal bovine serum (Sigma Aldrich, St. Louis, MO, USA) was added. After an incubation period of 4 days, the media was decanted and amido black staining was performed. The plaques were counted and PRNT50 titres were calculated.
Serum Cytokine Level Estimation
ELISA-based estimation (Immunotag, St. Louis, MO, USA) was performed to assess the levels of IL-4, IL-6, IL-10, IFN-γ and TNF-α in hamster sera samples as per the manufacturer's instructions.
Lung Histopathological Evaluation
Formalin-fixed lung tissue samples were processed using an automated tissue processor and were stained by routine hematoxylin and eosin staining. The samples were coded and blindly scored by a pathologist. The bronchiolar (degeneration, epithelial loss), alveolar parenchymal (edema, exudation, mononuclear infiltration, emphysema, pneumocyte hyperplasia, septal thickening) and vascular lesions (congestion, hemorrhages, perivascular infiltrations) were graded for severity on a score from 0 to 4 (0 = no changes, 1 = minimal, 2 = mild, 3 = moderate, 4 = severe). The cumulative scores for each group on 3, 7 and 14 DPI were compared and statistically analyzed. The scores of the re-infected animals on 7 DPI were compared with the scores of the primary infection group on 7 DPI.
Data Analysis
GraphPad Prism version 9.2.0 (GraphPad software Inc., San Diego, CA, USA) software was used for the descriptive statistics and statistical analysis. Nonparametric Mann-Whitney tests were used for the analysis. p-values less than 0.05 were considered statistically significant. For the primary infection study, the comparison was performed daywise among Delta, Delta AY.1, and B.1 infected groups for viral load, virus shedding, body weight loss, NAb titres and histopathological scores. For cytokine response, the comparison was performed with the uninfected control hamster sera. For the re-infection study with each variant, the comparison was performed with the data for the primary infection of the corresponding variant.
Body Weight Changes in Hamsters after Primary Infection
The Delta AY.1-and B.1-infected group animals showed the maximum weight loss [mean ± standard deviation (SD)] of −2.048 ± 8.557% and −0.566 ± 9.432%, respectively, on 14 DPI, whereas in the Delta variant-infected group, the peak weight loss (−10.886 ± 3.46%) was observed on 8 DPI. The body weight loss observed in the Delta AY.1 group was significantly lesser in comparison to the Delta variant infection (Figure 2a).
Immune Response in Hamsters after Primary Infection
Anti-SARS-CoV-2 IgG response could be observed from day 7 in all primary infected groups (Figure 2b). A mean optical density (OD) ± SD of 0.65 ± 0.45, 1.00 ± 0.67, 1.02 ± 0.40 on 7 DPI and 1.25 ± 0.55, 0.62 ± 0.13 and 0.62 ± 0.11 on 14 DPI for the Delta AY.1, Delta and B.1 variant, respectively, was observed with 1:100 dilution of sera. Cross NAbs were detected in hamster sera against the Delta AY.1, Delta, B.1 and Beta variants (Figure 2c). The mean ± SD of NAb titre on 14 DPI in the Delta AY.1-infected group against the Delta AY.1, Delta, B.1 and Beta variants were 5265 ± 1504, 4544 ± 2824, 4159 ± 3062 and 3735 ± 2804, respectively. In the case of the Delta-and B.1-infected groups, mean NAb responses against the Delta AY.1, Delta, B.1 and Beta variants were 2848 ± 978, 2667 ± 1275, 680 ± 234, 73 ± 57 and 1020 ± 59, 485 ± 269, 925 ± 398 and 109 ± 77, respectively. Thus, a significant reduction in NAb titre against the Beta variant was observed with the Delta and B.1 variant infected animal sera, whereas a comparable response was observed in the case of the AY.1-infected group. The serum IL-6 level was elevated in all the infected groups. The B.1 group hamsters showed a significantly higher IL-6 level on 3, 7 and 14 DPI than the Delta group on 14 DPI (Supplementary Figure S1). In the case of the Delta AY.1 group, the increase was not statistically significant. Among the other cytokines analyzed in serum, i.e., IL-4, IL-10, IFN-γ and TNF-α, no significant increase was observed in comparison to the control animal sera.
Viral Shedding in Hamsters after Primary Infection
In the throat swab and the nasal wash samples, genomic RNA (gRNA) was detected with a decreasing trend until 12 DPI in the Delta AY.1-and the B.1-infected groups and until 14 DPI in the Delta variant-infected group. The average viral gRNA levels in the throat swab were significantly lower in the Delta AY.1-infected group on 6, 8 and 10 DPI, as were the subgenomic (sg) RNA levels on 8 DPI in comparison with the Delta variant group (Figure 3a,b). The nasal wash and faeces viral RNA loads were also significantly lower in the Delta AY.1-infected group (Figure 3c-f).
Viral Load in Organs in Hamsters after Primary Infection
In the nasal turbinates, gRNA and sgRNA could be detected until 14 DPI in all the groups. The Delta AY.1 group did not show any significant differences in the gRNA or sgRNA levels of the nasal turbinates in comparison with the other groups (Supplementary Figure S2a,b). gRNA and sgRNA was detected in the lungs of hamsters in both the Delta and B.1 groups until 14 DPI. In the Delta AY.1 group, a comparatively lower viral load (mean ± SD = 4.55 × 10 4 ± 9.0 × 10 4 ) was observed on 7 DPI in the lungs, and complete clearance was seen by 14 DPI (Supplementary Figure S2c,d). Other organs, such as the brain (1/5), heart (2/5) and large intestine (2/5), in hamsters in the Delta AY.1-infected group showed sgRNA positivity on 3 DPI but none of the hamsters in the Delta-and B.1-infected groups showed positivity in non-respiratory organs.
Lung Pathological Changes after Primary Infection
Grossly, the lungs of 2/10 animals in the Delta AY.1 group and of 7/10 animals in the Delta variant group sacrificed on 7 and 14 DPI after the primary infection showed hemorrhagic lesions. A few focal hemorrhages were only seen in the case of B.1 infection. On 3 DPI, the vascular pathological changes in the lungs were minimal in all groups, and mild pneumonic changes were observed in the alveolar parenchyma. The pneumonic changes were minimal in the Delta AY.1 group on 7 DPI, which became pronounced by 14 DPI (in three out of five animals infected). By 7 DPI, inflammatory changes became severe in the Delta variant group, characterized by severe congestion/hemorrhages, alveolar consolidation, loss of bronchial epithelium, septal thickening, pneumocyte hyperplasia and cellular infiltration in the alveolar interstitial space, peribronchial and perivascular area. In the case of B.1-infected hamsters, pneumonic changes became more pronounced by 7 DPI. The highest lung cumulative score was observed in the Delta-infected group on 7 DPI (Supplementary Figure S2e).
Reduced Disease Severity in Hamsters Post Re-Infection
After re-infection, body weight loss in all the infected groups was minimal irrespective of the variant used for infection (Figure 4a Figure S3).
In the re-infected animals, only focal hemorrhagic foci were seen in the lungs, contrary to the pronounced gross lesions seen with primary infection (Supplementary Figure S4). The lung-body weight ratio, which was found to have increased in the Delta variant hamsters after primary infection, was found not to have increased in the re-infected hamsters (Supplementary Figure S5).
In the case of the Delta-and B.1-re-infected group, the lung pathological changes observed on 7 DPI were milder in comparison to those observed with primary infection. The lungs from the B.1-re-infected group showed mild alveolar parenchymal inflammatory cell infiltration, septal thickening and bronchiolar epithelial loss, whereas the lungs from the hamsters of the Delta variant-re-infected group showed mild bronchial/vascular changes and moderate alveolar changes (consolidation, alveolar septal thickening and pneumocyte hyperplasia). In the case of AY.1-re-infected animals, similar disease severity to that of two out of five naive AY.1-infected animals were observed, characterized by mild congestion/hemorrhages and moderate alveolar parenchymal consolidation, septal thickening and pneumocyte hyperplasia, as well as mononuclear cellular infiltration ( Figure 5).
Reduced Viral RNA Shedding and Lung Viral RNA Load in Re-Infected Hamsters
The viral RNA shedding through the nasal and oral cavity was reduced post reinfection in the hamsters. On 2, 4 and 6 days post re-infection, viral RNA shedding in the throat swabs and nasal washes was significantly lesser in the Delta-and B.1-re-infected hamsters in comparison to the viral RNA load observed in hamsters post primary infection (Figure 6a-f). In the faeces samples of the B.1-and Delta-re-infected groups, also, the reduction was evident. Although the viral RNA load in the AY.1 group showed a reduction, the values were not statistically significant. Unlike the viral RNA shed through the nasal cavity, the viral RNA load in the nasal turbinates was comparable in both the re-infected and the primary infected groups on 7 DPI in all the groups (Figure 7a-c). However, the viral RNA load in the lungs was reduced. Lung gRNA levels were significantly lower in the B.1-re-infected group, whereas in the case of Delta variant re-infection, only a minimal reduction was seen, and in the AY.1 group a viral RNA load was observed comparable to that of the primary infection on 7 DPI. Live virus titration was performed on the lung samples collected on 7 DPI from all the groups and no titre could be detected.
Discussion
The Delta variant and its sub-lineages have become the dominant SARS-CoV-2 lineages world-wide. The higher rate of transmissibility, increased disease severity and immune evasion potential of the Delta variant has alerted the scientific community to be vigilant about the mutating variants [2,3]. We have studied properties of the Delta AY.1 variant in comparison with the Delta and B.1 variants in Syrian hamsters and the potential for reinfection in hamsters by Delta variants 3 months post recovery from SARS-CoV-2 infection, the period for which antibodies are reported to persist in humans after infection.
Delta AY.1 variant infection in Syrian hamsters produced mild disease characterized by negligible weight loss, lower viral load in upper and lower respiratory tracts and mild pneumonic changes in the lungs. Comparable cell entry efficiency was reported by a recent study for the Delta sub-lineages with K417N mutation in comparison with the wildtype SARS-CoV-2 with the D614G mutation. This mutation tends to affect ACE2 binding affinity moderately [28]. The lower prevalence rate of the variant worldwide, i.e., less than 0.5% to date after its initial detection, points to the less efficient transmission/binding affinity of the variant [3]. Increased disease severity or risk of hospitalization has been reported for Delta variant infection compared to the earlier SARS-CoV-2 strains [29,30]. We also observed a higher degree of disease severity induced by the Delta variant in hamsters. Pathological changes observed in the lungs were similar to those reported during the acute phase of infection in humans [31,32]. Diffuse congestion and hemorrhages were observed grossly in the lungs and the histopathological changes were mostly of epithelial and vascular types. Microthrombi formation was not observed in the lungs of hamsters, unlike humans. Hamsters exhibited weight loss as the characteristic symptom of severe disease, unlike flu like symptoms and respiratory distress in humans [31].
The K417N mutation in the Delta AY.1 variant was found to be critical for neutralization resistance against some potent NAbs against SARS-CoV-2 [33,34]. Here, we observed a comparable neutralization efficiency in Delta AY.1-infected hamster sera against the Delta, B.1 and Beta variants, suggesting that the presence of the K417N mutation may not confer an advantage in terms of immune evasion, at least against these variants. Similar results of comparable neutralization efficiency of the Delta variant with the K417N mutation and the Delta variant has been reported in pseudo-virus neutralization studies [28]. Yadav et al. 2021 have reported comparable neutralization for the Delta AY.1 and Delta variants with the sera of naive BBV152 vaccinees, recovered cases with full vaccination and breakthrough cases in comparison to the B.1 variant [35]. The available vaccines against SARS-CoV-2 have shown reduced protection against symptomatic disease/infection by the Delta variant [2]. We observed cross NAbs against the variants studied here after infection and a boost in titres after re-infection. However, a significantly lower neutralization titre was observed against the Beta variant in the case of the Delta and B.1 variant-infected animal sera after primary infection. The B.1.351 variant is known worldwide for its immune escape property due to the mutations K417N, E484K and N501Y in the RBD of the spike region [33].
Many cytokines have been reported to be increased in severe COVID-19 patients and a few of them, such as IL-6, IL-8, IL-10 and TNF-α, are considered to be indicators of severe disease [36,37]. The increased production of cytokines can lead to cytokine storm and worsening of the disease prognosis [38]. IL-6, IL1-beta and TNF-α increase has been reported in hamsters infected with SARS-CoV-2 [39]. Here, we have observed increased IL-6 cytokine levels after primary infection in hamsters with SARS-CoV-2 variants. IL-6 is an important cytokine in host responses against viral infection [40]. The increase in IL-6 levels observed here could have contributed to the host response in control of infection. Other inflammatory cytokines, such as IL-4, IFN-γ and TNF-α, were also increased in comparison to control animals during the acute phase of infection. IL-6 level increase was independent of the lung histopathological score. The B.1 variant-infected animals showed the highest average lung sgRNA loads on 7 and 14 DPI and the highest increase in serum IL-6 levels. We found no aggravation in cytokine responses post re-infection. A detailed study of cell-mediated response was not possible because of the non-availability of reagents specific for hamsters.
The protective immunity conferred by prior SARS-CoV-2 infection is similar to that of vaccination [41]. Natural infection generates an effective mucosal immune response, unlike intramuscular vaccination [42]. Re-infection with SARS-CoV-2 and other human coronaviruses has been reported [6][7][8][9][10][11]. The re-infections are common 12 months after initial infection in the case of seasonal HCoVs. A trend of reduced virus replication and an increase in NAb titres were observed in hamsters following re-infection irrespective of the variants used for infection. The neutralizing titres against the Beta variant were also comparable following re-infection, contrary to the differences observed after primary infection. Polyclonal antibody response is generated by natural infection. Antibody evolution through somatic mutations and repeated antigen exposure could improve protection against newer emerging variants [43]. Thus, reinfections tend to improve neutralizing abilities against multiple variants.
Earlier research has shown that prior COVID-19 infection reduces virus replication and thus decreases transmission efficiency in Syrian hamsters re-infected 29 days after initial infection [19]. A recent re-infection study performed 15 months post primary infection in hamsters demonstrated protection against lung disease caused by the Delta variant and the prevention of transmission to naive hamsters [20]. Even though a reduction was seen in viral shedding, the nasal turbinate viral load remained comparable in the present study. We have used a high virus dose of 10 5 TCID50 for reinfection studies here which also might have contributed to this. This finding highlights the importance of maintaining COVID-19appropriate behavior till herd immunity is achieved in the population. Reinfection studies with lower virus doses and studies on the persistence of mucosal responses after infection need to be explored for better understanding of respiratory tract protection following re-infection. The hypothesis of high virus dose infection as a cause of re-infection is proved here. Furthermore, studies with lower virus doses for primary infection should be explored to understand whether the protective immune response generated by a low virus dose exposure could prevent further infection, as the majority of human SARS-CoV-2 infections are mild and asymptomatic.
Prior infection did not confer sterilizing immunity in the present study, as reported earlier in rhesus macaques and hamster models [18,43,44]. These studies were performed within a month after recovery from the primary infection, in contrast to our study. Experimentally re-infected or vaccinated animals can shed SARS-CoV-2 through the upper respiratory tract [19,[44][45][46][47]. The exact immune correlates of protection from infection are still not known. Other than the NAb levels post infection, the cellular as well as the local mucosal response tends to play an important role in protection against reinfection [48]. Limited mucosal immunity could be a probable reason for the high viral RNA load observed in the upper respiratory tract.
There are reports of varying disease severity in re-infected individuals [9,11,15,16]. Wang et al., 2021 have reported 68.8%, 18.8% and 12.5% of similar, worse and mild disease severity in re-infection cases [11]. Severe disease has also been reported with Delta variant re-infection [15,16]. We did not observe any aggravation of lung disease in the re-infected animals with any of the variants used in this study. In the case of the Delta AY.1-re-infected group, sgRNA clearance was observed in 75% of the animals by 7 DPI but the average viral RNA load remained comparable to that of primary infection and with the viral RNA load of animals of the Delta-as well as the B.1-re-infected group. The lung histopathology score severity was similar to that of the Delta variant-re-infected group in the animals. Considering all these observations, it seems that there was some amount of protection conferred, and a study with a large sample size would be appropriate to reach more definitive conclusions.
To conclude, the Delta AY.1 variant produced mild disease in a hamster model and did not show any evidence of neutralization resistance, as speculated previously. Re-infection with higher virus doses of the Delta variant or the B.1 variant 3 months after B.1 variant infection reduced virus shedding/disease severity and increased NAb levels, irrespective of the variants studied. The findings of this study indicate that primary SARS-CoV-2 infection can reduce the severity of secondary infection by the Delta variant, although it cannot confer sterilizing immunity or guarantee protection from a secondary infection.
Supplementary Materials:
The following supporting information can be downloaded at: [URL]: //www.mdpi.com/article/10.3390/v14030596/s1, Figure S1: Serum cytokine levels in hamsters post SARS CoV-2 infection; Figure S2: SARS-CoV-2 viral RNA load in organs of hamster's post infection; Figure S3: Serum cytokine levels in hamsters post SARS CoV-2 re-infection; Figure S4: Pathological changes observed in lungs after primary infection and re-infection; Figure S5: Lungs body weight ratio in hamsters after primary infection and re-infection. Data Availability Statement: All the data related to the study are available in the manuscript.
==
Domain: Biology Medicine
|
[{"added": "2021-12-02T14:09:18.019Z", "created": "2021-11-30T00:00:00.000", "id": "244802258", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/1999-4915/14/3/596/pdf?version=1647180451", "pdf_hash": "974be3a514d041267a0ccd7c0ed81c6d23c66a4b", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:37002", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "f927f9c1a47ef122e0a55755ebfc3a7a6e0884d8", "year": 2022}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Quorum Sensing and Virulence of Pseudomonas aeruginosa during Lung Infection of Cystic Fibrosis Patients
Pseudomonas aeruginosa is the predominant microorganism in chronic lung infection of cystic fibrosis patients. The chronic lung infection is preceded by intermittent colonization. When the chronic infection becomes established, it is well accepted that the isolated strains differ phenotypically from the intermittent strains. Dominating changes are the switch to mucoidity (alginate overproduction) and loss of epigenetic regulation of virulence such as the Quorum Sensing (QS). To elucidate the dynamics of P. aeruginosa QS systems during long term infection of the CF lung, we have investigated 238 isolates obtained from 152 CF patients at different stages of infection ranging from intermittent to late chronic. Isolates were characterized with regard to QS signal molecules, alginate, rhamnolipid and elastase production and mutant frequency. The genetic basis for change in QS regulation were investigated and identified by sequence analysis of lasR, rhlR, lasI and rhlI. The first QS system to be lost was the one encoded by las system 12 years (median value) after the onset of the lung infection with subsequent loss of the rhl encoded system after 17 years (median value) shown as deficiencies in production of the 3-oxo-C12-HSL and C4-HSL QS signal molecules respectively. The concomitant development of QS malfunction significantly correlated with the reduced production of rhamnolipids and elastase and with the occurrence of mutations in the regulatory genes lasR and rhlR. Accumulation of mutations in both lasR and rhlR correlated with development of hypermutability. Interestingly, a higher number of mucoid isolates were found to produce C4-HSL signal molecules and rhamnolipids compared to the non-mucoid isolates. As seen from the present data, we can conclude that P. aeruginosa and particularly the mucoid strains do not lose the QS regulation or the ability to produce rhamnolipids until the late stage of the chronic infection.
Introduction
The onset of the chronic lung infection with Pseudomonas aeruginosa in CF patients is preceded by intermittent colonization [1] usually with environmental strains [2]. The chain of events leading to the establishment of a persistent infection is mainly due to the biofilm forming capacity of P. aeruginosa with important contributions from individual virulence factors such as elastase [3], LPS [4], rhamnolipids [5] and alginate [6]. We have demonstrated that rhamnolipid plays a major role in the defense against the cellular components of the immune system, especially against the polymorphonuclear neutrophilic leukocytes (PMNs) which dominate the immune response in the CF lung [7][8][9]. P. aeruginosa respond to the presence of PMNs by upregulating synthesis of a number of virulence determinants including rhamnolipids, all of which are able to cripple and eliminate cells of the host defense which support a 'launch a shield' model by which rhamnolipids surround the biofilm bacteria and on contact eliminate incoming PMNs [9].
Production of several P. aeruginosa virulence factors is coordinated by a cell density monitoring mechanism termed Quorum Sensing (QS) [10][11][12]. P. aeruginosa employ two dominating QS system the las and the rhl encoded system. Both systems feature specific signal molecules for separation of the processes, 3-oxo-C12-HSL and C4-HSL respectively. The basic AHL QS system is comprised of an I gene encoding the AHL synthetase and a R gene encoding the receptor. During the growth of the bacteria, system specific signal molecules are produced by the synthetase, the I protein. The signal molecules produced by the bacteria bind to the receptor, the R-protein, the AHL-responsive transcriptional activator. The regulator proteins contain two functional domains. The signal molecule binding region, which is located in the N-terminal portion of the protein and a helix-turn-helix motif (HTH) located in the Cterminal, which is responsible for the protein binding to the target promoters [13][14][15]. Within these systems a third analogous receptor, the QscR operates with 3-oxo-C(12)-HSL to modulate gene expression of a specific regulon which overlaps with the two other las and rhl regulons [16]. P. aeruginosa has an additional QS regulatory pathway termed the Pseudomonas quinolone signal (PQS) system [17]. In vitro the QS systems of P. aeruginosa have been shown to be hierarchically arranged, with the las system on top, controlling the rhl system [18] and the PQS system positioned as a mediator functionally positioned between the las and rhl systems. However, it has been proposed that the rhl system can be activated independently of the las system, and it has been suggested that PQS system controls this activation [17]. This was further substantiated in a recent paper, where the authors provided evidence that rhl system is able to overcome the absence of the las system by activating specific LasR-controlled functions, including production of 3-oxo-C(12)-HSL and PQS [19].
When the chronic lung infection in CF patients is established it is well recognized that P. aeruginosa isolated from the sputum differ phenotypically from the initial intermittent strains even though they produce similar pulse field gel electrophoresis patterns and therefore are considered isogenic [20,21]. Loss of epigenetic regulatory systems such as QS is one of the dominating changes that occur during the adaptive process of the bacteria in the CF lung [22]. Different models accounting for the selection of non-functional QS systems have been reported. One such model focuses on the special nutrient availability in the CF lung which P. aeruginosa has to adapt to [22]. This model is supported by comparison of the genomes obtained from different CF isolates [23] suggesting that P. aeruginosa has the potential to act in a range of environmental conditions. Furthermore, the authors suggest that the bacterium acquires or discards genomic segments in order to optimize its genomic repertoire for the present specific environment. Another model focuses on the fact that P. aeruginosa is exposed to oxygen radicals which in turn induce genetic mutations [23,24]. We have recently demonstrated that the polymorphonuclear leukocytes (PMNs) are the major contributors of oxygen radicals in CF sputum [25]. It is therefore likely that oxygen radicals are derived from the PMNs. Recently, the cooperative behavior of mixed populations of bacteria has been studied using populations including both QS wild-type and lasR mutants [26]. These studies have introduced the concept of ''cheaters'' (the QS mutants) exploiting the functional QS systems of other members of the population [26,27]. It might be that in CF lungs, although P. aeruginosa lasRmutants may accumulate, QS-active members of the population are still maintained for the benefit of all members of the bacterial community.
Based on these observations we aimed to correlate the changes that occur in the QS systems with expression of virulence during stages of intermittent and chronic lung infections in CF patients. The capability to produce 3-oxo-C12-HSL and C4-HSL signal molecules and the sequences of lasR and rhlR encoding the receptor-transcriptional regulators as well as the lasI and rhlI encoding the synthethases were investigated in a large number of randomly collected CF isolates, (pairs of mucoid and non-mucoid if available) obtained from the intermittent or chronic stages of lung infection. The dynamics of the functionality of QS systems in the clinical strains were correlated to rhamnolipids and elastase production as well as to the mutational frequencies of the isolates. Our results show that functionality of the rhl encoded system is maintained longer than the las system during the chronic infection, especially in the mucoid isolates, providing evidence for the possible role of QS inhibitors in the treatment of early as well as late stages of P. aeruginosa infections.
QS functionality and duration of infection
Loss of QS regulation is generally considered a hallmark of chronic virulence and has been described for several P. aeruginosa CF isolates [22,28]. To investigate the dynamics of the QS loss at different stages of the P. aeruginosa lung infection, we determined the production of QS signal molecules of isolates collected from patients with intermittent colonization as well as chronic lung infection at different time points.
The P. aeruginosa CF isolates from the intermittently colonized patients showed significantly higher frequency of strains with simultaneous production of both QS molecules (x 2 test p,0.0001) and higher levels of rhamnolipid production (median The CF isolates were divided in four groups according to their ability to produce QS molecules: a group producing both 3-oxo-C12-HSL and C4-HSL (n = 58), a group producing only C4-HSL (n = 63), a group producing only 3-oxo-C12-HSL (n = 16) and a group not producing QS molecules (n = 73). A significant difference was found between the duration of the chronic lung infection of the CF patients harboring the isolates belonging to the different groups (table 1). The majority of the CF isolates (63) were not producing 3-oxo-C12-HSL after 12 years of infection while only a small proportion (16 isolates) were 3-oxo-C12-HSL producers but lost the ability to produce C4-HSL. Importantly, the lost of both QS molecules was found first after 17 years of infection. This shows that the abilities to produce 3-oxo-C12-HSL and C4-HSL signal molecules are lost at different time points during the chronic lung infection and particularly interesting is the finding of C4-HSL molecules in isolates from the late stages of the infection. This indicates that the rhl system is functional even in the late phases of the chronic lung infection and suggests that the Lasindependent regulation of rhl system is maintained during the chronic lung infection. These data emphasize that the shielding through rhamnolipid production might play an important role during the first 17 years of infection. Significant differences in the level of rhamnolipid (table 1) and elastase (table 1) were found between the four groups of QS signal molecule producers concurring that these virulence factors require a functional QS system for expression.
QS and mucoidity
Early occurrence of mucoid P. aeruginosa in the sputum of CF patients has been correlated to a poor prognosis [5,29]. Mucoidity has been shown to be selected for in the CF lung due to the protective role of alginates against oxygen radicals from activated PMNs [30]. In addition, as judged from flow-cell and animal experiments, mucoid isolates form more robust biofilms [31,32]. Investigations of QS functionality and connected phenotypes expressed by mucoid and non-mucoid isolates that were obtained from the chronically infected patients, showed that a significantly higher proportion of mucoid isolates produced C4-HSL compared to their non-mucoid counterparts (x 2 test, p = 0.02). Furthermore, this number was found to correlate with significantly higher amounts of rhamnolipids (median [ranges] = 4.5 [0-72.8] mg/ml) produced by mucoid compared with non-mucoid isolates (median [ranges] = 0[0-48]), (p = 0.02, Mann-Whitney) ( Figure 2).
Thus, mucoid isolates may be protected against the antimicrobial properties of the PMNs not only by alginate but also due to the production of rhamnolipids. This is in accordance with recent data from an animal model of chronic lung infection which showed that persistence against the host defense was maintained in mucoid but lost in nonmucoid isolates during the chronic lung infection of one CF patient [33]. This difference in the functionality of the QS system between mucoid and non-mucoid isolates strongly support that different adaptation strategies are employed by the two phenotypes [5,31].
QS and clonal distribution
The typing analysis showed that three previously identified bacterial clones entitled DK-1 (39 isolates), DK-2 (29 isolates) and NO (26 isolates) were represented among the 238 CF isolates. Clones DK-1 or ''red'' and DK-2 or ''blue'' are two dominant clones in the Copenhagen CF Center and clone NO is a clone identified among the Norwegian isolates [32][33][34]. The rest of the isolates were considered non-clonally related.
Significant differences in the functionality of the QS systems were found among the various clonal groups. While functionality loss of both the las and rhl systems was found in 75% of the DK-2 strains, this phenotype was encountered in only 46% and 33% of the NO and DK-1 isolates, respectively. These differences in the ability of the isolates to produce QS signal molecules were associated with significant differences in the ability of the isolates to produce rhamnolipids ( Figure 3). Importantly, we always saw a positive correlation between production of C4-HSL and rhamnolipids but no correlation between 3-oxo-C12-HSL and rhamnolipid production. This is in accordance with findings by us [35] and others who showed that PAO1 do not require a functional las system for expression of rhl and pqs controlled genes.
Several isolates belonging to the DK-2 clone did not harbor a lasR gene as shown by the lack of gene amplification which in turn suggests that this particular mutant of the DK-2 clone might have spread among CF patients after the apparent loss of QS signal recognition, or that a deletion hotspot exists at the particular chromosomal position. However, this second option is statistically very unlikely. Similar results were obtained with isolates belonging to the NO and DK-1 Table 1. Duration of the chronic lung infection of CF patients harboring P. aeruginosa isolates producing both C4-HSL and 3-oxo-C12-HSL, either C4-HSL or 3-oxo-C12-HSL or none of the QS molecules and distribution of the rhamnolipid and elastase levels in CF P. aeruginosa isolates producing both, one or none of the QS molecules. clones, although these isolates were found to harbor a functional rhl system. This suggests that dissemination and establishment within a community of CF patients does not require a functional LasRsystem. Alternatively, LasR proficient bacterial subpopulations might have been present in the initial infection but these subpopulations were not identified in our study. However, it is important to mention that we identified two CF patients that harbored QS-proficient DK-2 bacteria, suggesting that several evolutionary lineages develop in the CF population as recently published by Wilder [36].
QS signal molecules and sequence of the QS genes
To investigate the cause of QS loss we performed sequence analysis of the genes lasR and rhlR encoding the QS-regulators LasR and RhlR as well as of the genes lasI and rhlI encoding the signal molecule synthetases LasI and RhlI, respectively. The analysis showed that the wild-type sequences of lasI were conserved among the CF P. aeruginosa isolates. From 238 isolates, we only found a single occurrence of a loss of function mutation in lasI gene and intact lasR gene. However, the vast majority of the isolates presented point mutations in rhlI (data not shown) especially C249A leading to D83E which interestingly has also been found in isolates from CF patients attending the Oregon Health and Science University [36]. The measurements of C4-HSL were not affected by these point mutations. This indicates that these point mutations have no effect on the functionality of the gene and its encoded product. Mutations preferentially occurred in the genes encoding the regulatory proteins, in accordance with previous observations [37].
CF isolates with mutations in the regulatory genes lasR or rhlR produced significantly less 3-oxo-C12-HSL (x 2 p,0.0001) and C4-HSL (x 2 p,0.0001) respectively and lower levels of rhamnolipids compared to the isolates with wild-type genes ( Figure 4A and 4B).
The type of mutations identified in lasR and rhlR genes are presented in Figure 5 [38]. The complementation assays showed that the identified point mutations were responsible for the phenotypes (marked in yellow in figure 5). In the signalbinding domain, particular interesting are the mutations causing a Tyr56 to Cys exchange and a Thr75 to Lys exchange as both Tyr56 and Thr75 have been shown to be important for the binding of signal molecules to LasR [39]. In addition, Pro74, Ala105 and Gly113 were all amino acids that have been described as important for the multimerization and function of LasR (marked by squares in figure 5) [38]. Several of the mutations described in this study have been found by other investigators in lasR mutants of P. aeruginosa obtained under in vitro evolution experiments (encircled in figure 5) [22,27,38,40]. This Amino acid changes that have been previously shown to impair the LasR function [37,38] are marked in squares and amino acid changes that have been previously described in in vitro studies [22,39,48] the rhlR gene were identified in 24 out of 39 DK-2 isolates. The most frequently encountered mutation was a 3 bp deletion at position 519 (4 isolates) or 520 (18 isolates) with loss of Leu 173. These gene changes explained the basis for the loss of both 3-oxo-C12-HSL and C4-HSL signal molecules in the DK-2 clone.
Mutations in QS genes and mutability of the isolates
Increase in mutation frequencies leading to a weak mutator phenotype of the isolates was found to correlate with the loss of functionality of either lasR or rhlR (Figure 7) and occurred after a mean of 15 years. However, strong mutators occurred late during chronic infection (mean 19.7 years) and correlated to accumulation of mutations in both lasR and rhlR (Figure 7). These data are in accordance with previous observations showing that lasR mutants of P. aeruginosa obtained in in vitro evolution experiments were not hypermutable [27]. We propose that mutations in either of the QS regulatory genes can occur in isolates with non-or weak mutator phenotype, followed in time by the occurrence of strong mutators with increased accumulation of mutations which disables the entire QS system. This sequence of events in the CF lung might be explained by an impaired protection of the QS deficient strains against the mutagenic effects of reactive-oxygen species liberated by activated PMNs due to their decreased production of catalase and superoxide-dismutase.
Conclusion
The current knowledge regarding the chronic P. aeruginosa infection of the CF lung suggests that P. aeruginosa adjust its phenotypes to the environment of the CF lung. We have previously proposed that QS and especially the production of rhamnolipids are important for the initial stage of infection providing the bacteria with an immune shield which is protective against the antimicrobial activity of PMNs. As seen from the present data we can conclude that P. aeruginosa isolates may lose LasR dependent QS but keep the capability of RhlR dependent QS regulation enabling production of a number of important host damaging virulence factors particularly rhamnolipids. This capability is maintained till late in the chronic infection, in particular in mucoid isolates. Our results show that the mucoid isolates are protected from the antimicrobial activity of PMNs in the CF lung by both alginate which act as a ROS scavenger but also by their ability to produce rhamnolipids which provide a shield against cellular components of the innate immune response. This also suggest that a treatment with drugs interfering with QS and in particular the lower hierarchy of QS regulated virulence factors such as rhamnolipids might be useful not only in the early stages of the infection but also in the treatment of chronic infections with QS producing strains.
CF patients and bacterial strains
In the present study, 238 P. aeruginosa isolates obtained from 152 Scandinavian CF patients have been investigated. The random collection consist of 31 non-mucoid isolates from intermittently colonized patients, 35 non-mucoid isolates from chronically infected patients and 86 pairs of mucoid and nonmucoid isolates (172 isolates) from chronically infected patients. The 152 patients were distributed in five CF centers in Scandinavia as follows: Copenhagen, 22 intermittently and 62 chronically infected patients; Aarhus, 13 chronically infected patients; Lund, 9 intermittently and 16 chronically infected patients; Uppsala, 11 chronically infected patients and Oslo, 19 chronically infected patients. The mean duration of the lung infection in chronically infected patients was 15 years (from 1 and up to 32 years).
The CF patients were considered chronically infected when P. aeruginosa was cultured in the sputum for six consecutive months or serum precipitating antibodies to P. aeruginosa $2 by crossedimmuno-electrophoresis [41].
Sputum samples obtained by expectoration or endolaryngeal suction were Gram-stained and examined under the microscope to confirm the origin from the lower airways with the exception of the samples from Norway. The sputum samples of the 152 Scandinavian CF patients were plated on Blue agar plates (a modified Conrad Drigalski medium selective for Gram-negative rods, Statens Serum Institute, Copenhagen, Denmark containing peptone 10 g, yeast extract 5 g, NaCl 5 g, agar 11 g, detergent 0,05 g, Sodium thiosulphate 1 g, bromthymolblue 0,1 g, lactose 9 g and glucose 0,4 g).
When mucoid and non-mucoid P. aeruginosa isolates were simultaneously isolated from sputum samples, both phenotypes were collected from the Blue agar plate (Gram negative selective growth media).
Genotyping by pulsed-field gel electroforesis (PFGE)
All isolates were typed by PFGE as described previously using SpeI enzyme [42,43]. After PFGE, the band patterns were visualized by ethidium bromide staining and then photographed (GelDoc TM imaging system, Bio-Rad, Munich). The patterns were analyzed by Fingerprinting TM II software, Bio-Rad, CA, USA). The clonal relatedness of the individual pairs of mucoid and nonmucoid P. aeruginosa was confirmed according to Tenover [44]. Isolates with PFGE patterns that differ from each other by two to three bands were considered clonally related, as this pattern is consistent with a single genetic event, i.e. a point mutation or an insertion or deletion of the DNA. Isolates with PFGE patterns that differed by more than three bands were considered to belong to different strains.
Measurement of mutant frequencies in cultures of P. aeruginosa isolates
Mutation frequencies can be determined via fluctuation analysis [45]. However, because fluctuation analyses are laborious and the present study comprised 238 P. aeruginosa isolates, we took advantage of the fact that mutation frequencies and mutant frequencies are proportional in sufficiently large cultures. To determine mutant frequencies each bacterial isolate was grown overnight in 100 ml Luria-Bertani (LB) medium, upon which 20 ml culture was centrifuged at 3,000 rpm for 10 min, and resuspended in 1 ml LB medium. A 100 ml volume of undiluted, 10 21 diluted and 10 22 diluted was plated on LB plates containing 300 mg/ml rifampicin and on LB plates containing 500 mg/ml streptomycin. A 100-ml volume of 10 27 to 10 210 dilutions was plated on LB plates. After incubation at 37uC for 48 h the numbers of CFU were counted, and the frequencies of rifampicin resistant and streptomycin resistant mutants were calculated.
DNA sequence analysis of lasR, rhlR, lasI and rhlI genes lasR, rhlR, lasI and rhlI genes from all isolates were PCR amplified using the primer sets described below. After purification (Promega Wisart purification kit, Madison,USA) the PCR products were sequenced on a Macrogen automatic DNA sequencer ABI3700. The number of reads was between two and four for each gene of each strain. The sequencing results were compared with the strain PAO1 sequence (www.pseudomonas. com) with DNASIS Max vesion 2.0 (Hitachi software Engineering), in order to determine the occurrence of sequence variations.
Complementation of the lasR mutations in the clinical P. aeruginosa isolates was done by electroporation of plasmid MH645 (Plac-lasR cloned in BamHI site of pBBR1-MCS5 [48]). The success of complementation was verified by the reestablishment of the protease activity.
For PCR amplification and sequencing the following primers were used. Determination of elastase activity, rhamnolipid and signal molecule production Bacteria from 280uC freeze stocks were plated onto blue agar plates (State Serum Institute, Denmark) and incubated at 37uC overnight. From each plate (representing an isolate) one colony was selected and grown as a overnight culture in either ox-broth (for elastase activity) or AB trace minimal medium containing 3 mM glucose (32) at 37uC with shaking (rhamnolipid and signal molecules production). The supernatant was either dialyzed against sterile water (for elastase activity) or sterile filtered using 0.2 mm pore filters (16543; Sartorius), frozen and kept for further analysis.
Elastase activity. Elastase activity was determined in a spectrophotometric assay using elastin-Congo red (Sigma) as a substrate, as previously described [49].
Rhamnolipid production. Liquid chromatography/electron spray ionization mass spectrometry (LC-ESI-MS) data on pure rhamnolipid was used to produce a standard curve for rhamnolipid B (concentration vs. total ionization current (TIC)). The rhamnolipid standards used for calculating the concentration curve were analyzed immediately prior to, as well as after analysis of the samples, in order to minimize potential differences in ionization levels of rhamnolipid between the samples. Rhamnolipid concentrations were normalized to the standard curve for rhamnolipid B. In the analysis the total rhamnolipid concentration was derived from the six major rhamnolipids, with the following masses [M+NH4]+: 668,4; 694,4; 696,4; 522,4; 548,4 and 550.4. These equate to C10-C10-rha-rha, an unidentified C10-C12D-rha-rha, C10-C12-rha-rha, and the respective mono-rhamnose derivatives. BHPLC-MS analysis was performed with an agilent 1100 series high performance liquid chromatography (HPLC) connected to a micromass LCT TOF MS.
Measurement of signal molecules. The measurements for QS signal molecules were performed in ''black 96 welled microtiter plates'' (Nunc, black PolyBase, USA), using specific QS reporter strains. The reporter strains used were previously described: MH205 (C4-HSL) [50] and MH155 (3oxo-C12-HSL) [51]. ABT minimal media supplemented with thiamin (25 mg/ml), 0.5% glucose and 0.5% casaminoacids was used to grow the reporter strains. For controls 10 mM N-butanoyl-L-homoserine lactone (BHL), 10 mM N-Dodecanoyl-DL-homoserine lactone (DDHL) and supernatant from a wild type P. aeruginosa was used. Overnight cultures of each strain were prepared in 5 ml LB and 5 ml ABT supplemented with 0.5% glucose and 0.5% casaminoacids, respectively. For QS signal molecule measurements, 2 fold dilutions of the sterile filtered cultures were performed with a final volume of 150 ml. To each well 150 ml of the appropriate monitor strain diluted to OD 450 0.1, was added.
The plate was then incubated and read in a Multi Label reader (wallac 1320Viktor, Perkin Elmer). Measurements of turbidity at OD 450 and fluorescence (excitation 485 nm and emission 535 nm) were done every 15 minutes for 17 hours. The temperature was kept at 37uC.
Ethics
The Danish, Norwegian and Swedish Research Ethics Committees approved the collection of bacteria and informed written consent was obtained from all patients.
Statistical analysis
The description and analysis of the data were carried out using StatViewH 5.0.1. software.
The distribution of the data did not follow in all groups a Normal distribution and therefore the data are presented as median [ranges] and the graphic representation was done by boxplots indicating the median, 10, 25, 75 and 90 centiles of the groups. The dots outside the whiskers represent single isolates with values smaller than in 10% or higher than in 90% of the isolates.
The nonparametric Mann-Whitney test was used for comparison between the different groups. Categorical data were analyzed in frequency tables that fulfilled the guidelines for '' a large sample'' approximation and x 2 test was used to test the null hypothesis (e.g. QS producers in mucoid vs nonmucoid isolates). The level of significance was 5%.
==
Domain: Biology Medicine
|
[{"added": "2014-10-01T00:00:00.000Z", "created": "2010-04-12T00:00:00.000", "id": "14211202", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0010115&type=printable", "pdf_hash": "47e5fdd49526749771000b6c97e6c427625e9608", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39656", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "47e5fdd49526749771000b6c97e6c427625e9608", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Impaired macrophage phagocytosis of bacteria in severe asthma
Background Bacteria are frequently cultured from sputum samples of severe asthma patients suggesting a defect in bacterial clearance from the airway. We measured the capacity of macrophages from patients with asthma to phagocytose bacteria. Methods Phagocytosis of fluorescently-labelled polystyrene beads, Haemophilus influenzae or Staphylococcus aureus by broncholaveolar lavage alveolar macrophages (AM) and by monocyte-derived macrophages (MDM) from non-asthmatics, mild-moderate and severe asthmatic patients was assessed using fluorimetry. Results There were no differences in phagocytosis of polystyrene beads by AMs or MDMs from any of the subject groups. There was reduced phagocytosis of Haemophilus influenzae and Staphylococcus aureus in MDMs from patients with severe asthma compared to non-severe asthma (p < 0.05 and p < 0.01, respectively) and healthy subjects (p < 0.01and p < 0.001, respectively). Phagocytosis of Haemophilus influenzae and Staphylococcus aureus by AM was also reduced in severe asthma compared to normal subjects (p < 0.05). Dexamethasone and formoterol did not suppress phagocytosis of bacteria by MDMs from any of the groups. Conclusions Persistence of bacteria in the lower airways may result partly from a reduced phagocytic capacity of macrophages for bacteria. This may contribute to increased exacerbations, airway colonization and persistence of inflammation.
Introduction
Patients with asthma are usually well-controlled with inhaled corticosteroids (CS) and long-acting β 2 -agonists, but a proportion of patients, described as severe asthmatics, continue to experience uncontrolled asthma in spite of these treatments [1]. These patients consume a significant proportion of medical resources in terms of pharmacological treatments, hospital admissions or use of emergency services, and time off work or school [2]. Respiratory infections are often associated with wheezing episodes and may also impact on the development and severity of asthma. Viral and bacterial infections or colonization with bacteria could lead to chronic lower airway inflammation, impaired mucociliary clearance, increased mucus production and worsening of asthma [3,4]. Both viral and bacterial infections have been recognized as drivers of asthma exacerbations [5]. Various bacterial species have been cultured from sputum samples of patients during exacerbations [6], and also during a stable period in moderate-to-severe asthma patients [7,8]. Gram-positive Staphylococcus aureus and Gram-negative Haemophilus influenzae were two of the most common bacterial species cultured from induced sputum of patients with severe asthma [8]. The more sensitive technique of 16S ribosomal RNA microarray to detect bacterial species in the lower airways has revealed an increase in bacterial burden and diversity in patients with mild-to-moderate asthma compared to nonasthmatic individuals [9,10]. Thus, there appears to be an increased propensity for asthmatic subjects to carry more bacterial pathogens in their lower airways.
Macrophages produce a variety of cytokines and mediators that are vital for immune and inflammatory responses in their response with external agents. Macrophages are also important for the removal of particulates and bacteria from the airways [11]. This removal of potentially pathogenic micro-organisms via phagocytosis is essential for maintaining a non-pathogenic environment within the lung. However, the phagocytic capacity of macrophages for bacteria may be impaired in various pulmonary conditions. Thus, lung macrophages from children with severe asthma showed reduced phagocytic responses to Staphylococcus aureus [12]. Similarly, defective efferocytosis by AMs from severe asthmatics but not from mildmoderate asthmatics has also been reported [13]. This defect has been postulated to underlie the chronic inflammatory state due to the accumulation of bacteria and apoptotic and necrotic cells in the airways. In chronic obstructive pulmonary disease (COPD), a number of studies have shown defects in both bacterial clearance [14,15] and efferocytosis [16] indicating impaired macrophage innate responses.
We hypothesize that alveolar macrophages (AMs) and macrophages derived from blood monocytes (MDMs) from subjects with asthma may show impaired phagocytosis of Haemophilus influenzae and Staphlycoccus aureus when compared to non-asthmatic subjects. In addition, we examined the potential effects of the asthma treatments, corticosteroids and β-adrenergic agonists, on these phagocytic responses in-vitro.
Study participants
Patients with asthma were recruited from the Asthma Clinic of the Royal Brompton Hospital, London. All patients demonstrated either an improvement in FEV 1 of >12% baseline FEV 1 after inhaling 400 μg salbutamol aerosol or bronchial hyper-responsiveness defined by a concentration of methacholine provoking a fall in FEV 1 of 20% or more (PC 20 ) of <8 mg/ml. Current and exsmokers of >5 pack-years were excluded. Patients on steroid-sparing agents such as methotrexate, those with a concurrent diagnosis of clinical bronchiectasis or with a respiratory tract infection requiring antibiotic treatment within 6 weeks of enrolment were excluded.
Severe asthmatics were defined according to the American Thoracic Society's major criteria of needing either continuous or near-continuous oral corticosteroids or high dose inhaled corticosteroids (2,000 μg beclomethasone-equivalent per day or more) or both in order to achieve a level of mildmoderate persistent asthma, and also by the presence of 2 or more minor criteria of asthma control [17]. Patients with non-severe asthma were those who did not fall into the severe asthma category and who used up to 1000 μg inhaled beclomethasone or equivalent dosage per day with well-controlled asthma. Healthy volunteers with no diagnosis of asthma and with a negative PC 20 (>16 mg/ml) were also recruited. The study protocol was approved by the Ethics Committee of Royal Brompton & Harefield NHS Trust/National Heart & Lung Institute, London UK. All volunteers gave their written informed consent.
Fibreoptic bronchoscopy and alveolar macrophage isolation
Fibreoptic bronchoscopy was performed using topical anaesthesia with lignocaine to the upper and lower airways under conscious sedation with intravenous midazolam. Warmed 0.9% (w/v) NaCl solution (50 ml × 4) was instilled into the right middle lobe and bronchoalveolar lavage (BAL) fluid was recovered by gentle hand suction. BAL fluid was centrifuged (400 g for 10 min) and the resultant cell pellet washed with Hanks' balanced salt solution (HBSS). Cells were then suspended in culture media (RPMI-1640, containing 10% (v/v) foetal calf serum (FCS), 100U/ml penicillin, 100 μg/ml streptomycin, 2.5 μg/ml amphotericin, 2 mM L-glutamine). Cytospins were prepared and stained with Diff Quick stain (Harleco, Gibbstown, NJ, USA) for differential cell counts. Alveolar macrophages (AMs) were purified by adhesion to the plastic well for 4 h and then cultured overnight prior to experimentation.
Phagocytosis assay
Phagocytosis was measured as described previously [15]. AMs and MDMs were exposed, for 4 h, to fluorescentlylabelled polystyrene beads (2 μM diameter) at a concentration of 50 × 10 6 beads/ml or to heat-killed non-typeable H. influenzae or to S. aureus, all labelled with Alexa-Fluor 488 conjugate (Invitrogen, Paisley, Scotland). AMs and MDMs were washed with Dulbecco-phosphate buffered saline and extracellular fluorescence was quenched with 1% (w/v) trypan blue at room-temperature. Trypan blue was aspirated and phagocytosis of fluorescent-labeled beads and bacteria by AMs or MDMs was measured using a Fluostar Optima fluorimeter (BMG LabTech, Aylesbury, Buckinghamshire) at excitation wavelength of 480 nm and emission wavelength of 520 nm. Data were expressed as relative fluorescent units (RFU). In order to determine that the beads or bacteria were being internalised, cells were exposed to cytochalasin D (5 μg/ml) for 30 min, prior to incubation with bacteria.
Confocal microscopy was used to visualize whether the beads or bacteria were internalised. MDM (2×10 5 cells) were cultured on well Lab-tek Permanox chamber slides and phagocytosis performed as described above. Cells were then incubated at 37°C in 5% (v/v) CO 2 with Cell-Tracker Red CMPTX dye (12.5 μM; Invitrogen, Paisley, Scotland) for 45 min to stain the cytoplasm. Cells were then fixed with 4% (w/v) paraformaldehyde. The nuclei were stained by incubation with 4'6-diamidino-2-phenylindole dihydrochloride (DAPI; 250 μM) for 3 min. Cells were viewed on a confocal microscope with Krypton-Argon laser fluorescence detector.
Data analysis
Results are expressed as mean ± SEM. A Kruskal-Wallis test or Dunnett's test was used for multi-group comparisons as appropriate. Differences between the effects of dexamethasone or formoterol or dexamethasone and formoterol treatment were analysed using Wilcoxon paired t-test. A Mann-Whitney test was also used where appropriate. Correlations were determined using Spearman rank correlation coefficient. A p value of <0.05 was considered significant.
Participant characteristics
Severe asthmatics had lower FEV 1 (% predicted) and FVC (% predicted) compared to both non-severe asthmatics and normal subjects (Table 1). They were also on higher doses of inhaled corticosteroids and reported a greater frequency of exacerbations. Of 14 patients with severe asthma recruited to the MDM study (Table 1), 10 were on daily oral prednisolone. In the severe asthmatics recruited to the AM study (Table 2), there was a trend towards higher eosinophil and neutrophils counts in BAL compared to both healthy and non-severe asthmatic subjects. Three subjects had studies performed on AMs and MDMs.
Phagocytosis by MDMs
Having established the assay conditions, MDMs from normal subjects and patients with non-severe and severe asthma were exposed to fluorescently-labeled beads or bacteria. There was no difference in phagocytosis of beads between the groups ( Figure 1A). However, there was impaired phagocytosis of fluorescently-labelled H. influenzae by MDMs from severe asthmatics compared to cells from non-severe asthmatic patients (p < 0.05) and normal subjects (p < 0.01; Figure 1B). MDM phagocytosis of fluorescently-labelled S. aureus was also impaired in severe asthma compared to non-severe asthma (p < 0.01) and normal subjects (p < 0.001). There was no difference in the ability of MDMs from non-severe asthmatics to phagocytose either bacterial species compared to normal subjects.
Phagocytosis by alveolar macrophages
AMs from all subject groups had a similar capacity to phagocytose beads ( Figure 2A). AMs from severe asthmatic patients exhibited a reduced capacity to phagocytose bacteria similar to that observed in MDMs ( Figure 2B, 2C). There was impaired phagocytosis of fluorescently-labelled H. influenzae in severe asthma compared to normal subjects ( Figure 2B; p < 0.05 (Kruskal-Wallis, post-hoc test)); for S. aureus, this impairment did not reach significance. However, there was a significant difference between the phagocytic response to S. aureus when healthy subjects were directly compared to severe asthmatics (p < 0.01; Mann-Whitney).
Internalisation of S. aureus by MDM
Confocal microscopy was used to confirm that particles and bacteria were being internalized ( Figure 4) and showed uptake of both beads and S. aureus within MDMs ( Figure 4A and B). In order to further confirm the process of internalization, MDMs were exposed to cytochalasin D (5 μg/ml), an inhibitor of actin filament polymerization, prior to incubation with S. aureus ( Figure 4C). This led to an inhibition of phagocytosis with S. aureus unable to enter the cells, localizing to the outer cell membrane of MDMs. Having established that cytochalasin D prevented phagocytosis, we determined whether the fluorimetric plate reader assay could distinguish between intra-cellular bacteria (phagocytosis) and outer membrane-associated bacteria. Phagocytosis experiments were then performed in the absence and presence of cytochalasin D. Phagocytosis of beads, H. influenzae and S. aureus was suppressed in MDMs from all patient groups as indicated by the difference in fluorescence units ( Figure 4D-F). There was a significant difference for all groups and particles between non-stimulated and Cyto D treated cells ( Figure 4D-F).
Effect of dexamethasone on phagocytosis
In order to investigate whether corticosteroids could influence the phagocytic response, MDMs isolated from normal subjects, and non-severe and asthmatics were exposed to dexamethasone prior to addition of beads or bacteria. Dexamethasone improved phagocytosis of beads and bacteria by MDM from normal subjects and nonsevere asthmatics to a small extent (~10%) at low concentrations, reaching significance at 10 -8 M, and also at 10 -7 M for beads only, but this effect was not seen at higher concentrations of dexamethasone ( Figure 5). Dexamethasone did not change the phagocytic response of MDMs obtained from patients with severe asthma. Dexamethasone did not affect cell metabolic activity as measured by MTT assay (Additional file 1: Figure S1A).
Effect of formoterol alone or in combination with dexamethasone
The effect of the long-acting inhaled β 2-agonist, formoterol, and the combination of dexamethasone and formoterol on phagocytosis of MDMs was investigated. Formoterol alone had no effect on phagocytosis of bacteria by MDMs in any of the subject groups ( Figure 6A-B). The combination of formoterol with dexamethasone at 10 -8 M each caused a small improvement of phagocytosis of H. influenzae by MDMs from normal subjects and non-severe asthmatics (p < 0.05, Figure 6C), but this was not seen in MDMs from severe asthmatic subjects ( Figure 6C and 6D). For phagocytosis of S. aureus by MDMs, there was also an improvement with dexamethasone and formoterol at 10 -8 M for normal individuals (p < 0.01, Figure 6D) and for non-severe asthmatics with dexamethasone and formoterol at 10 -8 and 10 -9 M (p < 0.05, Figure 6D). Formoterol did not affect cell metabolic activity as measured by MTT assay (Additional file 1: Figure S1B).
Discussion
We investigated whether macrophage clearance of bacterial species associated with asthmatic airways were altered in asthma. We showed that there was a reduction in phagocytosis of fluorescently-labelled H. influenzae and S. aureus by MDMs from patients with severe asthma compared with MDMs from healthy controls; however Figure 3 Relationship between phagocytosis and lung function parameters. MDM were derived from normal subjects (•, n = 14), non-severe (■, n = 14) and severe asthmatics (▲, n = 14) and exposed to either fluorescently-labeled H. influenzae (Panel A) or S. aureus (Panel B). Alveolar macrophages from normal subjects (•, n = 7) and patients with non-severe asthma (NSA; ■, n = 6) or severe asthma (SA; ▲, n = 8) were exposed to either fluorescently-labeled H. influenzae (Panel C) or S. aureus (Panel D). Correlations were determined using Spearman rank correlation coefficient.
this defect was not apparent in MDMs from non-severe asthmatics. We also showed that alveolar macrophages from severe asthma patients exhibited compromised phagocytic capacities for these bacterial species. However, the phagocytic response of these macrophages and MDMs to polystyrene beads was not altered in cells from patients with asthma, indicating that the defect was related specifically to these bacteria. These results are an extension of previous observations in children, where phagocytosis of inactivated S. aureus was impaired in alveolar macrophages from patients with severe asthma and to a lesser extent in those with moderate asthma [12]. This observation may underlie the increased risk of bacterial infections that has been described in patients with stable severe asthma [8,20]. In addition, this defect in bacterial phagocytosis may explain partly the presence of the altered microbiome that has been described in the airways of patients with asthma compared with healthy subjects [9]. A similar defect in the phagocytosis of bacterial pathogens by MDMs and AMs from patients with COPD has also been reported [15], also in association with an altered microbiome in the airways of patients with COPD [9,21]. MDMs and macrophages obtained from non-severe asthmatic patients did not show reduced phagocytosis of H. influenzae and S. aureus compared to normal subjects suggesting that the phagocytic defect was associated with more severe asthma. Severe asthma exhibits a number of similarities with COPD such as the presence of airway neutrophilia [22]. Our results are in line with a recent study showing that macrophages isolated from induced sputum from asthmatic subjects with increased neutrophil counts in the sputum were also less able to phagocytose apoptotic bronchial epithelial cells [23], indicating a defect in the process of efferocytosis in the airways of patients with asthma. The defect in phagocytosis of bacteria could lead to a persistence of airway bacteria resulting in airway neutrophilia through the release of neutrophilic chemokines such as CXCL8 and possibly to worsening of asthma in terms of severity of symptoms and recurrence of exacerbations. Recent work has indicated that in-vitro infection of lung tissue macrophages with rhinovirus resulted in an impairment of the phagocytosis of bacteria by alveolar macrophages [24]. This mechanism of defective bacterial phagocytosis by the presence of rhinovirus may be relevant to acute severe asthma, where a high percentage of positive bacterial cultures from sputum samples has also been reported [25]. The cellular and molecular mechanisms for this phagocytic defect of macrophages from patients with severe asthma are unclear. Macrophages are important for both innate and acquired immunity in the respiratory tract, and have a pivotal role in lung defense against viruses, bacteria and fungi [26]. The findings of this study indicate that, in severe asthma, an impairment of the ability of alveolar macrophages to phagocytose bacteria could lead to prolonged persistence of bacteria in the airways and lungs, which could contribute to the worsening of asthma control. In a previous study, we showed that the release of pro-inflammatory cytokines from peripheral blood mononuclear cells and alveolar macrophages from patients with asthma, stimulated with lipopolysacharide, was not different from that of patients with non-severe asthma [27,28], indicating that the inflammatory response to one constituent of gram-negative bacteria is not impaired in severe asthma. However, the response of these cells to whole bacteria is not known. Reactive oxygen species production by alveolar macrophages is required for bacterial killing [29], and is also important for phagocytosis. In mice lacking the antioxidant enzyme extracellular superoxide dismutase [30], macrophage phagocytosis of bacteria was impaired, suggesting that removal of reactive oxygen species within alveolar macrophages is required for normal phagocytosis.
There was a positive correlation between airflow obstruction as measured by FEV 1 and phagocytosis of bacteria, either S. aureus or H. influenzae, by alveolar macrophages and MDMs. The link between the reduced bacterial phagocytosis and airflow obstruction remains unclear. Airflow obstruction in severe asthma has been associated with eosinophilic inflammation, duration of asthma and thickening of the airways as measured by high resolution computed tomography [8]. It is possible that the and severe asthmatics (▲, n = 14) were pre-treated with dexamethasone for 1 h prior to exposure to fluorescently-labelled beads (Panel A), H. influenzae (Panel B) or S.aureus (Panel C) for 4 h. Phagocytosis was measured by fluorimetry and presented as mean ± SEM. A Mann-Whitney test was used to compare the effect of given concentration of dexamethasone with that of untreated samples. **p < 0.01 and ***p < 0.001 compared to untreated cells.
link may lie with airway wall remodelling, as this has been associated with the presence of chronic airflow obstruction. The negative correlation between the percentage of eosinophils in bronchoalveolar lavage fluid and the phagocytic response to S. aureus of macrophages from asthma patients also support a potential role for eosinophilic inflammation. Eosinophlic inflammation is an important source of oxidative stress in asthma [31], and oxidative stress has been linked to defective phagocytosis of bacteria [32].
Inhaled corticosteroids and β-agonists remain the mainstay of treatment of many patients with asthma [33]. Patients with more severe asthma are usually treated with high doses of these agents. In addition, the majority of the patients with severe asthma in our study were also on oral corticosteroid therapy. However, many patients with severe asthma may not achieve control and this lack of therapeutic response to corticosteroids has been attributed to the development of corticosteroid insensitivity [27,28]. Oxidative stress pathways [34] and activation of mitogenactivated protein kinases have been implicated as mechanisms of corticosteroid insensitivity in severe asthma [27,35]. Bacterial interactions with epithelial cells and macrophages can activate these pathways and therefore could underlie corticosteroid insensitivity.
Neither dexamethasone nor formoterol had any deleterious effects on the phagocytic response in MDM from any of the subject groups. Dexamethasone alone or combined with formoterol slightly improved the phagocytic response of MDM for both beads and bacteria obtained from healthy subjects but not from those with asthma. The significance of this small increase is unlikely to be clinically relevant. There was also no effect of the phagocytic response towards either H. influenzae or S. aureus in severe asthma by dexamethasone or formoterol, which contrasts to the increase in phagocytosis of bacteria by MDMs stimulated by budesonide previously reported in COPD patients [15]. However, in another study, budesonide decreased phagocytosis in alveolar macrophages from healthy never-smoked subjects by decreasing the number of cells with the ability to phagocytose as well as decreasing phagocytic capacity of these cells [36]. These shortterm studies will not categorically exclude the possibility that long-term usage of corticosteroids and long-acting β-agonists in severe asthma may have a detrimental effect on macrophage phagocytosis. Figure 6 Effect of formoterol alone or in combination with dexamethasone on phagocytosis. MDM were derived from from normal subjects (•, n = 5), non-severe asthmatics (■, n = 6) and severe asthmatics (▲, n = 5) were pre-treated with formoterol in the absence or presence of dexamethasone for 1 h prior to exposure to H. influenzae (Panels A and C) or S. aureus (Panels B and D) for 4 h. Phagocytosis was measured by fluorimetry and data presented as mean ± SEM. Differences between the effects of dexamethasone or formoterol or dexamethasone and formoterol treatment were analysed using Wilcoxon paired t-test, where *p < 0.05 and **p < 0.01 compared to untreated cells. D = dexamethasone, F = formoterol.
A possible shortcoming of the study could be the lower number of subjects recruited to the AM study as a result of the inherent problems associated with collecting BAL samples from severe asthmatics. However, these studies performed in AM were proof of principle studies to confirm, ex vivo, the findings of the MDM study. Moreover, although there are studies showing that macrolides [37] and antifungal agents [38] can improve the phagocytic capabilities of macrophages, our results show that in the presence of antifungal and/or antibacterial agents, AMs and MDMs from patients with asthma continue to display an inherent impaired phagocytic response compared to those from control subjects In summary, MDMs and AMs from severe asthma patients demonstrate reduced phagocytosis of fluorescentlylabelled H. influenzae and S. aureus. This may contribute to the bacterial colonisation of the lower airways and to the propensity for bacterial exacerbations in severe asthma. This defect is unlikely to result from an acute response to corticosteroid or β-adrenergic therapy. Further work is needed to determine the cellular and molecular basis of this phagocytic defect for bacteria.
Conclusions
Persistence of bacteria in the lower airways may result partly from a reduced phagocytic capacity of macrophages for bacteria. This may contribute to increased exacerbations, airway colonization and persistence of inflammation.
Additional file
Additional file 1: Figure S1. Effect of dexamethasone or formoterol on metabolic activity of MDM. MDM from normal subjects, non-severe asthmatics and severe asthmatics were treated with dexamethasone (A) or formoterol (B) for 1h and cells assayed for metabolic activity using MTT assay. Figure S2. Relationship between phagocytosis of S. aureus by AM and % eosinophils in BAL. Alveolar macrophages from normal subjects (•, n = 7) and patients with non-severe asthma (■, n = 6) or severe asthma (▲, n = 8) were exposed to fluorescently-labelled S. aureus. Correlations were determined using Spearman's rank correlation coefficient. Competing interests PKB has received project grant funding from GlaxoSmithKline. PJB has received project grant funding from GlaxoSmithKline and Astra-Zeneca. KFC has received project grant funding from GlaxoSmithKline and Pfizer. LED has received project grant funding Astra-Zeneca and Pfizer. The other authors have no competing interests.
Authors' contributions ZL participated in the design of the study, carried out all assays on MDMs, performed the statistical analysis and drafted the manuscript. QZ participated in the design of the study, and performed assays on AMs. CMRT participated in the design of the study and generated labelled Haemophilus influenzae. KR performed assays on AMs. DG performed bronchoalveolar lavage on subjects for the AM study. PJB participated in the design and coordination of the study. LED participated in the design and coordination of the study and critically revised the manuscript. PKB and KFC conceived the study, participated in the design and coordination of the study and wrote the manuscript. All authors read and approved the final manuscript.
==
Domain: Biology Medicine
|
[{"added": "2016-05-12T22:15:10.714Z", "created": "2014-06-27T00:00:00.000", "id": "642231", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://respiratory-research.biomedcentral.com/track/pdf/10.1186/1465-9921-15-72", "pdf_hash": "63ee63129265c70e09b306fa99aae467ae19c999", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:32907", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "994043e0f6f07059e9159a45babaebef4eb20ae2", "year": 2014}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Tumor and Stem Cell Biology C 1 GALT 1 Enhances Proliferation of Hepatocellular Carcinoma Cells via Modulating MET Glycosylation and Dimerization
Altered glycosylation is a hallmark of cancer. The core 1 b1,3-galactosyltransferase (C1GALT1) controls the formation of mucin-type O-glycans, far overlooked and underestimated in cancer. Here, we report that C1GALT1 mRNA and protein are frequently overexpressed in hepatocellular carcinoma tumors compared with nontumor liver tissues, where it correlates with advanced tumor stage, metastasis, and poor survival. Enforced expression of C1GALT1 was sufficient to enhance cell proliferation, whereas RNA interference– mediated silencing of C1GALT1 was sufficient to suppress cell proliferation in vitro and in vivo. Notably, C1GALT1 attenuation also suppressed hepatocyte growth factor (HGF)–mediated phosphorylation of the MET kinase in hepatocellular carcinoma cells, whereas enforced expression of C1GALT1 enhanced MET phosphorylation. MET blockade with PHA665752 inhibited C1GALT1-enhanced cell viability. In support of these results, we found that the expression level of phospho-MET and C1GALT1 were associated in primary hepatocellular carcinoma tissues. Mechanistic investigations showed that MET was decorated with O-glycans, as revealed by binding to Vicia villosa agglutinin and peanut agglutinin. Moreover, C1GALT1 modified the O-glycosylation of MET, enhancing its HGF-induced dimerization and activation. Together, our results indicate that C1GALT1 overexpression in hepatocellular carcinoma activates HGF signaling via modulation of MET O-glycosylation and dimerization, providing new insights into how O-glycosylation drives hepatocellular carcinoma pathogenesis. Cancer Res; 73(17); 1–11. 2013 AACR. Introduction Hepatocellular carcinoma is the fifth most common solid tumor and the third leading cause of cancer-related deaths worldwide (1). Because of late-stage diagnosis and limited therapeutic options, the prognosis of patients with hepatocellular carcinoma after medical treatments remains disappointing (2). Diverse posttranslational modifications control various properties of proteins and correlate with many diseases, including cancer (3). Although comprehensive genomic and proteomic analyses have identified many key drivers of hepatocellular carcinoma, the posttranslational modifications remain poorly understood (4, 5). Thus, elucidation of the precise molecular mechanisms underlying hepatocellular carcinoma progression is of great importance for developing new reagents to treat this aggressive disease. Glycosylation is the most common posttranslational modification of proteins, and aberrant glycosylation is often observed in cancers (6, 7). Accumulated evidence indicates that alterations in N-linked glycosylation are a hallmark of various liver diseases, including hepatocellular carcinoma (5, 8). For instance, expression of N-acetylglucosaminyltransferase-III and -V is increased in hepatocellular carcinoma (9, 10). An N-glycan profiling study identified novel N-glycan structures in serum as prognostic markers of hepatocellular carcinoma (11). In addition, a-1,6-fucosyltransferase can generate fucosylated a-fetoprotein (AFP), which provided a more accurate diagnosis of hepatocellular carcinoma from chronic liver diseases (12, 13). However, changes in O-linked glycosylation have been overlooked in the past. The O-glycosylation of proteins is difficult to explore, as consensus amino acid sequences of O-glycosylation remain unclear and effective releasing enzymes for O-glycans are not available (14). Recently, a systematic analysis of mucin-type O-linked glycosylation revealed that mucin type O-glycans are decorated not only on mucins but on various unexpected proteins, and functions of the O-glycosylation are largely unknown (15). Several lines of evidence indicate that O-glycosylation of proteins plays critical Authors' Affiliations: Department of Surgery, National Taiwan University Hospital; Graduate Institute of Anatomy andCell Biology, National Taiwan University College of Medicine; and Research Center for Developmental Biology and Regenerative Medicine, National Taiwan University, Taipei, Taiwan Note: Supplementary data for this article are available at Cancer Research Online ( [URL]/). Y.-M. Wu and C.-H. Liu contributed equally to this work. Corresponding Authors: Min-Chuan Huang, Graduate Institute of Anatomy and Cell Biology, National Taiwan University College of Medicine, No. 1, Sec. 1 Jen-Ai Road, Taipei 100, Taiwan. Phone: 886-2-23123456, ext. 88177; Fax: 886-2-23915292; E-mailand ReyHeng Hu, E-maildoi: 10.1158/0008-5472. CAN-13-0869 2013 American Association for Cancer Research. Cancer Research www.aacrjournals.org OF1 Research. on July 21, 2017. © 2013 American Association for Cancer cancerres.aacrjournals.org Downloaded from Published OnlineFirst July 5, 2013; DOI: 10.1158/0008-5472. CAN-13-0869
Introduction
Hepatocellular carcinoma is the fifth most common solid tumor and the third leading cause of cancer-related deaths worldwide (1). Because of late-stage diagnosis and limited therapeutic options, the prognosis of patients with hepatocellular carcinoma after medical treatments remains disappointing (2). Diverse posttranslational modifications control various properties of proteins and correlate with many diseases, including cancer (3). Although comprehensive genomic and proteomic analyses have identified many key drivers of hepatocellular carcinoma, the posttranslational modifications remain poorly understood (4,5). Thus, elucidation of the precise molecular mechanisms underlying hepatocellular carcinoma progression is of great importance for developing new reagents to treat this aggressive disease.
Glycosylation is the most common posttranslational modification of proteins, and aberrant glycosylation is often observed in cancers (6,7). Accumulated evidence indicates that alterations in N-linked glycosylation are a hallmark of various liver diseases, including hepatocellular carcinoma (5,8). For instance, expression of N-acetylglucosaminyltransferase-III and -V is increased in hepatocellular carcinoma (9,10). An N-glycan profiling study identified novel N-glycan structures in serum as prognostic markers of hepatocellular carcinoma (11). In addition, a-1,6-fucosyltransferase can generate fucosylated a-fetoprotein (AFP), which provided a more accurate diagnosis of hepatocellular carcinoma from chronic liver diseases (12,13). However, changes in O-linked glycosylation have been overlooked in the past. The O-glycosylation of proteins is difficult to explore, as consensus amino acid sequences of O-glycosylation remain unclear and effective releasing enzymes for O-glycans are not available (14). Recently, a systematic analysis of mucin-type O-linked glycosylation revealed that mucin type O-glycans are decorated not only on mucins but on various unexpected proteins, and functions of the O-glycosylation are largely unknown (15). Several lines of evidence indicate that O-glycosylation of proteins plays critical roles in cancer. O-glycans on major histocompatibility complex class I-related chain A (MICA) enhance bladder tumor metastasis (16), and O-glycosylation of death receptor controls apoptotic signaling in several types of cancer (17). In addition, our previous study showed that GALNT2, an O-glycosyltransferase, regulates EGF receptor activity and cancer behaviors in hepatocellular carcinoma cells (18). Therefore, understanding the roles of O-glycosylation in hepatocellular carcinoma may provide novel insights into the pathogenesis of hepatocellular carcinoma.
Core 1 b1,3-galactosyltransferase (C1GALT1) is a critical mucin-type O-glycosyltransferase that is localized in the Golgi apparatus (19,20). C1GALT1 transfers galactose (Gal) to Nacetylgalactosamine (GalNAc) to a serine (Ser) or threonine (Thr) residue (Tn antigen) to form the Galb1-3GalNAca1-Ser/ Thr structure (T antigen or core 1 structure; ref. 21). The core 1 structure is the precursor for subsequent extension and maturation of mucin-type O-glycans (22). C1GALT1 has been shown to regulate angiogenesis, thrombopoiesis, and kidney development (23,24). Although mucin-type O-glycosylation and C1GALT1 have been shown to play crucial roles in a variety of biologic functions, the expression and role of C1GALT1 in hepatocellular carcinoma remain unclear. Here, we found that C1GALT1 is frequently overexpressed in hepatocellular carcinoma and its expression is associated with poor survival of hepatocellular carcinoma patients. We therefore hypothesized that C1GALT1 can regulate the malignant growth of hepatocellular carcinoma cells and contribute to the pathogenesis of hepatocellular carcinoma.
Human tissue samples
Postsurgery frozen hepatocellular carcinoma tissues for RNA extraction and Western blotting and paraffin-embedded tissue sections were obtained from the National Taiwan University Hospital (Taipei, Taiwan). This study was approved by the Ethical Committees of National Taiwan University Hospital, and all patients gave informed consent to have their tissues before collection.
Cell culture
Human liver cancer cell lines, Huh7, PLC5, Sk-Hep1, and HepG2, were purchased from Bioresource Collection and Research Center in the year 2008. HA22T, SNU387, and HCC36 cells were kindly provided by Prof. Shiou-Hwei Yeh (National Taiwan University) in the year 2010. All cell lines were authenticated by the provider based on morphology, antigen expression, growth, DNA profile, and cytogenetics. Cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) containing 10% FBS in 5% CO 2 at 37 C. To analyze growth factor-induced cell signaling, cells were starved in serum-free DMEM for 5 hours and then treated with 25 ng/mL of hepatocyte growth factor (HGF) or insulin-like growth factor (IGF) at 37 C for 30 minutes.
Tissue array and immunohistochemistry
Paraffin-embedded human hepatocellular carcinoma tissue microarrays were purchased from SuperBioChips and BioMax. Arrays were incubated with anti-C1GALT1 monoclonal antibody (1:200) in 5% bovine serum albumin/PBS and 0.1% Triton X-100 (Sigma) for 16 hours at 4 C. After rinsing twice with PBS, SuperSensitive Link-Label IHC Detection System (BioGenex) was used and the specific immunostaining was visualized with 3,3-diaminobenzidine liquid substrate system (Sigma). All sections were counterstained with hematoxylin (Sigma).
Transfection
To overexpress C1GALT1, cells were transfected with pcDNA3.1/C1GALT1/mycHisplasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. Empty pcDNA3.1/mycHisplasmid was used as mock transfectant. The transfected cells were selected with 600 mg/mL of G418 for 14 days and then pooled for further studies.
RNA interference
Two siRNA oligonucleotides against C1GALT1 (5 0 -UUA-GUAUACGUUCAGGUAA GGUAGG-3 0 and 5 0 -UUA UGU UGG CUA GAA UCU GCA UUG A-3 0 ) and a negative control siRNA of medium GC were synthesized by Invitrogen. For knockdown of C1GALT1, cells were transfected with 20 nmol of siRNA using Lipofectamine RNAiMAX (Invitrogen) for 48 hours. The pLKO/ C1GALT1-shRNA plasmid and nontargeting pLKO plasmids were purchased from National RNAi Core Facility (Academia Sinica, Taipei, Taiwan). The short hairpin RNA (shRNA) plasmids were transfected with Lipofectamine 2000 and selected with 500 ng/mL of puromycin for 10 days. Knockdown of C1GALT1 in single colonies was confirmed by Western blotting.
Western blotting
Western blotting was conducted as reported previously (18).
Phospho-receptor tyrosine kinase array
A human phospho-receptor tyrosine kinase (p-RTK) array kit was purchased from R&D systems. Hepatocellular carcinoma cells were serum starved for 5 hours and then treated with 20% FBS for 30 minutes. Cells were lysed and 500 mg of proteins were subjected to Western blotting according to the manufacturer's protocol.
Cell viability, proliferation, and cell cycle
Cells (4 Â 10 4 ) were seeded in each well of 6-well plates with DMEM containing 10% FBS. Viable cells were analyzed by Trypan blue exclusion assay at 0, 24, 48, and 72 hours. Cell proliferation was evaluated by immunostaining with anti-Ki67 antibody (1:1000; Vector Laboratories). For cell-cycle analysis, 1 Â 10 6 cells were stained with propidium iodide (Sigma) for 30 minutes. The percentages of cells in G 1 , S, and G 2 -M phases were analyzed by flow cytometry (Becton Dickinson).
Deglycosylation and lectin pull down
Protein deglycosylation was carried out using an Enzymatic Protein Deglycosylation Kit (Sigma). Briefly, cell lysates were treated with neuraminidase or PNGase F at 37 C for 1 hour. For lectin blotting, 20 mg of cell lysate was separated by an 8% SDS-PAGE, transferred to polyvinylidene difluoride membrane (Millipore), and blotted with biotinylated VVA (1:10,000). For lectin pull-down assay, cell lysates (0.3 mg) were incubated with or without deglycosylation enzymes and then applied to VVA-or PNA-conjugated agarose beads at 4 C for 16 hours. The pulled down proteins were analyzed by Western blotting.
Tumor growth in immunodeficient mice
For tumor growth analysis, 7 Â 10 6 of hepatocellular carcinoma cells were subcutaneously injected into severe combined immnodeficient (SCID) mice (n ¼ 4 for each group). Tumor volumes were monitored for 36 or 56 days. Excised tumors were weighed and lysed for Western blotting and immunohistochemistry.
Dimerization of MET
To analyze dimerization of MET, hepatocellular carcinoma cells were incubated with or without 25 ng/mL of human HGF in DMEM on ice for 5 minutes. Cross-linker Bis(sulfosuccinimidyl) suberate (BS 3 , 0.25 mmol/L, Thermo Scientific) was added to cells and reacted at 37 C for 5 minutes. Cells were then transferred on ice for 10 minutes. Reactions were blocked by adding 50 mmol/L of Tris-HCl (pH 7.4). Cell lysates were separated by 6% SDS-PAGE and immunoblotted with anti-MET antibody.
Statistical analysis
Student t test was used for statistical analyses. Paired t test was used for the analyses of paired hepatocellular carcinoma tissues. Mann-Whitney U test was used to compare unpaired nontumor liver tissue and hepatocellular carcinoma tissues. Two-sided Fisher exact test was used for comparisons between C1GALT1 expression and clinicopathologic features. Kaplan-Meier analysis and the log-rank test were used to estimate overall survival. Pearson correlation test was used to assess C1GALT1 and phospho (p)-MET expression. Data are presented as means AE SD. P < 0.05 is considered statistically significant.
Expression of C1GALT1 is upregulated in hepatocellular carcinoma and correlates with advanced tumor stage, metastasis, and poor survival
We first investigated expression of C1GALT1 in hepatocellular carcinoma tissues. Paired hepatocellular carcinoma and adjacent nontumor liver tissues (n ¼ 16) were analyzed by realtime reverse transcriptase PCR (RT-PCR). Results showed that C1GALT1 mRNA was significantly upregulated in hepatocellular carcinoma tissues compared with adjacent nontumor liver tissues (paired t test, P < 0.05; Fig. 1A). Consistently, Western blotting showed that C1GALT1 protein was overexpressed in hepatocellular carcinoma tissues of paired specimens (Fig. 1B). We further conducted immunohistochemical analysis for 72 primary hepatocellular carcinoma tissues and 32 nontumor livers to investigate the expression of C1GALT1. The immunohistochemistry showed dot-like precipitates of C1GALT1 in the cytoplasm of hepatocellular carcinoma (Fig. 1C), which is similar to the intracellular localization of the Golgi apparatus in hepatocytes (25). We did not observe expression of C1GALT1 in surrounding stromal cells under our experimental conditions. The intensity of staining was scored according to the percentage of C1GALT1-positive cells in each sample (0, negative; þ1, < 20%; þ2, 20%-50%; þ3, > 50%). Our data revealed that C1GALT1 was highly expressed (þ2 and þ3) in 54% of hepatocellular carcinoma tumors, whereas only 19% of nontumor liver tissues expressed high levels of C1GALT1 (Mann-Whitney U Test, P ¼ 0.002; Fig. 1C and D). Consistently, results from tissue microarrays also showed increased expression of C1GALT1 in hepatocellular carcinomas compared with normal liver tissues (Supplementary Fig. S1). These results indicated that C1GALT1 expression was significantly higher in hepatocellular carcinoma than that in nontumor liver tissues.
We next investigated the relationship between C1GALT1 expression and clinicopathologic features in patients with hepatocellular carcinoma. We found that high expression of C1GALT1 correlated with advanced tumor stage (Fisher exact test, P < 0.05) and metastasis (Fisher exact test, P < 0.01) of hepatocellular carcinoma tumors (Table 1 and Supplementary Table S1). A Kaplan-Meier survival analysis showed that the survival rate of patients with hepatocellular carcinoma with high C1GALT1 expression was significantly lower than those with low C1GALT1 expression.(log-rank test, P ¼ 0.001; Fig. 1E). Collectively, these data suggest that C1GALT1 is frequently upregulated in hepatocellular carcinoma and its expression is associated with advanced tumor stage, metastasis, and poor survival in hepatocellular carcinoma.
C1GALT1 modifies mucin-type O-glycans on hepatocellular carcinoma cells
To investigate functions of C1GALT1 in hepatocellular carcinoma, knockdown or overexpression of C1GALT1 was conducted in multiple hepatocellular carcinoma cell lines. Western blotting showed that the average expression level of C1GALT1 was significantly higher in hepatocellular carcinoma cell lines compared with that of nontumor liver tissues (Fig. 2A). We conducted knockdown of C1GALT1 with two different C1GALT1-specific siRNAs in HA22T and PLC5 cells, which express high levels of C1GALT1 and overexpression of C1GALT1 in Sk-Hep1 and HCC36 cells as they express low levels of C1GALT1 (Fig. 2B). Immunofluorescence microscopy further confirmed the knockdown and overexpression of C1GALT1 in hepatocellular carcinoma cells and its subcellular localization in the Golgi apparatus (Supplementary Fig. S2). Furthermore, we showed that knockdown of C1GALT1 enhanced binding of VVA to glycoproteins, whereas overexpression of C1GALT1 decreased the VVA binding (Fig. 2B), indicating that C1GALT1 catalyzes the formation of Tn to T antigen. Seven proteins had evident changes in VVA binding, including p50, p60, p80, p90, p110, p140, and p260, as shown in Fig. 2B. Among them, p140 showed changes in all four tested cell lines. Consistently, flow cytometry showed that C1GALT1 altered VVA binding to the surface of hepatocellular carcinoma cells (Fig. 2C). These results indicate that C1GALT1 modulates the expression of mucin-type O-glycans on hepatocellular carcinoma cells.
C1GALT1 regulates hepatocellular carcinoma cell proliferation in vitro and in vivo
Effects of C1GALT1 on cell viability and proliferation of hepatocellular carcinoma cells were analyzed by Trypan blue exclusion assays and fluorescent staining of Ki67, respectively. Knockdown of C1GALT1 significantly suppressed cell viability, whereas overexpression of C1GALT1 enhanced cell viability (Fig. 3A). Furthermore, Ki67 staining showed that C1GALT1 modulated cell proliferation (Fig. 3B). We also found that knockdown of C1GALT1 led to G 1 -phase arrest in HA22T and PLC5 cells, whereas overexpression of C1GALT1 increased the number of cells in S phase in Sk-Hep1 cells (Supplementary Fig. S3).
To analyze the effect of C1GALT1 on tumor growth in vivo, we stably knocked down C1GALT1 with C1GALT1-specific shRNA in HA22T and PLC5 cells (Supplementary Fig. S4). Clone number 8 of HA22T cells and clone number 10 of PLC5 cells were subcutaneously xenografted in SCID mice. The results showed that knockdown of C1GALT1 significantly suppressed the volume and weight of hepatocellular carcinoma tumors. Immunohistochemistry of excised tumors for Ki67 expression revealed that knockdown of C1GALT1 suppressed tumor cell proliferation in vivo (Fig. 3C). Knockdown of the C1GALT1 in the tumors was confirmed by Western blotting (Supplementary Fig. S5). These data provide evidence that C1GALT1 can modulate hepatocellular carcinoma cell proliferation in vitro and in vivo.
C1GALT1 regulates phosphorylation of MET
Because RTKs are crucial for hepatocellular carcinoma proliferation (4,26) and their activities have been found to be regulated by O-glycosylation (18,27), we investigated whether C1GALT1 could affect RTK signaling pathways in hepatocellular carcinoma cells. A human p-RTK array was used to detect the tyrosine phosphorylation level of 42 different RTKs. Our data indicated that knockdown of C1GALT1 in HA22T cells decreased phosphorylation of ERBB3, MET, and EPHA2, whereas phosphorylation of VEGFR1 was increased (Fig. 4A). MET plays crucial roles in multiple functions in hepatocellular carcinoma, including cell proliferation (28), hepatocarcinogenesis (29), and metastasis (28). In addition, NetOGlyc 3.1 predicts one potential Oglycosylation site in the extracellular domain of MET (data not shown). Therefore, we further investigated the role of C1GALT1 in glycosylation and activation of MET in hepatocellular carcinoma cells. Our results showed that knockdown of C1GALT1 inhibited HGF-induced phosphorylation of MET at Y1234/5 and suppressed phosphorylation of AKT in HA22T and PLC5 cells (Fig. 4B, top). In contrast, overexpression of C1GALT1 enhanced HGF-induced activation of MET and increased p-AKT levels in Sk-Hep1 and HCC36 cells. In addition, C1GALT1 expression did not significantly alter IGF-I-induced signaling in all tested hepatocellular carcinoma cell lines (Fig. 4B, bottom). These results suggest that C1GALT1 selectively activates the HGF/MET signaling pathway.
To investigate the role of the MET signaling pathway in C1GALT1-enhanced cell viability, we treated hepatocellular carcinoma cells with PHA665757, a specific inhibitor of MET phosphorylation (30). Trypan blue exclusion assays showed that C1GALT1-enhanced cell viability was significantly inhibited by the blockade of MET activity (Fig. 4C). In addition, we observed that knockdown of C1GALT1 decreased HGFinduced cell migration and invasion, whereas overexpression of C1GALT1 enhanced HGF-induced cell migration and invasion (Supplementary Fig. S6).
We next analyzed whether C1GALT1 expression correlated with MET activation in primary hepatocellular carcinoma tissues. Western blotting (Fig. 4D) and Pearson test (Fig. 4E) from 20 hepatocellular carcinoma tumors showed a significant correlation (R 2 ¼ 0.73, P < 0.0001) between expression levels of C1GALT1 and phospho-MET. These results suggest that C1GALT1 could regulate MET activation in hepatocellular carcinoma.
C1GALT1 modifies O-glycans on MET and regulates HGF-induced dimerization of MET
To investigate the mechanisms by which C1GALT1 regulates HGF/MET signaling, we analyzed the effects of C1GALT1 on glycosylation and dimerization of MET in hepatocellular carcinoma cells. Because C1GALT1 is an O-glycosyltransferase, we first analyzed whether MET is O-glycosylated using VVA and PNA lectins, which recognize tumor-associated Tn and T antigen, respectively. Lectin pull-down assays with VVA or PNA agarose beads showed that endogenous MET expressed Tn and T antigens in all seven hepatocellular carcinoma cell lines tested (Supplementary Fig. S7). Moreover, we found that VVA binding to MET in HA22T cells was further increased after removal of N-glycans on MET with PNGaseF (Fig. 5A), indicating the specificity of VVA binding to O-glycans on MET. We also observed that removal of sialic acids by neuraminidase enhanced VVA binding, suggesting that MET expresses sialyl Tn in addition to Tn. These findings strongly suggest that MET expresses short mucin-type O-glycans in hepatocellular carcinoma cells.
To investigate whether C1GALT1 can modify O-glycans on MET, VVA binding to MET was analyzed in hepatocellular carcinoma cells with C1GALT1 knockdown or overexpression. We found that knockdown of C1GALT1 increased VVA binding to MET in both HA22T and PLC5 cells (Fig. 5B). Conversely, overexpression of C1GALT1 decreased VVA binding to MET in Sk-Hep1 and HCC36 cells. Consistently, we observed that removal of sialic acids enhanced VVA binding to MET in these cell lines. These findings indicate that C1GALT1 can modify O-glycans on MET in hepatocellular carcinoma cells.
We next explored the effects of altered O-glycosylation on MET properties. Our results showed that C1GALT1 expression did not significantly alter the protein level of MET analyzed by Western blotting (Fig. 5B) and flow cytometry (data not shown). Because HGF-induced dimerization of MET is an initial and crucial step for the activation of MET signaling (31), we analyzed whether C1GALT1 could affect MET dimerization. Our data showed that knockdown of C1GALT1 suppressed HGF-induced dimerization of MET in both HA22T and PLC5 cells. In contrast, overexpression of C1GALT1 enhanced
h 24 h 48 h 72 h 0 h 24 h 48 h 72 h 0 h 24 h 48 h 72 h 0 h 24 h 48 h 72 h
Ctr si C1GALT1 si-1 C1GALT1 si-2 dimerization of MET in Sk-Hep1 and HCC36 cells (Fig. 5C). These results suggest that C1GALT1 modifies O-glycans on MET and regulates HGF-induced dimerization of MET in hepatocellular carcinoma cells.
Discussion
This study showed that overexpression of C1GALT1 in hepatocellular carcinoma tissues was associated with advanced tumor stage, metastasis, and poor prognosis. C1GALT1 expression regulated hepatocellular carcinoma cell viability and proliferation in vitro and in vivo. The C1GALT1enhaced cell viability was inhibited by MET inhibitor. MET carried O-glycans, and these structures were modified by C1GALT1. Furthermore, C1GALT1 could regulate HGFinduced dimerization and activity of MET in hepatocellular carcinoma cells. Taken together, this study is the first to show that C1GALT1 was able to regulate hepatocellular carcinoma cell proliferation in vitro and in vivo, and modulation of O-glycosylation and activity of MET may be involved in this process. Our findings provide novel insights not only into the role of O-glycosylation in the pathogenesis of hepatocellular carcinoma but also in the development of reagents for hepatocellular carcinoma treatment.
In colon and breast cancer, an increase in the expression of short O-glycans, such as Tn, sialyl Tn, T, and sialyl T, often alters the function of glycoproteins and their antigenic property, as well as the potential of cancer cells to invade and metastasize (32). Short O-glycans have been developed as carbohydrate vaccines for cancer treatment (33). The expression of short O-glycans in human hepatocellular carcinoma has been reported (34,35). However, glycogenes responsible for these O-glycans and their functions in hepatocellular carcinoma remain largely unknown. Previously, we reported that GALNT1 and GALNT2 are the major GalNAc transferases in liver tissues and that GALNT2 modulates the sialyl Tn expression on EGF receptor in hepatocellular carcinoma cells and suppresses their malignant properties (18). Here, we further report that C1GALT1 expression is dysregulated in hepatocellular carcinoma and C1GALT1 modulates the expression of mucin-type O-glycans on hepatocellular carcinoma cell surfaces. We found that O-glycans can be decorated on MET, an important protooncogene in a variety of human cancers. Our data showed that MET from all tested seven hepatocellular carcinoma cell lines could be pulled down by VVA and PNA lectins, suggesting the presence of Tn and T antigens on MET. Removal of sialic acids by neuraminidase enhanced VVA binding to MET, indicating that some of the Tns are sialylated in hepatocellular carcinoma cells. Removal of N-glycans by PNGaseF further enhanced VVA binding to MET. Moreover, prediction of O-glycosylation sites using NetOGlyc 3.1 indicates that there is one potential O-glycosylation site in the extracellular domain of MET. These results strongly suggest the presence of O-glycans on MET. Glycosylation has long been proposed to control various protein properties, including dimerization, enzymatic activity, secretion, subcellular distribution, and stability of RTKs (14,36). However, most studies focused on effects of N-glycans on RTKs. Importantly, we found that C1GALT1 can modulate the O-glycans on MET and enhance dimerization and phosphorylation of MET. Because receptor dimerization is a key regulatory step in RTK signaling (37), it is highly possible that C1GALT1 modulates MET activity via the enhancement of its dimerization. To our knowledge, we are the first to report that MET expresses O-glycans and changes in these carbohydrates regulate the activity of MET. It will be of great interest to further investigate the exact structures and sites of O-glycans on MET to understand how Oglycosylation modulates the structure and function of RTKs.
Recent studies have shown that aberrant activation of MET signaling correlates with the increased cell proliferation, poor prognosis, and poor outcome of human hepatocellular carcinoma (38)(39)(40). HGF/MET signaling has been shown to promote invasion and metastasis of hepatocellular carcinoma cells (41,42). Our data show that C1GALT1 can increase dimerization and phosphorylation of MET. Consistent with previous findings, we also found that C1GALT1 can enhance HGFinduced migration and invasion. Targeting MET is considered to be an attractive strategy for treating many human cancers, including hepatocellular carcinoma (43). Thus, a complete understanding of the mechanisms by which the structure and function of MET signaling are regulated is crucial to improve the effect of MET-targeted therapies in human cancers. This study provides novel insights into the role of O-glycosylation in modulating MET activity. It will be important to further investigate whether changes in O-glycans on MET can affect hepatocellular carcinoma cell sensitivity toward targeted therapeutic drugs, including small-molecule inhibitors and therapeutic antibodies. We found that C1GALT1 expression modulates HGF-, but not IGF-mediated signaling, suggesting the selectivity of C1GALT1 activity toward certain RTKs. However, we observed that C1GALT1 expression changes binding patterns of VVA to several glycoproteins and knockdown of C1GALT1 in hepatocellular carcinoma cells also modulates phosphorylation of ERBB3, VEGFR1, and EPHA2, suggesting that there are other acceptor substrates, in addition to MET. Therefore, it remains possible that several signaling pathways may be involved in mediating the biologic functions of C1GALT1 in hepatocellular carcinoma cells. Thus, targeting C1GALT1 could have effects similar to those from targeting multiple RTKs. This study opens up avenues for treating cancers by targeting not only the receptors themselves but also their O-glycosylation regulators.
Figure 1 .
Figure 1. Expression of C1GALT1 in human hepatocellular carcinoma. A, expression of C1GALT1 mRNA in 16 paired hepatocellular carcinoma tissues. The mRNA levels of C1GALT1 were analyzed by real-time RT-PCR and normalized to GAPDH. Paired t test, Ã , P ¼ 0.013. B, Western blot analyses showing C1GALT1 expression in paired hepatocellular carcinoma tissues from A. Representative images are shown. N, nontumor liver tissue; T, tumor tissue. C, immunohistochemistry of C1GALT1 in paired primary hepatocellular carcinoma tissues. The brown stained cells in the tumor part are hepatocellular carcinoma cells, and those in the nontumor part are hepatocytes. The staining was visualized with a 3,3-diaminobenzidine liquid substrate system, and all sections were counterstained with hematoxylin. Representative images of tumor (left) and nontumor liver tissue (right) are shown. The negative control does not show any specific signals (bottom left of the top panel). Amplified images are shown at the bottom. Scale bars, 50 mm. D, statistical analysis of immunohistochemistry in hepatocellular carcinoma tissues. The intensity of C1GALT1 staining in tissues is shown at the top. Scale bars, 50 mm. Results are shown at the bottom. Mann-Whitney U Test, P ¼ 0.002. E, Kaplan-Meier analysis of overall survival for patients with hepatocellular carcinoma. The analyses were conducted according to the immunohistochemistry of C1GALT1. Probability of overall survival was analyzed after the initial tumor resection. Log-rank test, P ¼ 0.001.
Figure 2 .
Figure 2. C1GALT1 regulates O-glycosylation in hepatocellular carcinoma cells. A, expression of C1GALT1 in hepatocellular carcinoma cell lines. C1GALT1 expression in seven hepatocellular carcinoma cell lines and nine nontumor (N) liver tissues was analyzed by Western blotting. GAPDH was used as an internal control. Signals were quantified by ImageQuant5.1.Ã , P < 0.05. B, effects of C1GALT1 on binding of VVA to glycoproteins. Western blotting shows C1GALT1 expression after knockdown with two C1GALT1 siRNAs compared with control (Ctr) siRNA in HA22T cells and PLC5 cells. Overexpression of C1GALT1 in Sk-Hep1 and HCC36 cells transfected with C1GALT1/pcDNA3.1 plasmid (C1GALT1) was compared with pcDNA3.1 empty plasmid (mock). The changes in O-glycans on glycoproteins were revealed by Western blotting with biotinylated VVA. Proteins with evident changes in VVA binding are indicated by arrows.p140 changes in all tested cell lines are indicated by red arrows. C, effects of C1GALT1 on O-glycans of hepatocellular carcinoma cell surfaces. Surface O-glycans were analyzed by flow cytometry with FITC-VVA. Negative (À) refers to cells without addition of VVA-FITC.
Figure 3 .
Figure 3. C1GALT1 regulates hepatocellular carcinoma cell proliferation in vitro and in vivo. A, C1GALT1 modulated cell viability in vitro. Cell viability of HA22T, PLC5, Sk-Hep1, and HCC36 cells was analyzed by Trypan blue exclusion assays at different time points for 72 hours. The results are standardized by the cell number at 0 hour. Data are represented as means AE SD from three independent experiments.Ã , P < 0.05; ÃÃ , P < 0.01. B, effects of C1GALT1 on cell proliferation. Cells were immunofluorescently stained for Ki67 and Ki67-positive cells were counted under a microscope. Results are presented as means AE SD from three independent experiments.Ã , P < 0.05; ÃÃ , P < 0.01. C, effects of C1GALT1 on hepatocellular carcinoma tumor growth and proliferation in SCID mouse model. HA22T (top) and PLC5 (bottom) cells were subcutaneously injected into SCID mice. Four mice were used for each group. The volume of tumors was measured at different time points, as indicated (left). Mice were sacrificed at day 56 for HA22T cells and day 36 for PLC5 cells. Tumors were excised and weighted (middle). Cell proliferation of tumor cells was evaluated by immunohistochemical staining for Ki67, and representative images are shown (right). Results are presented as the mean AE SD from 4 mice for each group.Ã , P < 0.05.
Figure 4 .
Figure 4. C1GALT1 regulates activity of MET in hepatocellular carcinoma cells. A, human p-RTK array showing the effect of C1GALT1 on the phosphorylation of RTKs. Cell lysates of control and C1GALT1 knockdown HA22T cells were applied to p-RTK array including 42 RTKs. B, C1GALT1 modulates HGF-induced signaling in hepatocellular carcinoma cells. HA22T, PLC5, Sk-Hep1, and HCC36 cells were starved for 5 hours and then treated with (þ)/without (À) HGF (25 ng/mL) or IGF (25 ng/mL) for 30 minutes. Cell lysates (20 mg) were analyzed by Western blotting with various antibodies, as indicated. C, effects of MET inhibitor, PHA665752, on C1GALT1-enhanced cell viability. Sk-Hep1 and HCC36 cells were treated with PHA665752 at the indicated concentration and then analyzed by Trypan blue exclusion assays at 72 hours. Data are represented as means AE SD from three independent experiments.ÃÃ , P < 0.01. D, expression of C1GALT1 and p-MET in hepatocellular carcinoma tissues. Tissue lysates (20 mg for each tumor) were analyzed by Western blotting. Signals of Western blotting were quantified by ImageQuant5.1.b-actin was a loading control. E, correlation of C1GALT1 and p-MET expression in 20 hepatocellular carcinoma tumors. Pearson test was used to analyze the statistical correlation of C1GALT1 and p-MET expression in D.
Figure 5 .
Figure 5. C1GALT1 regulates glycosylation and dimerization of MET. A, MET is decorated with short O-glycans. Lysates of HA22T cells were treated with neuraminidase and/or PNGaseF and then pulled down (PD) by VVA agarose beads. The pulled down proteins were analyzed by immunoblotting (IB) with anti-MET antibody. The molecular mass is shown on the right. B, C1GALT1 modifies O-glycosylation of MET in hepatocellular carcinoma cells. Cell lysates were treated with (þ) or without (À) neuraminidase and then pulled down by VVA agarose beads. The pulled down glycoproteins were immunoblotted (IB) with anti-MET antibody. C, C1GALT1 regulates dimerization of MET in hepatocellular carcinoma cells. Hepatocellular carcinoma cells were starved for 5 hours and then treated with (þ) or without (À) 25 ng/mL of HGF. Cell lysates were cross-linked by BS 3 and then analyzed by Western blotting with anti-MET antibody. The arrows indicate the dimer (D) of MET, and the arrowheads indicate the monomer (M). Markers of molecular weight are shown on the left. GAPDH is an internal control.
Table 1 .
Correlation of C1GALT1 expression with clinicopathologic features.
==
Domain: Biology Medicine
|
[{"added": "2017-04-12T21:07:54.504Z", "created": "2013-09-01T00:00:00.000", "id": "10508715", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://aacr.figshare.com/articles/journal_contribution/Supplementary_Table_from_C1GALT1_Enhances_Proliferation_of_Hepatocellular_Carcinoma_Cells_via_Modulating_MET_Glycosylation_and_Dimerization/22396421/1/files/39842123.pdf", "pdf_hash": "9488d5b20f6aaca2a1b8847c5a15f9ffef07ca40", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:39462", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "0b845a37e92ed3cfc3da2a7d5bd8664a4d4a1058", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Corticotropin-releasing hormone and extracellular mitochondria augment IgE-stimulated human mast-cell vascular endothelial growth factor release, which is inhibited by luteolin
Background Autism spectrum disorders (ASDs) are neurodevelopmental disorders characterized by varying degrees of dysfunctional social abilities, learning deficits, and stereotypic behaviors. Many patients with ASDs have ‘allergy-like’ symptoms and respond disproportionally to stress. We have previously shown that the peptide neurotensin (NT) is increased in the serum of young children with autism and that can stimulate extracellular secretion of mitochondrial (mt)DNA which was also increased in the serum of these children. Methods Human mast cells were stimulated by corticotropin-releasing hormone (CRH), mitochondrial DNA, IgE/anti-IgE, either for 24 hours to measure vascular endothelial growth factor (VEGF) release by ELISA or for 6 hours or quantitative PCR. Results CRH augmented IgE/anti-IgE-induced human mast-cell release of VEGF and it also induced the expression of IgE receptor (FcεRI) on mast cells. Moreover, sonicated mitochondria also augmented VEGF release, and this effect was blocked by the natural flavone luteolin. Conclusion These results indicate that stress and infection-mimicking extracellular mitochondrial components augment allergic inflammation that may be involved in the early pathogenesis of ASDs. Moreover, luteolin inhibits these processes and may be helpful in the treatment of ASDs.
Introduction
Autism spectrum disorders (ASDs) are pervasive developmental disorders for which no distinct pathogenesis, biomarkers, or effective treatment have been identified. ASDs involve some immune dysfunction in the patient [1] or in the mother during gestation [2], and may have a neuroimmune component [3]. Many children with ASDs also have atopic features [4] or food allergies [5][6][7] that present as 'allergy-like' symptoms [7,8]. Such symptoms often occur in the absence of increased serum IgE levels or positive skin-prick tests, suggesting mast-cell activation by non-immune triggers [9]. Increased anxiety seems to be present in at least a subgroup of patients with ASDs, who may also be more prone to stress [10].
We previously showed that corticotropin-releasing hormone (CRH), secreted under stress, could induce release of vascular endothelial growth factor (VEGF) from human mast cells [11]. We found that the neuropeptide neurotensin (NT), which is present in both the brain and gut, is significantly increased in the serum of young children with autism [12]. It is interesting that the distribution of NT receptors is more concentrated in the brain Broca area [13], which regulates speech, a function commonly lost in children with autism [14]. We also found that the serum of the same patients had higher levels of extracellular mitochondrial (mt)DNA [15], and NT stimulated release of extracellular mtDNA from human cultured mast cells [15]. We also found that the natural flavonoid luteolin can inhibit the ability of IgE [16] and mercury [17] to induce VEGF release from human mast cells.
In the current study, we investigated the effect of CRH and mitochondria on VEGF release from IgE/ anti-IgE-stimulated human mast cells, the effect of CRH on gene expression of the high affinity IgE receptor (FcεRI), and the effect of the flavone luteolin on VEGF release.
Methods
The study was approved by the human institutional review board of Tufts Medical Center (Boston, MA, USA) under Exemption Number 4 for discarded samples without any identifiers.
Culture of human mast cells
Human umbilical cord blood was collected from mothers who had normal uncomplicated deliveries at Tufts Medical Center. Human cord blood-derived cultured mast cells (hCBMCs) were prepared using hematopoetic stem cells (CD34 + ) isolated by positive selection of CD34 + /AC133 + cells by magnetic cell sorting using an AC133 + cell isolation kit (Milletnyi Biotec, Auburn, CA, USA) as previously reported [18]. CD34 + cells were grown in serum-free expansion medium (StemSpan; StemCell Technologies, Vancouver, BC, Canada), supplemented with 100 ng/ml recombinant human stem cell factor (rhSCF; kindly supplied by Sweden Orphan Biovitrum AB, Stockholm, Sweden), 100 U/ml penicillin, 100 μg/ml streptomycin (Invitrogen, Carlsbad, CA, USA) and IL-3 (R&D Systems, Minneapolis, MN, USA) for the first 3 weeks, then in the serum-free expansion medium with 50 ng/ml IL-6 (Peprotech, Rocky Hill, NJ, USA) and for 8 to 10 weeks, with fetal bovine serum (Invitrogen/Gibco, Carlsbad, CA, USA) added from week 6. The purity of the hCBMCs was evaluated by immunocytochemical staining for tryptase [18]. hCBMCs cultured for 7 to 10 weeks were used for the experiments.
Mitochondrial preparation
A commercial kit (Mitochondria Isolation Kit for Cells; Pierce Scientific, Rockford, IL, USA) was used to isolate mitochondria from cultured mast cells. Mitochondria were isolated under sterile conditions at 4°C in accordance with the manufacturer's instructions, and then subjected to sonication for 2 minutes at 4°C to release all inner components. The mtDNA and protein concentrations were determined by UV spectrophometry (NanoDrop 2000; Thermo Scientific, Waltham, MA, USA). The purity of the mitochondrial fraction was confirmed by the absence of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and lactate dehydrogenase (markers of microsomal contamination) and of 5′ nucleotidase and glucose-6-phosphatase (markers of cytoplasmic contamination).
Quantitative PCR
Total RNA from cultured mast cells and human skin biopsies was isolated using a commercial kit (RNeasy Mini Kit; Qiagen, Valencia, CA, USA) and reagent (Trizol; Invitrogen) respectively, in accordance with the manufacturer's instructions. Reverse transcription was performed with 300 ng of total RNA using the iScript cDNA synthesis kit (BIO-RAD, Hercules, CA). To measure FcεRI expression, cells were incubated for 6 hours with CRH (Sigma-Aldrich, MA, USA) and quantitative PCR was performed using Taqman gene expression assays (Applied Biosystems, Foster City, CA, USA). Samples were run at 45 cycles using a real-time PCR system (7300; Applied Biosystems). Relative mRNA abundance was determined from standard curves run with each experiment. mRNA gene expressions were normalized to GAPDH endogenous control. (Hu, VIC TAMRA; Applied Biosystems)
Statistical analysis
All experiments were performed in triplicate (n = 3) and repeated (n = 5). Results are presented as mean ± SD. Data from stimulated and control samples were compared using the unpaired two-tailed, Student's t-test. p < 0.05 was considered significant.
Results
Because many children with autism have allergic symptoms are more anxious and over-react to stress, we investigated if CRH would affect allergic mast-cell activation. Addition of CRH (10 μmol/l) together with or after anti-IgE had no effect on anti-IgE-induced VEGF release (results not shown). Pretreatment of LAD2 cells for 24 hours with CRH (10 μmol/l) followed by 2 hours of incubation with IgE augmented VEGF release in response to anti-IgE ( Figure 1A). The amount of CRH required was high because LAD2 cells do not express many CRH receptors.
We then investigated if such augmentation might be due to increase in FcεRI gene expression. Incubation of hCBMCs with CRH (0.1, 1, 10 μmol/l) for 6 hours increased FcεRI gene expression by almost five-fold ( Figure 1B). Incubation of mast cells with CRH (10 μmol/l) for 48 hours increased FcεRI gene expression by almost 200-fold (results not shown).
We then investigated the effect of mitochondria. Treatment of hCBMCs with sonicated mitochondria (0.1 and 10 microgram/μl) stimulated some VEGF release (Figure 2), but addition of mitochondria to anti-IgE-stimulated mast cells significantly increased VEGF release (Figure 2). Pretreatment with luteolin (100 μmol/l) for 30 minutes completely inhibited the VEGF release induced by mitochondria and anti-IgE, and even caused it to drop below basal levels ( Figure 2).
Discussion
In this report, we show that CRH not only can augment allergic mast-cell release of VEGF, but can also induce FcεRI expression in these human mast cells. Our finding is specific, because the peptide substance P had been shown previously to decrease FcεRI gene expression [19], as also does lipopolysaccharide [20]. These results could explain how stress may worsen allergy-like symptoms in patients with ASDs [6,8,21]. It has previously been shown that CRH can augment NT-induced VEGF release [22]. Hence, CRH might augment both allergic and nonimmune mast-cell activation. The mechanism of such augmentation was not known. Increased anxiety seems to be present in at least a subgroup of patients with ASDs, who may also be more prone to stress [10]. A comparison of 34 adults with autism and 20 controls, matched for age, gender, and intellectual ability, found that patients with ASDs were three times as anxious as controls, and were significantly less able to cope with stress [23]. Acute stress can activate brain mast cells, an effect abolished by pretreatment with polyclonal antiserum to CRH [24]. Subsequently, CRH was reported to activate brain mast cells and increase blood-brain barrier permeability in rodents [25,26], particularly in brain areas containing mast cells [27]. The direct effect of CRH was documented by intradermal administration leading to increased vascular permeability in rodents and humans, through activation of CRHR-1 [28].
We also found that sonicated mitochondrial components at (10 microgram/μl) stimulates VEGF release, which also augments allergic stimulation of VEGF release from human mast cells. At the present, we are not sure which mitochondrial components are responsible for VEGF release. They may include ATP, mtDNA, or formyl peptides found in mitochondria. VEGF is also known to stimulate mitochondrial biogenesis [29], suggesting a possible paracrine effect on secreted VEGF on the mitochondria of neighboring cells.
Several studies have reported mitochondrial dysfunction in autism [30], which may involved a subset of children with autism [31,32]. Mitochondria are the primary energy-generating organelles in eukaryotic cells, and they participate in multiple intracellular processes, including calcium buffering [33]. However, mitochondria were originally bacteria that became symbiotic with eukaryotic cells, and are typically prevented from being released extracellularly by autophagy [34]. We previously found increased extracellular mtDNA in the serum of young children with autism [15]. The present results indicate that extracellular mitochondria components can augment allergic mast-cell stimulation. This action may be in addition to any direct effect that mitochondrial components may have on the immune system. For instance, damageassociated mitochondrial pattern are able to activate Tolllike receptor 9 on human peripheral polymorphonuclear leukocytes, leading to release of interleukin-8 [35].
Given that ASDs has been associated with brain inflammation and oxidative stress [1,3,36], we investigated the effect of the flavone luteolin, which has anti-inflammatory and anti-oxidant properties [37]. We found that luteolin 100 μmol/l was able to inhibit the augmenting effect at mitochondria on allergic human mast-cell activation. We used this concentration because it had been previously shown to cause maximal inhibition of mast cells and mast-cell-dependent stimulation of activated T cells [16]. Luteolin also blocks methyl mercury-induced VEGF release from human mast cells [17]. Myricetin, the structural analog of luteolin, can also inhibit mast-cell activation [38], and methyl mercury-induced mitochondrial dysfunction [39]. Luteolin also blocks activated peripheral blood mononuclear cells from patients with the inflammatory brain disease multiple sclerosis [40]. A new luteolin-containing dietary supplement was recently shown to have significant benefit in children with ASDs [41]. Luteolin may therefore be useful for the treatment of brain inflammation [40,42].
Conclusion
Augmentation of allergic and mitochondria-stimulated mast-cell activation by CRH secreted by stress may explain at least some of the symptoms of patients with Figure 2 Mitochondria augment VEGF release from IgE/anti-IgE-stimulated human mast cells, and inhibition by luteolin. (A) VEGF secretion from hCBMCs was measured after pretreatment with IgE 1 microgram/μl for 2 hours and then incubating with mitochondria (0.1 and 10 microgram/μl) and anti-IgE (10 microgram/μl) for 24 hours. Pretreatment with luteolin 100 μmol/l for 30 minutes completely inhibited VEGF release and dropped it even below basal levels. For all experiments, n = 5; *p < 0.05, **p < 0.01, ***p < 0.001 compared with control.
Competing interest
The authors declared that they have no competing interest.
==
Domain: Biology Medicine
|
[{"added": "2016-05-12T22:15:10.714Z", "created": "2012-05-04T00:00:00.000", "id": "14501467", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://jneuroinflammation.biomedcentral.com/track/pdf/10.1186/1742-2094-9-85", "pdf_hash": "a1a935d802b92cc0749c68d99946f6d3fee7d213", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:34858", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "a279a5b21c205ec176b82ccc2c8320f5de3848ca", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Targeted Deletion of Kynurenine 3-Monooxygenase in Mice
Background: Kynurenine 3-monooxygenase (KMO) is hypothesized to play a pivotal role in regulating tryptophan metabolism in health and disease. Results: Mice that were generated lacking KMO have alterations in the levels of several tryptophan metabolites. Conclusion: KMO is a critical regulator of tryptophan metabolism. Significance: KMO knock-out mice will be a useful research tool to dissect the biological and pathophysiological roles of tryptophan metabolism. Kynurenine 3-monooxygenase (KMO), a pivotal enzyme in the kynurenine pathway (KP) of tryptophan degradation, has been suggested to play a major role in physiological and pathological events involving bioactive KP metabolites. To explore this role in greater detail, we generated mice with a targeted genetic disruption of Kmo and present here the first biochemical and neurochemical characterization of these mutant animals. Kmo−/− mice lacked KMO activity but showed no obvious abnormalities in the activity of four additional KP enzymes tested. As expected, Kmo−/− mice showed substantial reductions in the levels of its enzymatic product, 3-hydroxykynurenine, in liver, brain, and plasma. Compared with wild-type animals, the levels of the downstream metabolite quinolinic acid were also greatly decreased in liver and plasma of the mutant mice but surprisingly were only slightly reduced (by ∼20%) in the brain. The levels of three other KP metabolites: kynurenine, kynurenic acid, and anthranilic acid, were substantially, but differentially, elevated in the liver, brain, and plasma of Kmo−/− mice, whereas the liver and brain content of the major end product of the enzymatic cascade, NAD+, did not differ between Kmo−/− and wild-type animals. When assessed by in vivo microdialysis, extracellular kynurenic acid levels were found to be significantly elevated in the brains of Kmo−/− mice. Taken together, these results provide further evidence that KMO plays a key regulatory role in the KP and indicate that Kmo−/− mice will be useful for studying tissue-specific functions of individual KP metabolites in health and disease.
Under physiological conditions, the kynurenine pathway (KP) 3 is responsible for Ͼ95% of tryptophan degradation in mammals, ultimately leading to the formation of NAD ϩ (1,2). Located in a pivotal position in the KP, kynurenine 3-monooxygenase (KMO) is a FAD-dependent enzyme that catalyzes the 3-hydroxylation of L-kynurenine ("kynurenine") in the presence of NADPH and molecular oxygen. KMO localizes to the outer membrane of mitochondria and is highly expressed in peripheral tissues, including liver and kidney, and in phagocytes such as macrophages and monocytes (3,4). In the central nervous system, KMO is expressed predominantly in microglial cells (5,6). Notably, cytokines stimulate KMO activity both in the periphery and in the brain when the immune system is activated (7,8).
Because of its central location in the metabolic cascade, KMO controls the synthesis of several bioactive KP metabolites, including 3-hydroxykynurenine (3-HK), quinolinic acid (QUIN), and kynurenic acid (KYNA), as well as anthranilic acid (Fig. 1). Kynurenine, the substrate of KMO, itself promotes tumor formation (9), regulates vascular tone, and generates regulatory T cells by activating the aryl hydrocarbon receptor and/or adenylate and guanylate cyclase pathways (10 -13). 3-HK, the primary enzymatic product of KMO, is a potent free radical generator that can trigger neuronal apoptosis (14 -16). Similar features have been described for the next KP metabolite in the major branch of the pathway, 3-hydroxyanthranilic acid (17). Its downstream product, QUIN, can also generate reactive free radicals (18), but is best known for its ability to produce excitotoxic lesions in the central nervous system by activating the NMDA subtype of glutamate receptors (1,19,20) and as a bioprecursor of the ubiquitous coenzyme NAD ϩ .
Selective pharmacological inhibition of KMO leads to the increased formation of kynurenic acid (KYNA) in a dead end side arm of the KP (Fig. 1). At physiological concentrations, this metabolite has several biological targets, acting as an antagonist of the ␣7 nicotinic acetylcholine receptor and the glycine coagonist site of the NMDA receptor (21,22) and as an agonist of the G protein-coupled receptor GPR35 (23) and the aryl hydrocarbon receptor (24). Together with its capacity to scavenge free radicals (25), these actions account for the neuromodulatory and neuroprotective properties of KYNA (20,26,27) and for its ability to affect cytokine expression and function (24,28).
KMO inhibition is increasingly recognized as a useful experimental approach to unravel the characteristics of KP metabolism in brain and periphery. This, as well as the realization that KP manipulation by specific KMO inhibitors may have therapeutic utility in human maladies ranging from neurodegenerative disorders and major depression to immunological diseases and cancer (1, 20, 29 -32), has not only prompted the development of several potent enzyme inhibitors (33-36) but recently also led to the first reported crystal structure of KMO (37).
To expand the experimental armamentarium for the study of KP biology in mammals, we produced mice with a targeted deletion of the KMO gene. We describe here the generation of Kmo Ϫ/Ϫ mice and present an analysis of KP metabolites and enzymes in brain, liver, and plasma of these animals. Our study revealed novel characteristics of mammalian KP metabolism, indicating that the newly generated mutant mice will serve as a useful research tool for dissecting the proposed roles of bioactive metabolites in physiology and pathology.
EXPERIMENTAL PROCEDURES
Generation of Kmo Ϫ/Ϫ Mice-The Kmo conditional targeting construct was generated using the 4517D vector, kindly provided by Dr. Richard Palmiter (University of Washington). This vector contains a Pgk:neomycin resistance (NeoR) cassette for positive selection, which is flanked by FRT sites to allow FLP recombinase-mediated excision following selection of homologous recombinants (38). The vector also contains a 5Ј diphtheria toxin cassette and a 3Ј HSV:thymidine kinase cassette, both for negative selection. The targeting construct was designed such that exon 5 of Kmo is flanked by loxP sites and is thus targeted for deletion. A C57Bl/6J genomic DNA fragment containing the entire Kmo genomic region was obtained in the pBACe3.6 Bacterial Artificial Chromosome (BAC) vector (BACPAC Resource Center, Children's Hospital Oakland Research Institute, Oakland, CA). The right arm of the targeting vector was derived from a 5.7-kb ApaI fragment from the Kmo BAC that was shotgun cloned into the pKS-Bluescript plasmid (Stratagene, Santa Clara, CA). A 1.9-kb KpnI-PacI fragment from this construct was further subcloned into the PmeI site of 4517D downstream of the NeoR cassette, generating the 4517D (1.9-kb KpnI-PacI) construct. The left arm of the construct was derived from a 5.2-kb KpnI fragment from the above BAC, which was shotgun cloned into pKS-Bluescript. A double-stranded oligonucleotide encoding both a loxP site and a HpaI site was cloned into the StuI site of this 5.2-kb KpnI fragment, thereby generating a HpaI site and destroying the StuI site. The 5.2-kb KpnI loxP fragment was cloned into the 4517D (1.9-kb KpnI-PacI) construct upstream of the NeoR cassette using Acc651 (a neoschizomer of KpnI), yielding the final targeting construct.
The Kmo targeting construct was linearized at the 5Ј end with NotI and electroporated into a mouse ES cell line, derived from the C57Bl/6J strain of mice. NeoR-positive ES cell clones were selected, and 96 were screened by PCR for correct integration on the 3Ј end of the clone. Of the clones, 28 of 96 were positive, and they were screened by Southern blotting using a probe external to the targeting construct to confirm correct integration of both ends of the construct, yielding seven positive clones. Positive clones were injected into CBA inbred strain-derived murine blastocysts to produce chimeric founders, which were identified by coat color and mated to C57Bl/6J wild-type breeders to confirm germ line transmission, which was ascertained by coat color and PCR genotyping of the NeoR cassette. This founder line was then bred to mice carrying FLPrecombinase (39) to excise the NeoR cassette. This transgenic line of mice was designated as Kmo flox(ex)5-NeoR⌬ and bred to -actin Cre-C57Bl/6J (40) mice to generate KMO knock-out FIGURE 1. Major metabolites of the kynurenine pathway of tryptophan degradation. KMO, located in a pivotal position of the pathway, is responsible for the conversion of kynurenine to neurotoxic metabolites (such as 3-hydroxykynurenine and quinolinic acid) in the main branch of the enzymatic cascade. However, kynurenine can also be enzymatically converted to anthranilic acid or to the neuroprotective metabolite kynurenic acid. The kynurenine pathway normally accounts for Ͼ95% of tryptophan metabolism in mammals.
mice (Kmo Ϫ/Ϫ ), which were subsequently backcrossed to the FVB/N background for at least six generations, using the Speed Congenics Service (The Jackson Laboratory) to identify the optimal breeders.
Detection of KMO by Immunoblotting-Mice were perfused with 0.9% NaCl, and the right hemi-brain was fixed with 4% paraformaldehyde (prechilled to 4°C) for 48 h and then reduced to 2% paraformaldehyde for storage at 4°C. One hemibrain, as well as a portion of the liver, was snap frozen after saline perfusion and stored at Ϫ80°C until use. The tissues were then defrosted on ice and homogenized with 400 l of ice-cold radioimmunoprecipitation assay buffer (150 mM NaCl, 1% Igepal CA-630, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris-HCl, pH 8.0, 1 mM -mercaptoethanol and proteinase inhibitor mixture tablet) in a prechilled 1.5-ml microcentrifuge tube with a matching pestle until no visible tissue chunks remained. Two hundred ml of ice-cold radioimmunoprecipitation assay buffer were then added to the homogenate. After rehomogenization (30 s), the samples were kept on ice for 30 min and then centrifuged (Ͼ11,000 ϫ g) for 90 min at 4°C. The supernatants were collected in aliquots and stored at Ϫ80°C for future use. Protein concentration was determined using the Bradford assay (Bio-Rad). Samples containing 25 and 100 g of total liver and brain protein, respectively, were mixed with 8% SDS loading buffer and kept at room temperature for 5-10 min. Samples were then loaded on a 12% Tris-HEPES protein polyacrylamide gel and run at 100 V for ϳ55 min. After separation, the protein was transferred to a nitrocellulose membrane, which was blocked with 5% nonfat milk in 0.1% TBST (1ϫ TBS plus 0.1% Tween 20). The membrane was probed with KMO antibody (CHDI-90000271; 21 Century Biochemicals; 1:1,000 for brain, 1:5,000 for liver), and HRP-conjugated anti-rabbit secondary antibody (1:5,000). Protein was detected by chemiluminescence (ECL kit; GE Healthcare) according to the manufacturer's instructions.
Kynureninase (EC 3.7.1.3) Using 3-HK as the Substrate-The original tissue homogenate was diluted 1:40 (brain) or 1:4,000 (liver) (v/v) in 5 mM Tris-HCl buffer (pH 8.4) containing 10 mM 2-mercaptoethanol and 50 M pyridoxal-5Ј-phosphate. 100 l of the preparation were then incubated for 2 h at 37°C in a solution containing 90 mM Tris-HCl buffer (pH 8.4) and 4 M DL-3-HK, in a total volume of 200 l. The reaction was stopped by the addition of 50 l of 6% perchloric acid. To obtain blanks, tissue homogenate was added at the end of the incubation, i.e., immediately prior to the denaturing acid. After removing the precipitate by centrifugation (16,000 ϫ g, 15 min), 25 l of the resulting supernatant were applied to a 5-m C 18 reverse phase HPLC column (Adsorbosil; 150 ϫ 4.6 mm; Grace, Deerfield, IL) using a mobile phase containing 100 mM sodium acetate (pH 5.8) and 1% acetonitrile at a flow rate of 1.0 ml/min. In the eluate, the reaction product, 3-hydroxyanthranilic acid, was detected fluorimetrically (excitation wavelength, 322 nm; emission wavelength, 414 nm; S200 fluorescence detector; PerkinElmer Life Sciences). The retention time of 3-hydroxyanthranilic acid was ϳ4 min.
Kynureninase (EC 3.7.1.3) Using Kynurenine as the Substrate-Essentially identical incubation conditions were used to determine enzyme activity with kynurenine as the substrate. In this case, the original tissue homogenate was either used undiluted (brain) or diluted 1:400 (liver) (v/v) in the same 5 mM Tris-HCl buffer described above, and 80 l of the preparation were incubated for 2 h in the presence of 6 M kynurenine. Following termination of the reaction, 25 l of the supernatant were subjected to HPLC (as above), and the enzymatic product, anthranilic acid, was detected fluorimetrically in the eluate (excitation wavelength, 340 nm; emission wavelength, 410 nm; S200 fluo-rescence detector; PerkinElmer Life Sciences). The retention time of anthranilic acid was ϳ7 min.
Quinolinic Acid Phosphoribosyltransferase (QPRT; EC 2.4.2.19)-The original tissue homogenate was either used undiluted (brain) or diluted 1:120 (liver) (v/v). Forty l of the preparation were incubated for 2 h at 37°C in a solution containing 50 mM potassium phosphate buffer (pH 6.5), 1 mM MgCl 2 , 1 mM phosphoribosylpyrophosphate, and 20 nM [ 3 H]QUIN (30 nCi) in a total volume of 0.5 ml. Blanks were obtained by including the QPRT inhibitor phthalic acid (500 M) in the incubation solution. The reaction was terminated by placing the tubes on ice, and particulate matter was separated by centrifugation (16,000 ϫ g, 10 min). Newly formed [ 3 H]nicotinic acid mononucleotide was recovered from a Dowex AG 1 ϫ 8 anion exchange column and quantified by liquid scintillation spectrometry (43).
For the determination of KP metabolites, samples were thawed and processed as follows.
KYNA-Plasma was diluted (1:10, v/v), and tissues were homogenized (brain, 1:5; liver, 1:50; w/v) in ultrapure water. Twenty-five l of 6% perchloric acid were added to 100 l of the samples. After thorough mixing, the precipitated proteins were removed by centrifugation (16,000 ϫ g, 15 min). Twenty l of the resulting supernatant were subjected to HPLC analysis, and KYNA was assessed as described above (see "Kynurenine Aminotransferase II").
3-HK-Plasma was diluted (1:2, v/v), and tissues (brain or liver) were homogenized (1:5, w/v) in ultrapure water. Twentyfive l of 6% perchloric acid were added to 100 l of the samples. After thorough mixing, the precipitated proteins were removed by centrifugation (16,000 ϫ g, 15 min). Twenty l of the resulting supernatant were subjected to HPLC analysis, and 3-HK was assessed as described above (see "Kynurenine 3-Monooxygenase").
Anthranilic Acid-Anthranilic acid was measured according to the method of Cannazza et al. (44). Plasma was diluted (1:10, v/v), and tissues were homogenized (brain, 1:20; liver, 1:40; w/v) in ultrapure water. Twenty-five l of 6% perchloric acid were added to 100 l of the samples. After thorough mixing, the precipitated proteins were removed by centrifugation (16,000 ϫ g, 15 min). Thirty l of the resulting supernatant were subjected to HPLC analysis. Anthranilic acid was isocratically eluted from a 5-m C 18 reverse phase column (YMC-Pack Pro C18; 250 ϫ 4.6 mm; YMC) using a mobile phase containing 20 mM sodium acetate and 10% methanol (pH adjusted to 5.5 with glacial acetic acid) and was detected fluorimetrically (excitation wavelength, 316 nm; emission wavelength, 420 nm; S200 fluorescence detector; PerkinElmer Life Sciences). The retention time of anthranilic acid was ϳ18 min.
Tryptophan, Kynurenine, and QUIN-Tryptophan, kynurenine, and QUIN levels in brain and liver, and QUIN in plasma, were quantified by GC/MS. To this end, plasma was diluted (1:10, v/v), and tissues were homogenized (1:20, w/v) in an aqueous solution containing 0.1% ascorbic acid. Fifty l of a solution containing internal standards (500 nM [ 2 H 5 ]L-tryptophan, 10 M [ 2 H 6 ]L-kynurenine, and 50 nM [ 2 H 3 ]QUIN) were added to 50 l of the tissue preparation, and proteins were precipitated with 50 l of acetone. After centrifugation (13,700 ϫ g, 5 min), 50 l of a methanol:chloroform mixture (20:50, v/v) were added to the supernatant, and the samples were centrifuged (13,700 ϫ g, 10 min). The upper layer was added to a glass tube and evaporated to dryness (90 min). The samples were then reacted with 120 l of 2,2,3,3,3-pentafluoro-1-propanol and 130 l of pentafluoropropionic anhydride at 75°C for 30 min, dried down again, and taken up in 50 l of ethyl acetate. One l was then injected into the GC. GC/MS analysis was carried out with a 7890A GC coupled to a 7000 MS/MS (Agilent Technologies, Santa Clara, CA), using electron capture negative chemical ionization (45).
For kynurenine measurement in plasma, 50 l of 6% perchloric acid were thoroughly mixed with 100 l of plasma, and the precipitated proteins were removed by centrifugation (16,000 ϫ g, 15 min). Twenty l of the supernatant were applied to a 3-m C 18 reverse phase column (HR-80; 80 ϫ 4.6 mm; ESA), and kynurenine was isocratically eluted using a mobile phase containing 250 mM zinc acetate, 50 mM sodium acetate, and 3% acetonitrile (pH 6.2) at a flow rate of 1 ml/min. Kynurenine was detected fluorimetrically (excitation wavelength, 365 nm; emission wavelength, 480 nm; S200 fluorescence detector, PerkinElmer Life Sciences). The retention time of kynurenine under these conditions was ϳ6 min. DECEMBER 20, 2013 • VOLUME 288 • NUMBER 51
JOURNAL OF BIOLOGICAL CHEMISTRY 36557
NAD ϩ -The levels of NAD ϩ were quantified using the EnzyChrom TM NAD ϩ /NADH assay kit according to manufacturer's instructions. Briefly, 20 mg of tissue (brain or liver) were washed with cold PBS and homogenized in 100 l of NAD ϩ extraction buffer. The samples were then heated at 60°C for 5 min, and 20 l of the assay buffer and 100 l of NADH extraction buffer were added, thoroughly mixed, and centrifuged (16,000 ϫ g, 5 min). The resulting supernatant was diluted 1:5 with ultrapure water. Forty l of the diluted supernatant were then mixed with 90 l of the working reagent mixture and incubated at room temperature for 15 min. The absorbance was read both at time 0 and 15 min on a plate reader, with the reference and measuring filters set at 450 and 570 nm, respectively.
Microdialysis-The mice were anesthetized with chloral hydrate (360 mg/kg, intraperitoneal) and mounted in a David Kopf stereotaxic frame (Tujunga, CA). A guide cannula (outer diameter, 0.65 mm) was positioned over the striatum (anteroposterior, 0.5 mm anterior to bregma; lateral, 1.2 mm from midline; and ventral, 1.1 mm below the skull) and secured to the skull with an anchor screw and acrylic dental cement. A concentric microdialysis probe (membrane length, 1 mm; SciPro, Sanborn, NY) was then inserted through the guide cannula. The probe was connected to a microinfusion pump set to a speed of 1 l/min and perfused with Ringer solution containing 144 mM NaCl, 4.8 mM KCl, 1.2 mM MgSO 4 , 1.7 mM CaCl 2 (pH 6.7). Samples were collected every 60 min for 8 h. The concentration of KYNA was determined in the microdialysate as described (46), and data are reported without correction for recovery from the microdialysis probe.
Generation and Validation of Kmo
Ϫ/Ϫ Mice-We employed ES cell-based transgenesis to generate transgenic mice with a targeted disruption of the Kmo gene. The Kmo targeting construct was designed with exon 5 of the Kmo locus flanked by loxP sites (Fig. 2A), such that Cre-mediated recombination is predicted to delete exon 5 and generate a putative null allele-a nonsense frameshift mutation with a premature stop in exon 7, which is expected to lead to nonsense-mediated degradation of the mutant mRNA. In combination with Cre recombinase, the Cre/loxP system allows for tissue-specific and temporal deletion of a candidate DNA sequence. Homologous recombination in murine ES cells was used to generate a putative conditional loxP allele of Kmo (Kmo flox(ex5) ) as described (38). Correct targeting of the construct was confirmed by PCR and Southern hybridization (Fig. 2, B and C).
Mice carrying the (Kmo flox(ex5) ) allele were hypomorphic for KMO activity (data not shown), likely because of the presence of the neomycin resistance cassette (NeoR) in intronic sequences (38). A true Kmo conditional allele (Kmo flox(ex)5-NeoR⌬ ) was generated by excision of the NeoR cassette, which is flanked by FRT recombination sites (see "Experimental Procedures"). Mice carrying the Kmo flox(ex)5-NeoR⌬ allele were mated to transgenic mice expressing Cre recombinase ubiquitously under the -actin promoter to generate progeny lacking KMO systemically and throughout all developmental stages (hereafter referred to as Kmo Ϫ/Ϫ mice). We confirmed correct targeting of the Kmo locus by analysis of KMO protein levels in the brain and liver by immunoblotting, finding that Kmo Ϫ/Ϫ mice did not express KMO protein in either the brain or the liver (Fig. 2D). Notably, elimination of KMO activity in Kmo Ϫ/Ϫ mice did not cause embryonic lethality, did not lead to overt phenotypic abnormalities during postnatal development or in adulthood, and did not influence breeding or normal Mendelian inheritance (data not shown).
Targeted Deletion of Kynurenine 3-Monooxygenase in Mice
KP Enzymes and Metabolites in the Liver of Kmo Ϫ/Ϫ Mice-We next determined KMO activity in the liver of ϳ2-monthold mice and observed that enzyme activity was eliminated in the Kmo Ϫ/Ϫ mice (Fig. 3A). This indicated that KMO is the sole enzyme responsible for catalyzing the production of 3-HK from kynurenine in the mouse liver. Aside from KMO, none of the KP enzymes tested (KAT II, kynureninase (using either kynurenine or 3-HK as substrate), 3-HAO, and QPRT) displayed significant changes in activity in Kmo Ϫ/Ϫ mice compared with wild-type (Kmo ϩ/ϩ ) mice of the same age (Fig. 3, B-F). As expected, KMO activity was ϳ50% of wild-type levels in brain and liver tissues of Kmo ϩ/Ϫ mice (data not shown). A detailed analysis of these mice will be presented elsewhere. 4 The hepatic levels of tryptophan tended to be lower in Kmo Ϫ/Ϫ mice than in Kmo ϩ/ϩ littermate controls, but the difference did not reach statistical significance (Fig. 4A). In contrast, levels of kynurenine, the substrate of KMO, were significantly increased (ϳ5-fold) in Kmo Ϫ/Ϫ mice (Fig. 4B), whereas levels of the KMO product (3-HK) and its downstream metabolite (QUIN) were reduced to 42 and 3%, respectively, of wildtype levels (Fig. 4, C and F). Thus, despite the elimination of KMO, 3-HK was still detected in the liver of the mutant mice. Hepatic NAD ϩ levels were slightly, but not significantly, lower in Kmo Ϫ/Ϫ compared with Kmo ϩ/ϩ mice (Fig. 4G). In line with the elevated kynurenine content, the levels of KYNA (the product of the irreversible transamination of kynurenine by KAT) and anthranilic acid (the product of kynureninase) were significantly increased (ϳ69-and ϳ5-fold, respectively) in the liver of Kmo Ϫ/Ϫ mice (Fig. 4, D and E).
KP Enzymes and Metabolites in the Brains of Kmo Ϫ/Ϫ Mice-Similar to the liver, brain KMO activity was abolished in the Kmo Ϫ/Ϫ mutants (Fig. 6A), whereas none of the other KP enzyme activities (KAT II, kynureninase, 3-HAO, and QPRT) differed between genotypes (Fig. 6, B-F). However, examination of KP metabolites in the brain revealed remarkable differences from the periphery. Thus, cerebral 3-HK levels in the knock-out animals were very low (ϳ10% of wild-type controls; Fig. 7C), whereas QUIN levels were only slightly reduced (to ϳ80% of Kmo ϩ/ϩ controls) (Fig. 7F). Moreover, although brain KYNA levels were elevated (ϳ12-fold) in Kmo Ϫ/Ϫ mice (Fig. 7D), this increase was far less pronounced than in the periphery (cf. Figs. 4D and 5C). Similar to the results obtained in the liver, the brain levels of tryptophan were modestly decreased in Kmo Ϫ/Ϫ mice, kynurenine and anthranilic acid levels were significantly elevated (ϳ2-and ϳ3-fold, respectively; Fig. 7, A, B, and E), and NAD ϩ levels were not significantly different from Kmo ϩ/ϩ controls (Fig. 7G).
Extracellular Levels of KYNA in the Striatum of Kmo Ϫ/Ϫ Mice-We next performed in vivo microdialysis in the striatum of awake, behaving mice to measure basal extracellular KYNA in Kmo Ϫ/Ϫ mice. As shown in Fig. 8, which illustrates basal levels averaged over an 8-h collection period, KYNA levels were significantly increased (ϳ6-fold) in the mutant animals compared with wild-type controls.
DISCUSSION
We describe here the generation and characterization of Kmo Ϫ/Ϫ mice, confirming that genetic disruption of the Kmo locus is sufficient to eliminate KMO detection by immunoblotting and to reduce enzymatic activity in brain and liver to undetectable levels. Although 3-HK was still clearly measurable in the mutant animals and our study therefore does not categorically rule out the existence of functional KMO isoforms, this indicates that a single KMO is largely responsible for the production of 3-HK from kynurenine in mice. Moreover, none of the other KP enzymes analyzed in liver and brain showed any significant changes in adult Kmo Ϫ/Ϫ mice. The mutant mice were viable and overtly healthy, demonstrating that the deleted KMO was not essential for embryonic and postnatal survival.
Tryptophan levels were marginally reduced in the brain, but not in the liver, of knock-out compared with wild-type mice, suggesting that the elimination of KMO has only a small effect on the activity of the upstream enzymes, indoleamine 2,3-dioxygenases and tryptophan 2,3-dioxygenase, which initiate catabolism along the KP (Fig. 1). However, further analysis of Kmo Ϫ/Ϫ mice revealed several biochemical features that sup-port the idea that KMO normally plays a critical role in the regulation of KP metabolism. Thus, the steady-state levels of kynurenine, the substrate of KMO, were substantially increased both in the periphery and in the brain of the mutant animals. It is likely that this, in turn, accounted for the fact that the levels of two primary degradation products of kynurenine, KYNA and anthranilic acid, were also significantly elevated in Kmo Ϫ/Ϫ mice. Notably, these effects did not involve changes in the activities of KAT II and kynureninase, the enzymes that catalyze the formation of KYNA and anthranilic acid, respectively, from kynurenine ( Fig. 1). Thus, the accumulation of KYNA and anthranilic acid in Kmo Ϫ/Ϫ mice further suggests that KATs and kynureninase are operating below saturation under physiologic conditions.
Although the levels of 3-HK, the product of KMO, were substantially reduced in the liver, serum, and brain of Kmo Ϫ/Ϫ mice, specific deletion of the enzyme did not totally eliminate 3-HK from either of the three tissues examined. The liver of mutant animals, in particular, contained considerable residual 3-HK (Fig. 4C), and additional studies will be required to identify both its source and its possible functional significance.
Targeted Deletion of Kynurenine 3-Monooxygenase in Mice
Interestingly, our study revealed substantial quantitative differences in KP metabolism between the periphery and the brain as a result of the KMO deletion. For example, the increase in KYNA levels in Kmo Ϫ/Ϫ mice was far more pronounced in the liver and even more so in the plasma, than in the brain, whereas the increases in the levels of anthranilic acid in the same animals were of comparable magnitude (ϳ3-5-fold) in periphery and brain. This phenomenon may be related to the fact that in the brain KAT II is contained in astrocytes, whereas KMO and kynureninase are localized in microglial cells (5,47). The cellular segregation of the two KP arms, which does not exist in the periphery, may therefore prevent the preferential conversion of kynurenine to KYNA in the brain of Kmo Ϫ/Ϫ mice.
The quantitatively similar increase in anthranilic acid levels in brain and periphery of mutant mice probably accounted for the fact that only a ϳ20% reduction in QUIN levels was observed in the brain of Kmo Ϫ/Ϫ compared with wild-type mice, whereas QUIN was practically eliminated in the liver and plasma of the same animals (cf. Figs. 4, 5, and 7). Thus, in stark contrast to the periphery, anthranilic acid is far superior to 3-HK as a bioprecursor of 3-hydroxyanthranilic acid in the brain (48) and should therefore be much better suited to sustain cerebral QUIN levels when 3-HK formation is compromised. Our results also may indicate that the anthranilic acid branch of KP metabolism is the minor pathway for kynurenine catabolism in all tissues. Interestingly, in brains of Kmo Ϫ/Ϫ mice, the anthranilic acid to QUIN conversion is likely the dominant pathway for QUIN formation and may represent a compensatory mechanism to form QUIN when KMO activity is compromised. More importantly, our results indicate surprisingly that pharmacological inhibition of KMO in the brain is unlikely to dramatically lower QUIN levels, suggesting that peripheral inhibition of KMO may be sufficient to confer neuroprotection (30). Unfortunately, 3-hydroxyanthranilic acid levels were below the limits of assay sensitivity in the present study and were therefore not available for testing this hypothesis directly. However, we examined whether the differential effects of KMO deletion on QUIN levels in brain and liver carried over to the important downstream metabolite NAD ϩ . No significant differences in NAD ϩ content were seen between wild-type and FIGURE 6. KP enzyme activities in brain. A, KMO activity is eliminated in cortical homogenates of Kmo Ϫ/Ϫ mice. B-F, compared with wild-type (Kmo ϩ/ϩ ) mice, the activities of KAT II, kynureninase (using either kynurenine or 3-HK as a substrate), 3-HAO, and QPRT are unchanged in Kmo Ϫ/Ϫ mice. The data are the means Ϯ S. E. (n ϭ 6 -8/group). Statistical analysis was performed using Student's t test. ***, p Ͻ 0.001 versus Kmo ϩ/ϩ . mutant animals in either tissue, supporting the notion that alternative mechanisms readily maintain normal levels of NAD ϩ , even when the KP is severely compromised by elimination of KMO (49 -51).
Work in a large number of in vivo and in vitro model systems has revealed that kynurenine itself, as well as several KP metabolites, acting through a remarkably broad array of biological mechanisms (28,52), normally operate as signaling molecules in physiological processes ranging from immune regulation (53) to cognitive functions (20). Although not investigated here, the newly generated Kmo Ϫ/Ϫ mice can therefore be expected to have abnormalities that are related to the impairments in KP metabolites described in the present study. Thus, in the periphery, the massive reduction in QUIN may compromise the function of the spleen and lymphoid tissue (54), and increased KYNA may adversely influence the function of endothelial cells (55). Functional consequences of fluctuations in KP metabolites are best documented in the brain, where decreased 3-HK and elevated KYNA levels are known to reduce neuronal vulnerability (16,20). Notably, increased brain KYNA levels also FIGURE 7. KP metabolites in brain. A, compared with wild-type (Kmo ϩ/ϩ ) controls, tryptophan levels are slightly reduced in the brain of Kmo Ϫ/Ϫ mice. B, D, and E, the levels of kynurenine (B), KYNA (D), and anthranilic acid (E) are significantly elevated in Kmo Ϫ/Ϫ mice. C, the levels of 3-HK are significantly reduced in the mutant animals. F, Kmo Ϫ/Ϫ mice exhibit a moderate, but significant, reduction in QUIN levels compared with Kmo ϩ/ϩ controls. G, no significant differences between the levels of NAD ϩ in Kmo ϩ/ϩ and Kmo Ϫ/Ϫ genotypes. The data are the means Ϯ S. E. (n ϭ 6 -7/group). Statistical analysis was performed using Student's t test. *, p Ͻ 0.05; ***, p Ͻ 0.001 versus Kmo ϩ/ϩ . DECEMBER 20, 2013 • VOLUME 288 • NUMBER 51 cause a variety of distinct cognitive impairments (56 -59). In a preliminary study evaluating such possible abnormalities in Kmo Ϫ/Ϫ mice, which may be a direct consequence of the increase in extracellular KYNA levels in the brain (Fig. 8), we indeed observed a pronounced deficit in contextual memory in the mutant animals (60).
Targeted Deletion of Kynurenine 3-Monooxygenase in Mice
Related to their role(s) in physiology, peripheral and central KP metabolites have recently received considerable attention as potential agents in human diseases ranging from cancer (61) to intestinal syndromes (62) and disorders of the endocrine (63), immune (64,65), and central nervous systems (20, 66 -69). Causal connections have been postulated either because genetic or biochemical KP abnormalities were found in specimens obtained from patients, as in schizophrenia (70 -76), or because studies in animals had predicted etiologically significant links. In some instances, such as the neurodegenerative disorder Huntington disease, the case for causality is strengthened by converging evidence from both human tissues and relevant animal models (77,78).
These insights and hypotheses, in turn, stimulated efforts to target individual KP enzymes pharmacologically to normalize functional impairments caused by KP metabolites. A rich armamentarium of selective and potent enzyme inhibitors is now available, and these compounds are not only used as experimental tools but also hold considerable promise as therapeutic agents in a number of human diseases (see Refs. 1, 20, 29, and 67 for recent reviews). Because of its pivotal position in the KP, KMO has attracted special attention in this regard. Mostly tested in the neurosciences so far, specific KMO inhibitors, including the prototypical Ro 61-8048 (35), indeed have significant biological activity in vivo. These studies have revealed, for example, anticonvulsive, neuroprotective, and anti-dyskinetic effects of KMO inhibition (30,66,79,80) and led to the acceptance of KMO as a bona fide drug target (37). Notably, genetic elimination of KMO suppresses toxicity in yeast and Drosophila models of Huntington disease (32,81), suggesting that Kmo Ϫ/Ϫ mice, too, may show resistance to endogenous or external insults. Nonetheless, pharmacological inhibition of KMO might also lead to undesirable consequences on the nervous and immune systems (20) that will have to be evaluated carefully in animals prior to clinical studies in humans.
In summary, the newly produced Kmo Ϫ/Ϫ mice should be an attractive experimental tool for the further elucidation of the role of the KP in mammalian physiology and pathology. This role involves the bi-directional cross-talk between periphery and brain (20), so that gene deletion in individual tissues will be of significant interest. Because our Kmo Ϫ/Ϫ mice were generated using the Cre/loxP system, studies currently in progress therefore target deletion of the Kmo locus in specific organs to characterize the regulatory functions of the enzyme in distinct biological compartments.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T03:33:44.296Z", "created": "2013-11-04T00:00:00.000", "id": "6718565", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "http://www.jbc.org/content/288/51/36554.full.pdf", "pdf_hash": "1c2655e28a19041b90527d6dc2321efe83ee3451", "pdf_src": "Highwire", "provenance": "peS2o-0075.json.gz:35491", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "98758d655cf0a047e494bbb795c51cb75eb29a0f", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Imaging metabolic syndrome
Metabolic syndrome is a fast growing public health burden for almost all the developed countries and many developing nations. Despite intense efforts from both biomedical and clinical scientists, many fundamental questions regarding its aetiology and development remain unclear, partly due to the lack of suitable imaging technologies to visualize lipid composition and distribution, insulin secretion, β-cell mass and functions in vivo. Such technologies would not only impact on our understanding of the complexity of metabolic disorders such as obesity and diabetes, but also aid in their diagnosis, drug development and assessment of treatment efficacy. In this article we discuss and propose several strategies for visualization of physiological and pathological changes that affect pancreas and adipose tissue as a result of the development of metabolic diseases.
Molecular imaging
For centuries, physicians have been striving to observe the structures and functions of the human body without the need to cut it open or to perform a biopsy. The discovery of X-rays in 1895 laid the foundation for the major advances made much later in the 1970s, i.e. taking a picture of the human body by X-ray to give a detailed image. At the same time, magnetic resonance imaging (MRI, see Box 1) was discovered and developed for medical imaging, as was the use of radioisotopes and ultrasound. A modern radiology department uses all these techniques for the diagnosis and therapy evaluation of a range of diseases.
In parallel, life scientists wanted to look inside the living cells to see the architecture and makeup of life at the molecular level. The microscope in the first instance, followed by the use of light, electrons, X-rays and magnetic signals to observe molecular details of structure and function in isolated cells have all contributed to the enormous advances that have been made in understanding life. Several imaging tools, their resolution and applications in the biomedical sciences are depicted in Fig 1. In this genomic era, imaging the working products of the genome inside a living cell, in a model organism of human disease (e.g. mouse models) and in human patients is one of the most powerful approaches to advance our knowledge. In the field of molecular medicine, imaging is also a way to link pre-clinical biomedical research and clinical practice. Imaging biomarkers can provide surrogate endpoints for clinical trials or shortcuts to drug development.
By definition, molecular imaging is the non-invasive visualization in space and time of normal as well as abnormal cellular processes at a molecular or genetic level. It can be used to characterize and measure particular biological processes in living organisms. The term 'molecular imaging' is used in a variety of ways to describe (1) the imaging of endogenous molecules that are present in a living system; (2) the use of foreign targeted or activatable reporter agents that sense specific molecular targets or cellular processes; (3) the use of labelled or natural substrates to follow particular pathways and (4) the introduction of genes to express protein products that can be detected directly or indirectly.
In this article, rather than an extensive review of the literature, we present our own perspective of how an important clinical problem-the metabolic syndrome-can be tackled using molecular imaging tools. Our group includes a highly multidisciplinary team of engineers, physicists, chemists, biologists Metabolic syndrome is a fast growing public health burden for almost all the developed countries and many developing nations. Despite intense efforts from both biomedical and clinical scientists, many fundamental questions regarding its aetiology and development remain unclear, partly due to the lack of suitable imaging technologies to visualize lipid composition and distribution, insulin secretion, b-cell mass and functions in vivo. Such technologies would not only impact on our understanding of the complexity of metabolic disorders such as obesity and diabetes, but also aid in their diagnosis, drug development and assessment of treatment efficacy. In this article we discuss and propose several strategies for visualization of physiological and pathological changes that affect pancreas and adipose tissue as a result of the development of metabolic diseases. and medical researchers, working together as a team to bring individual ideas and technical knowledge. The problem we chose to address, metabolic syndrome, describes a group of metabolic abnormalities that raise the risk of cardiovascular disease and type 2 diabetes, including hyperinsulinaemia, dyslipidaemia, central obesity and hypertension (Reaven, 1993). Here we will consider obesity and type 2 diabetes in particular and will use these diseases as examples to illustrate how some of the outstanding questions in the area may be approached by using novel imaging techniques. We will discuss the limitations of existing technologies and examine some recent developments that might provide alternative ways of thinking in the field. It will become clear that such solutions require close collaboration between scientists in the engineering and physical sciences with biomedical and clinician scientists.
Imaging ectopic fat accumulation
Ectopic lipid deposition, i.e. lipid accumulation in tissues/ organs other than white adipose tissue (WAT), such as liver and muscle, is often associated with metabolic abnormalities, including insulin and leptin resistance (Muoio & Newgard, 2006). One of the questions in the metabolic disease field is how lipids are accumulated ectopically, and which particular lipid species cause the most severe damage leading to insulin and leptin resistance during diabetes and obesity development.
Arguably, magnetic resonance is probably the most suitable technology to study the role of lipids non-invasively in vivo (see Box 2). Although measurement of total fat and three-dimensional (3D) reconstruction and quantification of various fat depots can be done routinely and reliably with MRI aided with
Bridge the Gap
Weiping Han et al.
Glossary
Arterial spin labelling A functional magnetic resonance imaging method used in measuring blood flow.
b-cell
A type of endocrine cell in the pancreas that secretes insulin.
Cardiovascular disease
A class of diseases that involve the heart or blood vessels, e.g. stroke.
Contrast agents
A substance used to enhance the contrast of structures or fluids within the body in medical imaging.
Dynamic nuclear polarization (DNP)
Results from transferring spin polarization from electrons to nuclei, thereby aligning the nuclear spins to the extent that electrons are aligned. DNP is one of several techniques for hyperpolarization.
Dyslipidaemia
A condition with abnormal amount of lipids in the blood. Hyperlipidaemia, a form of dyslipidaemia, is often seen in people with hyperinsulinaemia.
Exocytosis
A process by which a cell directs the contents of secretory vesicles out of the cell membrane.
Gradient spin echo
A variety of MRI sequences that allow the measurement of diffusion coefficients.
Hyperinsulinaemia
A condition in which there are excess levels of circulating insulin in the blood. It is often found in people with type 2 diabetes.
Hypertension
A chronic medical condition in which the blood pressure is elevated. Persistent hypertension increases the risk for strokes, heart attacks and heart failure.
Insulin
A hormone released from pancreatic b-cells, which is central to regulating energy and glucose metabolism in the body.
Isotopomer analysis
It provides the possibility of distinguishing between molecules of the same compound on which only one atom contains an isotope and those with two or more atoms labelled by the same isotope; it allows the measurement of isotope isomer distributions to calculate the fluxes through a biochemical network.
Krebs cycle
A series of enzyme-catalysed chemical reactions of central importance in all living cells that use oxygen for cellular respiration.
Leptin
A hormone released from white adipose tissue, which is a key regulator of energy homeostasis. Defective leptin signalling is the leading biological basis for obesity.
Magnetic resonance spectroscopy
A specialized technique that allows detection of biochemical reactions non-invasively in vivo, e.g. measurement of ATP and phosphocreatine levels in phosphorus spectroscopy.
Obesity
A medical condition with excess body fat accumulation. Body mass index is used to define people as overweight (BMI between 25 and 30 kg/m 2 ) or obese (BMI ! 30 kg/m 2 ).
Pancreatic islets
The pancreatic islets contain the endocrine cells in the pancreas, including glucagon-secreting a-cells and insulin-secreting b-cells.
PET (positron emission tomography)
A nuclear medicine imaging technique which produces a 3D image or picture of functional processes in the body.
Radioisotope
An atom with an unstable nucleus, characterized by excess energy that may be imparted to a newly created radiation particle within the nucleus, or to an atomic electron.
1D MRS (magnetic resonance spectroscopy) (Liu et al, 2010;Springer et al, 2010), it remains a challenge to determine the extent of saturated and unsaturated lipids within a tissue compartment. In particular, detecting the saturated and unsaturated lipids of intramyocellular lipid (IMCL) and extra-myocellular lipid (EMCL) pools in skeletal muscle is a challenge because of the small spectral chemical shift between the olefinic protons of the two lipid pools (Fig 2A). The degree of unsaturation within the IMCL and EMCL pools is of significant clinical importance (Boesch, 2007), as the effects of fatty acid on Bridge the Gap Approaches to image metabolism BOX 1: Magnetic resonance imaging and magnetic resonance spectroscopy MRI/MRS is based on a physical phenomenon called nuclear magnetic resonance (NMR). A positively charged hydrogen nucleus (proton) exposed to a magnetic field can absorb energy from the magnetic pulse and radiate this energy out at a particular frequency (the resonance frequency (RF), $64 MHz at 1.5 T). As the perturbed proton returns to its equilibrium state (relaxation), the RF wave can be detected by a radiofrequency coil and the relaxation process can be described by two-time constants-longitudinal relaxation time (T 1 ) and transverse relaxation time (T 2 ). T 1 and T 2 reflect the micro-or macro-environment of the proton such as diffusion, magnetic susceptibility, temperature, binding to proteins and flow. Therefore, pathologies that change these properties, e.g. haemorrhages, can be detected. The resonance frequencies of protons in different chemicals are characteristically 'shifted' by the chemical environments of the protons and therefore spectral analysis of the NMR signal enables identification and quantification of metabolites in tissues and in vivo. Besides 1 H, nuclei of atoms, including 13 C, 23 Na and 31 P, also have magnetic properties. They can be used to study a range of chemicals in living systems that contain these atoms. Figure 1. Imaging modalities, their spatial resolution and applications. Bioimaging encompasses a range of imaging modalities that provide spatial and functional analyses from molecules in the angstrom level to human brain and heart in the centimetre level. NMR, nuclear magnetic resonance; SERS, surface-enhanced Raman scattering; CT, computed tomography; MRI, magnetic resonance imaging; PET, positron emission tomography. metabolic signalling and energy metabolism are modulated by degree of unsaturation (Vessby et al, 2002). To overcome the limitations of inadequate spectral dispersion in 1D MRS measurements, one possible solution is to use spatially resolved 2D MRS techniques ( Fig 2B). A recent study demonstrated the 2D approach to estimate the degree of unsaturation within IMCL pool, and found a direct link between the extent of obesity and unsaturated IMCL, lending further support to the current knowledge of dysregulated lipid metabolism in obesity (Velan et al, 2008).
Bridge the Gap
Weiping Han et al.
BOX 2:
The advantages of magnetic resonance to study the role of lipids non-invasively in vivo MR is well suited for studying lipids non-invasively in humans or animal models of diabetes or obesity. First, the signals from fat and water in tissue are relatively easy to separate in MR images, and it is routine to use MRI in clinical studies to evaluate fat distribution. Second, 1 H magnetic resonance spectroscopy (MRS) can be used to measure the amount of 'liquid lipid' in triglycerides, i.e. lipids that have fast movement and not lipids that compose structures like membranes. Third, signals in the MR spectra from saturated and unsaturated fatty acid chains can be distinguished, providing a route for the molecular characterization of in vivo fat. Fourth, intracellular and extracellular lipid signals in skeletal muscle and possibly in other organs can be separated by 1 H MRS (Boesch et al, 1997). All these approaches have been used in humans and rodent models for obesity and diabetes, and thus provide a powerful link between pre-clinical investigations of animal models and clinical studies of patients. For example, the signal from IMCL correlates with insulin resistance in both human (Meex et al, 2010) and animal models such as the Zucker rat (Korach-Andre et al, 2005).
Imaging of insulin secretion and insulin granule exocytosis in model organisms
Understanding insulin secretion and its regulation mechanisms in vivo as well as being able to measure b-cell mass over the disease progression is instrumental for drug target screening, validation and for evaluation of therapeutic strategies. Since the discovery of insulin almost a century ago, research efforts have been focussed on understanding insulin secretion mechanisms, and more recently, on the molecular regulation of insulin granule exocytosis. These studies established cellular mechanisms governing insulin secretion (Fig 3), but many of its details have not been tested in in vivo settings, due to the lack of suitable tools to measure and quantify insulin secretion at high spatial and temporal resolutions. Traditionally, insulin secretion is measured by enzyme-linked immunosorbent assay (ELISA) or radioimmunoassay (RIA). These methods suffer from poor temporal resolution, complete lack of spatial resolution and delayed results. In model organisms, optical imaging of genetically introduced exocytosis markers may provide an alternative, with high spatial and temporal resolutions, along with instantaneous quantitative information. Figure 4 depicts one strategy for visually monitoring insulin granule movement and exocytosis, and for providing instantaneous quantification of insulin secretion. The optical sensor for exocytosis can be genetically introduced into mouse b-cells under the control of mouse insulin I promoter (Lu et al, 2009) or other b-cell specific promoters. Intravital fluorescent microscopic imaging may be used to examine insulin granule exocytosis systemically by tail vein injections of reagents that impact insulin secretion. Such a strategy may also be combined with surface-enhanced Raman spectroscopic (SERS) detection of insulin and glucose as detailed in Fig 5. In this SERS-based approach, a SERS tag (e.g. an organic molecule immobilized on a gold nanoparticle) serves to label and track an analyte such as insulin. Although label-free detection of analytes on a nanoparticle surface is theoretically possible, this has not been demonstrated in a published study. Since SERS tags use molecules tethered to the surface of gold or silver nanoparticles, it has limited or no toxicity (Faulds et al, 2004), particularly in gold-based systems. In some cases, the whole SERS active substrate could be implanted in mouse to obtain quantitative sensing of glucose (Stuart et al, 2006) upon laser excitation (see Fig 5B for an example of an implantable substrate prepared by lithographic techniques). Moreover, advantages such as high information content, multiplexing capability, lack of extensive sample preparation, high tissue penetration and a highly sensitive detection limit to the level of single molecules (Qian et al, 2008) have led to various applications of SERS in biosensing, e.g. in cells (Kim et al, 2006), tissues and circulating tumour cell detection in human whole blood . The essential requirement for the SERS-based method is that an analyte or its reporter lies close to the nanoparticles or nanoroughened surface of noble metals such as gold or silver. Table 1 lists the means to attach an analyte molecule to the metal surface and its applications. The SERS technique has been used in mouse models for in vivo tumour targeting and detection Qian et al, 2008;von Maltzahn et al, 2009;Xiao et al, 2009;Zavaleta et al, 2009). In the context of metabolic sensing, the promise of SERS lies on the detection of glucose and its derivatives, and of insulin and other peptide hormones using SERS-active nanoparticles attached to a fibre-optic sensor (Zhang et al, 2007). Even though optical fibre-based SERS biosensing is still at its infancy (Shi et al, 2009), we anticipate that this technique will be of great use as the fibre can be readily configured for in vivo applications to allow SERS-based metabolic studies in living animals in the near future.
In vivo imaging of pancreatic b-cell mass and function
Besides defective insulin secretory processes, reduced b-cell mass with consequent decreased production and secretion of insulin also contributes to the development of type 2 diabetes. Whether there is a minimal number of pancreatic islets needed to maintain proper blood glucose levels or a sizeable reduction (e.g. by 50%) in b-cell mass directly results in diabetes is not yet
Bridge the Gap
Approaches to image metabolism . Cellular and molecular regulation of insulin secretion. The cellular events leading to insulin secretion start with a rise in glucose level in the blood, which quickly leads to glucose uptake into pancreatic b-cells. Glucose in the cells then undergoes glycolysis and Krebs cycle to produce ATP, resulting in an increased ATP/ADP ratio, and consequent closure of K ATP -channels. Membrane depolarization from K ATP -channel closure opens L-type calcium channels, allowing calcium influx into the cells, which then triggers insulin granule exocytosis and the release of insulin into blood. GluT-2, glucose transporter 2; GK, glucokinase; TCA, tricarboxylic acid cycle; Syt7, synaptotagmin-7; K-ATP, ATP-sensitive potassium channel; L-Ca, L-type calcium channel.
clear (Kahn et al, 2009). It is thus important to quantify b-cell mass in vivo and to correlate it with glucose homeostasis in the course of diabetes development. Furthermore, determining bcell mass and function in vivo is essential to assess tissue-and cell-based therapies for diabetes.
From the drug development angle, an animal model that allows detection of b-cell mass in vivo will be extremely useful to evaluate the potential of drug candidates in preserving b-cell function and number, but imaging the pancreas in small animals is challenging as it is a thin layer of tissue of irregular shape between other large organs. Imaging pancreatic islets to evaluate the b-cell mass is even more difficult because the islets occupy only 1-2% of the total pancreas volume. In addition, motion artefacts due to respiration and cardiac pulsation may deteriorate the quality of high resolution imaging needed for resolving the pancreas. Due to the lack of difference in proton density and MRI relaxation times between b-cells and surrounding pancreatic tissues, b-cells cannot be differentiated based on these intrinsic contrasts in MRI. Although contrast agents can be designed to target specific receptors on cells, the limited number of receptors on b-cell surface requires high sensitivity, affinity and specificity of the contrast agent (for review of past efforts on developing b-cell imaging probes, see Schneider, 2008). As this likely represents the best translatable approach for non-invasive imaging of human b-cell mass, ongoing efforts are aimed at identifying novel b-cell specific biomarkers (Flamez et al, 2010;Ueberberg et al, 2010). So far, limited success has been achieved in imaging b-cell mass in rodents by MRI of transplanted islets or b-cells labelled with iron oxide nanoparticles or gadolinium (Gd) (Evgenov et al, 2006;Zheng et al, 2005), by bioluminescent imaging of genetically introduced luciferase gene in b-cells (Park & Bell, 2009), by Mn 2þ -enhanced MRI (Antkowiak et al, 2009) and by positron emission tomography (PET) imaging of dihydrotetrabenazine (DTBZ) bound Type 2 vesicular monoamine transporters (VMAT2) (Souza et al, 2006). We propose here two approaches to generate pancreatic b-cell specific labels that may allow quantification of b-cell mass in vivo.
Small molecule probes
Commonly used antibody-based imaging techniques are limited to tissue/cell surface targets (not intracellular) and repeated treatment may induce adverse immune responses. Tagging the antibodies with suitable labels (e.g. small molecules, metal complexes or nanoparticles) is an added challenge. Genetic modification of specific protein targets with fluorescence proteins such as green fluorescent protein (GFP) provides target specific labelling platforms in live systems but this approach is not applicable to a clinical setting limiting their usage to cell culture or animal model studies. Instead, small molecule probes are ideal for clinical application as demonstrated in many currently used drug molecules. Drug-like (2006); Yonzon et al (2004) imaging probes will also be easy-to-use universal research tools, which can be used in cell or animal studies, without additional genetic modifications. We note the revolutionary application of Fura dyes for calcium imaging; this powerful small molecule probe was the technical foundation for the booming of calcium signalling field during the last several decades (Grynkiewicz et al, 1985). Despite their great potential, target specificity remains a challenge in developing small molecule probes. We discuss two approaches to target probes to particular molecules or cells in Box 3. Fluorescent molecules are the most popular small molecule probes due to their high sensitivity and exceptional ease of handling compared to their radioactive counterparts. However, the intrinsic limit of light penetration of the fluorescent probe precludes its applications in deep tissue imaging. They can, however, be modified by attaching an MRI-, PET-or SPECT (single photon emission computed tomography)-compatible functional group or single atomic radioisotope, which would render them suitable for deep tissue imaging in animal or clinical studies. The general work flow of diversity oriented fluorescence C. Examples of some potential modifications to convert Glucagon Yellow into PET or SPECT probes. The 99m Tc ligand could also be replaced by a Gd chelate to form an MRI contrast agent.
BOX 3: How to target small molecule fluorescent probes
There are two general approaches to target small molecule fluorescent probes: one is analyte-oriented and the other diversity-oriented. The conventional analyte-oriented approach combines known analyte binding motifs to fluorescent molecules through a linker. Many fluorescence-based sensors have been developed through this approach, but each individual development requires a major effort in both designing and synthesizing the sensors and the sensor's application remains limited to the pre-selected specific analytes. To overcome this limitation and to accelerate novel sensor discovery, an alternative diversity oriented fluorescence library approach (DOFLA) was introduced recently (Lee et al, 2009b). In DOFLA, structurally diverse library compounds are generated by combinatorial chemistry and the library compounds are screened in high throughput manner either in purified analyte or against whole cell/tissue/organism. In addition to its high speed of novel probe discovery, this approach is advantageous over conventional method especially when the molecular target is not known, i.e. the whole cell can be used for screening, and the target can be identified later (system to target approach). DOFLA has been successfully used to generate probes against a broad range of targets including DNA (Lee et al, 2003), RNA (Li & Chang, 2006), b-amyloid (Li et al, 2004(Li et al, , 2007, GTP (Wang & Chang, 2006), heparin (Wang & Chang, 2008), chymotrypsin (Wang & Chang, 2008), human serum albumin , glutathione Min et al, 2007) and myotube cell-state discrimination (Wagner et al, 2008).
library approach (DOFLA) (see Box 3) and strategies to convert a particular probe for MRI, PET and SPECT imaging are summarized in Fig 6. We have recently applied this approach to identify Glucagon Yellow as an a-cell-selective probe (Lee et al, 2009a). We believe that pancreatic b-cell specific probes can be identified by adopting the approaches described in Box 3, and that the identified fluorescent probe can be converted as in Fig 6 to allow quantification of b-cell mass in vivo.
Genetic approach
An alternative to the small molecule approach is by genetic introduction of complementary DNA (cDNAs) encoding proteins that bind to MRI-, PET-or SPECT-compatible probes, or that can accumulate iron and form MR-detectable iron particles. Although this cannot be adapted to clinical usage, applying it to animal models could generate useful pre-clinical models to evaluate therapeutic compounds for their ability to preserve bcell mass. Several studies have demonstrated the use of ironbinding proteins such as ferritin and a bacterial protein MagA in in vitro and in vivo imaging (Cohen et al, 2005(Cohen et al, , 2007Genove et al, 2005;Goldhawk et al, 2009;Zurkiya et al, 2008). One can envisage the use of transgenic mice expressing one of these proteins specifically in pancreatic b-cells for the detection and quantification of b-cell mass.
Functional assessment of pancreas in vivo
Pancreatic blood flow, blood volume and vascularization can provide valuable information on the pancreatic islet function and the re-establishment of b-cell function after transplantation. Dynamic contrast enhanced MRI with intravenous injection of Gd-based contrast agent has been used to estimate these haemodynamic parameters in humans (Coenegrachts et al, 2004;Yu et al, 2009). Since one passage of the contrast agent through circulation can be as short as 5 s in mice (Nyman et al, 2008) and 20 s in human, this makes it difficult to calculate blood flow from the very fast contrast agent kinetics. Instead, the time-to-peak and the area under curve of the contrast agent kinetics are commonly used as an estimate of blood volume and vascularization (Hathout et al, 2007;Medarova et al, 2007). We recently demonstrated that similar contrast agent enhanced MRI can be applied to study blood flow in mouse pancreas (Lee et al, 2009c). Another approach for quantifying blood flow is arterial spin labelling (Schraml et al, 2008), but this has its own share of challenges as well. A method based on gradient spin echo (GRASE) image acquisition can significantly reduce the susceptibility artefacts (unpublished observations) and could be applied in abdominal imaging. The present MR technologies cannot compete with the stunning time-resolved blood flow patterns of exposed mouse pancreas in 3D obtained by linescanning confocal microscopy (Nyman et al, 2008). However, image registration and correlation of high resolution optical image data with MRI can provide a way for translating animal model observations to in vivo and to human studies. Imaging glucose, its derivatives and the activation of b-cells in vivo is another important functional readout of the pancreas.
Glucose and glycogen can be detected by MRI using specific paramagnetic lanthanide complexes, which generate a chemical exchange saturation transfer (CEST) effect (Zhang et al, 2003). Glucose distribution in the liver has been mapped ex vivo in this manner (Ren et al, 2008). However, due to strong magnetic field heterogeneity and motion, it will prove to be difficult to apply this approach in in vivo studies. Using manganese ion (Mn 2þ ) as a calcium analogue allows the measurement of calcium influx into pancreatic b-cells after glucose stimulation (for review, please see Koretsky & Silva, 2004). Indeed, Mn 2þ -enhanced MRI has been used to observe b-cell activity in isolated b-cells and in mouse pancreas in vivo (Antkowiak et al, 2009). Although the toxicity of Mn 2þ may limit its application in humans, this non-invasive technique complements the restriction of optical fluorescent imaging.
Although MRI is sensitive to different physical and physiological parameters in vivo, it detects the ensemble of all the parameters and cannot differentiate individual components. This limits our ability to image multiple cell/islet populations and/or biological processes in parallel to understand their interactions and dynamics. One strategy to overcome this limitation is by using frequency-shifting contrast agents to change the resonance frequency of the water instead of changing the T 1 or T 2 relaxation times (Zabow et al, 2008). This allows images to be generated at different frequencies like quantum dots in optical imaging. Alternatively, one can combine the use of conventional relaxation agents (e.g. iron oxide nanoparticles or Mn 2þ ) and a CEST contrast agent (Gilad et al, 2009). A potential application is to track transplanted b-cells with iron oxide (or a frequency-shifting agent) and to monitor their function using Mn 2þ -or Zn 2þsensitive agents and/or CEST effect.
New technologies in imaging metabolic diseases
MRS using carbon ( 13 C) is ideally suited to the study of metabolism due to the extensive range of compounds that can be detected and the ability to attribute signals to the different carbon atoms within individual molecules. 13 C studies of metabolism in cells and in vivo have been conducted since the early 1980s, pioneered by Robert Shulman. Bailey et al carried out the first experiments on isolated perfused rat hearts using 13 C-enriched sodium acetate (Bailey et al, 1981). A wealth of work has since followed on cardiac metabolism, despite limitations of the method caused by both the low natural abundance of the MRSvisible isotope of carbon ( 13 C) and the low level of magnetic polarization normally achievable. The low natural abundance of 13 C has required most work to be conducted using 13 C-enriched molecules and even then long scan times are required, resulting in the study of metabolic steady-state conditions.
Recently, a new technique-DNP-MR-has produced a practical method to enhance magnetic polarization levels by more than 10,000-fold Golman et al, 2003) (Fig 7). The technique combines the solid-state methods of dynamic nuclear polarization (DNP) (Bailey et al, 1981) with a rapid dissolution procedure to produce stable injectable solutions (in vivo stability of approximately 60 s) (Golman et al, 2006a;Malloy et al, 1987). DNP-MR allows visualization of 13 C-labelled metabolites in vivo and, more importantly, their enzymatic transformation into other species Tyler et al, 2008). Initial trials of the DNP-MR method focussed on the study of hyperpolarized [1-13 C]pyruvate (Golman et al, 2006b) and indicate that its rapid conversion to [1-13 C]lactate, [1-13 C]alanine and bicarbonate (H 13 CO 3 À ) should provide a sensitive test for cardiac metabolism. In the heart, the metabolism of pyruvate and the activity of mitochondrial pyruvate dehydrogenase (PDH) play key roles in oxidative metabolism (Chen et al, 2007;Schroeder et al, 2009). The development of metabolic imaging with hyperpolarized MR Golman et al, 2006a) enabled unprecedented visualization of the biochemical mechanisms of normal and abnormal metabolism (Chen et al, 2007;Day et al, 2007;Golman et al, 2006b;Merritt et al, 2007), such as measuring PDH flux in vivo. In vivo, the hyperpolarized tracer [1-13 C]pyruvate rapidly generates the visible metabolic products [1-13 C]lactate, [1-13 C]alanine and bicarbonate (H 13 CO 3 À ), which exist in equilibrium with carbon dioxide ( 13 CO 2 ). Because it is the PDH-mediated decarboxylation of pyruvate into acetyl-CoA that produces 13 CO 2 , monitoring the production of hyperpolarized H 13 CO 3 À should enable a direct, non-invasive measurement of flux through the PDH enzyme complex (Chen et al, 2007), which is highly dependent on PDH activity in vivo.
The Krebs cycle is fundamental to cardiac energy production, and is often implicated in energetic imbalances characteristic of heart disease. To date, Krebs cycle flux has been measured using 13 C-MR spectroscopy with isotopomer analysis; however, this approach is limited to the study of steady-state metabolism only and has limited in vivo applications. Hyperpolarized [2-13 C]pyruvate was demonstrated in a recent study to serve as a tracer to monitor Krebs cycle metabolism in the isolated perfused heart directly (Schroeder et al, 2009). Hyperpolarized [2-13 C]pyruvate was infused into healthy hearts, and the metabolic products with sufficient MR signal for detection at high temporal resolution were identified. The time courses of the formation of each of these metabolites gave kinetic information describing the relationships among cytosolic metabolism of [2-13 C]pyruvate, PDH-mediated oxidation of [2-13 C]pyruvate and its subsequent incorporation into the Krebs cycle. In addition, hyperpolarized [2-13 C]pyruvate was infused at the moment of reperfusion into globally ischaemic hearts, to identify differences in [2-13 C]pyruvate metabolism in the reperfused myocardium. This study demonstrated that Krebs cycle metabolism can be directly and instantaneously monitored by using hyperpolarized [2-13 C]pyruvate as a metabolic tracer, and that new information can be obtained about the coordination of glycolysis, pyruvate oxidation and Krebs cycle flux in the normal and post-ischaemic myocardium (Schroeder et al, 2009). Hence, hyperpolarized 13 C enables non-invasive, in vivo and real time assessment of metabolic processes, with the possibility of quantification, e.g. enzyme-mediated flux of metabolite intermediates (PDH, LDH (L-lactate dehydrogenase), CA). When coupled with metabolic imaging, it is possible to study the spatial distribution of metabolites and to measure in vivo pH at multiple time points (e.g. see Fig 7C). The technique may be applied to study the impact of carbohydrate metabolism on disease progression, and to evaluate the efficacy of therapeutic interventions longitudinally, e.g. to monitor glucose metabolism at different stages of diabetes development. Since a-cells, but not b-cells possess high level of surface transporters for pyruvate (monocarboxylase transporters) (Ishihara et al, 2003), a potential application of hyperpolarized imaging of pyruvate is to study the physiological and pathological changes specifically in pancreatic a-cells.
Although technical details are still being worked out, future applications of the technique in lipid metabolism will be highly significant in understanding changes in lipogenesis during development of obesity and other metabolic diseases.
Bridge the Gap
Weiping Han et al.
Bridge the gap
The Gap Metabolic syndrome, a pandemic problem in most of the developed countries and many developing nations, poses a significant burden to the health care of the affected countries. Despite intense efforts from both basic and clinical scientists, many fundamental questions regarding its aetiology and development remain unanswered. The fine details of in vivo insulin secretion regulation remain elusive. The threshold of b-cell mass needed to keep glucose level in check is not known. In addition, biomarkers that reflect in vivo bcell mass or pancreatic activity are also needed as endpoints in clinical trials or for early drug development assays. Although many technologies are available to address these issues in vitro or ex vivo, what is urgently needed is clinical applicable technologies and proper animal models that can be used to understand the disease and test potential therapeutic candidates.
The Bridge
We believe imaging represents the most suitable technology in addressing these questions, with a track record of successful pre-clinical applications and clinical adaptation. Answering the questions above requires the development of suitable imaging tools and this Bridge the Gap article proposes a series of approaches that could be used to tackle these and other relevant matters in the field of metabolic diseases.
Concluding remarks
Despite intense efforts from both basic and clinical scientists, many fundamental questions regarding the aetiology and development of the metabolic syndrome remain unanswered. We believe imaging represents the most suitable technology in addressing such questions, with a track record of successful preclinical applications and clinical adaptation. We discussed many imaging possibilities and approaches in this article and summarized them in Table 2. While some are currently in use, many others are still in early phase of development or of translation to clinical application. The diversity and the complexity of these approaches highlight the need for close collaborations among biologists, clinicians, chemists, physicists and engineers to develop suitable imaging tools that allow visualization of metabolic processes in vivo.
==
Domain: Biology Medicine
|
[{"added": "2017-06-22T20:54:26.809Z", "created": "2010-06-01T00:00:00.000", "id": "14759101", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.1002/emmm.201000074", "pdf_hash": "c9dd1c39d1dacd350fee16bbbe5042d75aae7c2e", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:36862", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "f61c19bb6b407928dcb9adf6b5cfe550a6ccc23e", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The expanding spectrum of neurological disorders of phosphoinositide metabolism
ABSTRACT Phosphoinositides (PIPs) are a ubiquitous group of seven low-abundance phospholipids that play a crucial role in defining localized membrane properties and that regulate myriad cellular processes, including cytoskeletal remodeling, cell signaling cascades, ion channel activity and membrane traffic. PIP homeostasis is tightly regulated by numerous inositol kinases and phosphatases, which phosphorylate and dephosphorylate distinct PIP species. The importance of these phospholipids, and of the enzymes that regulate them, is increasingly being recognized, with the identification of human neurological disorders that are caused by mutations in PIP-modulating enzymes. Genetic disorders of PIP metabolism include forms of epilepsy, neurodegenerative disease, brain malformation syndromes, peripheral neuropathy and congenital myopathy. In this Review, we provide an overview of PIP function and regulation, delineate the disorders associated with mutations in genes that modulate or utilize PIPs, and discuss what is understood about gene function and disease pathogenesis as established through animal models of these diseases.
Introduction
Phosphatidylinositol (PtdIns) and its derivatives are membrane lipids that are composed of a phosphatidic acid (PA) backbone esterified to a myo-inositol ring. This inositol ring can be reversibly phosphorylated at three positions (D3, D4, D5) to generate a group of seven phosphoinositides (PtdIns phosphates or PIPs). PtdIns synthesis begins with the phosphorylation of diacylglycerol (DAG) into PA by DAG kinase. Next, PA is converted into cytidine diphosphate (CDP)-DAG by phosphatidate cytidylyltransferase (Lykidis et al., 1997;D'Souza et al., 2014). The enzyme CDIPT (also known as PtdIns synthase or PIS) then combines CDP-DAG with myo-inositol in the endoplasmic reticulum (ER) to make PtdIns (Lykidis et al., 1997).
To improve the understanding of the involvement of PIPs in neurological function, a comprehensive list of genes involved in PtdIns binding and PIP metabolism is presented in Table S1. This list was generated from the existing literature (e.g. Di Paolo and De Camilli, 2006;Sasaki et al., 2009;Balla, 2013 and references therein) and from an InterPro database ( [URL]/) search of genes containing a PIP-binding protein domain (Mitchell et al., 2019). There are 25 PtdIns/PIP kinases (including regulatory subunits), 31 PIP phosphatases and 742 genes with PIP-binding domains. Among these 798 genes, mutations in 163 are associated with a confirmed human genetic disorder as curated in the Online Mendelian Inheritance in Man (OMIM) database ( [URL]. omim.org/). Intriguingly, 83 of the disease-associated genes feature neurological or neuromuscular involvement, highlighting 1 the importance of PIP regulation in neurological function (Table S2). It is important to note that many genes with a PIP-binding domain in this list have not been functionally validated to determine whether they indeed bind PIPs; thus, this list should serve primarily as a reference for future studies.
In this Review we present the human neurological disorders associated with mutations in PIP-modulating enzymes (Table 1) and in proteins with PIP-binding domains, and discuss the mechanistic relationship between disease and PIP metabolism by focusing on patient-related information and data from cell and animal models. For a more in-depth review of the regulation of PIPs and their involvement in non-neurological disorders, we refer readers to several excellent reviews on this topic (McCrea and De Camilli, 2009;Billcliff and Lowe, 2014;Staiano et al., 2015;Burke, 2018).
Neuromuscular diseasesskeletal myopathies Myotubular myopathy
One of the best-characterized neuromuscular disorders of PIP metabolism is the severe childhood muscle disease X-linked myotubular myopathy (XLMTM; MIM 310400) (Laporte et al., 1997;Blondeau et al., 2000;Biancalana et al., 2003;Jungbluth et al., 2008;Lawlor et al., 2016;Dowling et al., 2009;Amburgey et al., 2017). XLMTM is associated with profound muscle weakness that manifests as a failure to achieve motor milestones, wheelchair and ventilator dependence, and death before the age of 10 years in most cases (Amburgey et al., 2017). The disorder is defined by a unique set of skeletal muscle biopsy features that include centrally located nuclei, disorganized perinuclear organelles and myofiber hypotrophy (Dowling et al., 2002;Lawlor et al., 2016). The primary ultrastructural abnormality observed in XLMTM is the disrupted appearance and function of a muscle substructure called a triad (T-tubules and sarcoplasmic reticulum).
Vertebrate models have provided important insights into how MTM1 loss causes muscle disease, including the key observation of triad abnormalities. The first model generated was a mouse Mtm1 knockout (KO) (Buj-Bello et al., 2002), which showed that loss of There are seven phosphorylation states of phosphatidylinositol that are targeted by kinases (blue) and phosphatases (green). Perturbations in these pathways are associated with a growing number of neurological disorders, as indicated. Dotted arrows and question marks indicate conversions that have yet to be supported in the literature.
Mtm1 protein results in muscle pathology, even though the gene is ubiquitously expressed. The Mtm1 KO mouse has subsequently been instrumental for testing treatment strategies, including the evaluation of adeno-associated virus (AAV)-based gene replacement therapy (Buj-Bello et al., 2008;Childers et al., 2014). It was also used to identify Pik3c2b, which encodes a PIP kinase, as a genetic modifier of the disease (Sabha et al., 2016). In mice, Pik3c2b KO both prevents and completely reverses the Mtm1 KO phenotype (Sabha et al., 2016). Interestingly, the knocking out of Pik3c3, which encodes another phosphoinositide 3-kinase (PI3K), significantly worsens the Mtm1 KO phenotype (Sabha et al., 2016). These results suggest that MTM1 and PIK3C2B co-regulate a specific pool of endosomal PtdIns3P that is distinct from that regulated by PIK3C3 (Cao et al., 2007;Gavriilidis et al., 2018). PIK3C3 can generate PtdIns3P at several membrane compartments, including nascent autophagosomal membranes and various intermediates of the endolysosomal pathway (Reifler et al., 2014;Ketel et al., 2016), and conditional KO of Pik3c3 in skeletal muscle causes lysosomal dysfunction and disrupts autophagy (Reifler et al., 2014). These phenotypes associated with Pik3c3 KO might explain the worsening of the Mtm1 phenotype in double mutants, as both zebrafish and mouse MTM1 mutants show impaired autophagic flux (Dowling et al., 2010;Fetalvero et al., 2013).
Two zebrafish models of XLMTM have also been created (Dowling et al., 2009;Sabha et al., 2016). An mtm1 morphant was instrumental in uncovering the triad abnormalities that are now considered a key pathological change in this disease (Dowling et al., 2009). This model was also used to identify and evaluate drug therapies, such as pyridostigmine, which was shown to have efficacy in the mouse model and is now routinely used in an offlabel manner in patients with XLMTM . Subsequently, an exon 5 mtm1 mutant was generated and used to demonstrate that pharmacological and genetic perturbation of specific PI3Ks ameliorates mtm1 phenotypes (Sabha et al., 2016). There is also a Labrador dog model with an MTM1 mutation (Beggs et al., 2010). AAV gene replacement therapy has been tested in this model, which has, in turn, provided the critical safety and proofof-concept data necessary for translation of this therapy to a first-in-human clinical trial (Childers et al., 2014).
A mouse Bin1 KO mutant has been generated that has a recessive perinatal lethal phenotype and skeletal muscle weakness (Muller et al., 2003). Intriguingly, this severe phenotype is rescued by lowering Dnm2 levels using a similar strategy to that employed for Dnm2 reduction in Mtm1 KO mice . These data imply that Mtm1 and Bin1 negatively regulate Dnm2 in mice. In addition, Bin1 overexpression rescues the Mtm1 KO mouse phenotype (Lionello et al., 2019), further supporting a model of critical interplay and counterbalancing between these three proteins.
The exact nature of this interplay is still not understood, nor is it clear how these interactions contribute to the CNM disease process. Of note, the field is currently hindered by the lack of a pre-clinical model of DNM2-related CNM that phenocopies the clinical aspects of the human disease and thus could be used to develop and test potential therapies. Other models exist, including a mouse Dnm2 knock-in with a mild phenotype (Buono et al., 2018;Fongy et al., 2019), a zebrafish morphant (Bragato et al., 2016) and a transient dominant dnm2 zebrafish model (Gibbs et al., 2014), but additional models are clearly needed.
In addition, a very rare form of CNM with cardiomyopathy (found in three unrelated individuals) is caused by recessive mutations in the striated muscle preferentially expressed protein kinase (SPEG) gene (Agrawal et al., 2014). SPEG is a serine/threonine kinase that interacts with MTM1 as assessed by a two-hybrid assay (Agrawal et al., 2014) and by co-immunoprecipitation (Agrawal et al., 2014;Gavriilidis et al., 2018). A mouse KO of Speg resembles human CNM clinically and histopathologically and has a severe phenotype similar to that of the Mtm1 KO mouse (Liu et al., 2009). Given the phenotypic similarity between XLMTM and patients with SPEG mutations, and the physical interaction between SPEG and MTM1, it is tempting to speculate that SPEG plays some role in the regulation of PIP metabolism and/or endolysosomal trafficking. Charcot-Marie-Tooth (CMT) disease is a degenerative disorder of the peripheral nervous system with a global prevalence of 1/2500 (Bird et al., 1998a;Timmerman et al., 2014;Magy et al., 2018). It usually presents as a slowly progressing disease, although age of onset varies from infancy to adulthood and severity can range from mild walking difficulties to a complete inability to ambulate independently (Bird et al., 1998a;Timmerman et al., 2014;Magy et al., 2018). Typical symptoms include distal muscle weakness and sensory loss, distal muscle atrophy and skeletal deformities including, in particular, pes cavus and hammertoes (Bird et al., 1998a). CMT is categorised into subtypes based on nerve pathology (affecting either the myelinating Schwann cell or the axon that it ensheaths) and mode of inheritance (dominant, recessive or X-linked). Nerve pathology is typically defined by nerve conduction studies: slow conduction velocities signify demyelinating CMT, whereas decreased action potential amplitudes signify axonal CMT (Bird et al., 1998a). Given that CMT is extremely genetically heterogeneous, with causative mutations in more than 80 genes identified to date (Timmerman et al., 2014;Magy et al., 2018), there has been a gradual shift towards a gene-based classification system (Magy et al., 2018). The proteins encoded by CMT genes have diverse cellular and molecular functions, the elucidation of which has also led to seminal insights into the cell biology of the peripheral nervous system (Tazir et al., 2014).
Charcot-Marie-Tooth disease type 4B
Mutations in genes involved in PtdIns metabolism cause autosomal recessive subtypes of CMT (i.e. CMT type 4; CMT4). These subtypes generally represent severe demyelinating peripheral neuropathies, with onset during early childhood that cause profound disability (Bird et al., 1998b). CMT4B1 (MIM 601382) is caused by mutations in the myotubularin family member MTMR2 (Bolino et al., 2000), whereas CMT4B2 (MIM 604563), which can also feature glaucoma, is caused by mutations in MTMR13 (SBF2) (Senderek et al., 2003), which encodes a phosphatase-dead myotubularin-related protein that interacts directly with MTMR2 (Robinson and Dixon, 2005). CMT4B3 (MIM 615284), which has a similar presentation to the other two subtypes, is caused by mutations in MTMR5 (SBF1) (Nakhro et al., 2013), which also encodes a catalytically inactive myotubularin family member that interacts with MTMR2 (Kim et al., 2003). These three subtypes share a distinctive nerve pathology, featuring generalized loss of large myelinated nerve fibers and focally folded myelin sheaths (Bird et al., 1998b).
MTMR2/5/13 are all members of the myotubularin gene family, of which MTM1 is the canonical member. As with MTM1, MTMR2 is a 3-phosphatase located at the endosome that acts on both PtdIns3P and PtdIns(3,5)P 2 . Its stability and activity are modulated by MTMR5 and MTMR13 (Kim et al., 2003;Robinson and Dixon, 2005). MTMR2 is implicated in endosomal traffic, as is MTM1, although some evidence suggests that MTM1 and MTMR2 act on separate endosomal fractions with different PIP preferences in vitro (Cao et al., 2008). That said, when Mtmr2 is overexpressed in the skeletal muscle of the Mtm1 KO mouse, it can rescue the KO phenotype (Raess et al., 2017), indicating functional redundancy between these two MTMR family members.
In Schwann cells (the cell type affected in CMT4B), MTMR2 and MTMR13 localize to the cytoplasm and to endomembrane punctae (Ng et al., 2013). The role(s) of MTMR2, MTMR13 and MTMR5 in normal Schwann cell biology, and why their loss causes demyelination, is incompletely understood. In addition, it is not clear why mutations of the two modulators of MTMR2 (MTMR5 and MTMR13) result in the same phenotype and do not exhibit functional compensation.
Mtmr2, Mtmr5 and Mtmr13 mouse KOs have been generated that phenocopy each other and show initial, early-onset dysmyelination [at postnatal day 3 (P3)] with myelin outfoldings, followed by later nerve degeneration (Bolino et al., 2004;Bonneick et al., 2005;Firestein et al., 2002;Robinson et al., 2008;Tersar et al., 2007). By using these mice, as well as cell culture overexpression, one study has postulated that these proteins regulate epidermal growth factor receptor (EGFR) recycling and thus phospho-Akt signaling (Berger et al., 2011). However, another study that examined peripheral nerve extracts from these KO mutant mice at a different time point did not find evidence of altered Akt signaling (Ng et al., 2013), suggesting a lack of EGF modulation. An alternative hypothesis has been proposed, based on two-hybrid and direct binding assays, that MTMR2 interacts with the scaffold protein discs large 1 (Dlg1) and the motor protein kinesin 13B (Kif13b) (Bolino et al., 2004;Bolis et al., 2009). These studies determined that Dlg1 and Kif13b are mis-expressed and mis-localized in Mtmr2 KO mice and postulated, based on these molecular interactions and the association of Dlg1 with the exocyst protein Sec8 (Exoc4), that MTMR2 negatively regulates myelin membrane formation, counterbalancing the normal synthesis of myelin.
Mtmr2 KO mice have been used to identify novel therapeutic targets for CMT4B. Haploinsufficiency of the PIP 5-phosphataseencoding gene Fig4, which is required for both the synthesis and breakdown of PtdIns(3,5)P 2 , lowers PtdIns(3,5)P 2 levels and ameliorates aspects of the Mtmr2 KO phenotype in mice (Vaccari et al., 2011). Thus, Fig4 and Mtmr2 genetically interact. A recent study (Bolino et al., 2016) has also demonstrated that an inhibitor (apilimod) of the PIP 5-kinase PIKFYVE, which interacts with FIG4 and generates PtdIns(3,5)P 2 , can improve myelination of Mtmr2 KOderived Schwann cells in vitro. Apilimod appears to be safe in humans (Gayle et al., 2017;Krausz et al., 2012) and future studies aim to assess its efficacy in vivo in Mtmr2 KO mice. Worth noting is the fact that MTMR2 and FIG4, mutations that are discussed extensively in the next section, have a complex relationship. Whereas Fig4 haploinsufficiency improves the neuropathy phenotype of Mtmr2 KO mice, Mtmr2 mutation does not appear to positively modulate the Fig4 phenotype (and in fact increases motor neuron loss) (Vaccari et al., 2011(Vaccari et al., , 2015. Another potential treatment strategy has been developed based on the role of neuregulin 1 (Nrg1) type III in regulating myelin formation and the thickness of the myelin sheath. The observation of over-folded myelin (a potentially 'hypermyelinating' phenotype) in CMT4B1 led Bolino and colleagues to propose the inhibition of Nrg1 signaling as a treatment (Bolino et al., 2016). They tested this hypothesis using Niaspan, an FDA-approved drug that activates TNF-alpha-converting enzyme, an Nrg1 pathway inhibitor, and found that it rescues the myelin defect observed in Mtmr2 KO mice (Bolino et al., 2016).
CMT4J: neuropathy caused by FIG4 mutation
Another subtype of CMT type 4, called CMT4J, is caused by recessive mutations in FIG4 (also known as SAC3) Nicholson et al., 2011). CMT4J is characterized by variable age of onset, often with asymmetry in the muscle groups affected, and periods of rapid clinical deterioration (Li et al., 2013). Needle electromyography studies reveal non-uniform slowing of conduction velocities and evidence of denervation (Li et al., 2013). Genetically, patients with CMT4J typically carry one missense change (most commonly the recurrent p. I41T mutation) and one nonsense mutation Nicholson et al., 2011). Two FIG4 nonsense mutations cause a severe neurodegenerative syndrome, Yunis-Varon Syndrome (YVS; MIM 216340) described below (Campeau et al., 2013), with features that are distinct from CMT4J.
FIG4 is a PtdIns(3,5)P 2 -specific 5-phosphatase (Rudge et al., 2004) that exists in a molecular complex with the scaffold protein VAC14 and the 5-kinase PIKFYVE, which together regulate the metabolism of PtdIns(3,5)P 2 (Gary et al., 2002;Rudge et al., 2004;Jin et al., 2008). Beyond evolutionary explanations, the precise reason that a kinase and phosphatase with opposing roles exist in the same complex is still unknown; however, it is hypothesized that this arrangement enables dynamic spatiotemporally regulated control of PtdIns(3,5)P 2 (Jin et al., 2008). Much of what we know about FIG4 has emerged from studies in yeast (Gary et al., 2002;Rudge et al., 2004;Jin et al., 2008) and from a spontaneous mouse mutant strain called pale tremor ( plt) . The plt mouse has an abnormal coat color, manifests a progressive neurodegenerative phenotype that includes limb weakness, spasticity and early death, and extensive vacuolization in the neurons of the central and peripheral nervous system . The underlying genetic cause of the plt phenotype is an insertion mutation at the Fig4 locus that completely abrogates Fig4 protein production. Because of the striking peripheral nerve phenotype seen in these mice, Meisler and colleagues screened patients with CMT and identified FIG4 mutations in four unrelated individuals . Subsequently, numerous CMT patients have been found to have biallelic FIG4 mutations (i.e. CMT4J).
Loss of Fig4 in mice is associated with decreased levels of PtdIns(3,5)P 2 and with slight increases in PtdIns3P, reflecting the role of FIG4 in stabilizing PIKFYVE expression and function. Indeed, a Pikfyve hypomorphic mouse with 65-90% reduction in PIKFYVE protein levels across tissues shows moderate to severe neurodegeneration and dies perinatally (Zolov et al., 2012), highlighting the importance of PIKFYVE in neuronal development. Likewise, PIKFYVE likely directly generates all of the PtdIns(3,5)P 2 in the cell, which is then used to generate the majority of PtdIns5P (Zolov et al., 2012).
The VAC14 gene encodes the FIG4 and PIKFYVE scaffolding protein VAC14. Vac14 mutant mice (one KO model and one missense mutant) have a similar phenotype to plt mice Jin et al., 2008). Importantly, Vac14 mutants show reduction in PtdIns(3,5)P 2 levels Jin et al., 2008). Of note, two patients with a severe neurodegenerative phenotype of progressive debilitating dystonia, loss of motor milestones and deep gray matter changes on brain MRI, characterized as striatonigral degeneration, childhood-onset (MIM 617054), were reported to have biallelic (compound heterozygous) mutations in VAC14 (Lenk et al., 2016). In addition, one patient with YVS (discussed below) harbored mutations in VAC14 (Lines et al., 2017). Taken together, these data suggest that mutations in FIG4 and VAC14 result in depletion of PtdIns(3,5)P2 via the loss of PIKFYVE activity (despite the role of FIG4 as a 5-phosphatase) and are associated with similar severe neurodegenerative phenotypes.
On a cellular level, reduction in PtdIns(3,5)P 2 affects the PIPsensitive lysosomal calcium channel TRPML1 (MCOLN1; Li et al., 2016). Reduced PtdIns(3,5)P 2 results in reduced calcium release from TRPML1; this calcium flux is required for the proper maturation of lysosome vesicles, and its reduction is associated with impaired lysosomal motility and tubulation and aberrant autophagy via impaired autophagosome-lysosome fusion . Of note, TRPML1 mutations result in the disease mucolipidosis type 4 (MIM 252650), which features neuronal vacuolization but has different clinical manifestations compared to FIG4-associated CMT4J or YVS (Bargal et al., 2000). TRPML1 agonists have been developed and provide interesting candidates for treating FIG4related disorders. In fact, a chemical agonist that increases calcium flux from TRPML1 can reduce vacuole formation in fibroblasts and in dorsal root ganglia explants derived from Fig4 mice (Zou et al., 2015). Despite the clear functional interplay between FIG4 and TRPML1, why mutations in these genes cause different phenotypes in not known, though it likely indicates that each gene product has functions beyond lysosomal PtdIns(3,5)P 2 -dependent calcium release via TRPML1.
Other FIG4-related neurological disorders
Since the discovery of FIG4 mutations in CMT4J patients, FIG4 mutations have been associated with other conditions, such as YVS, which is caused by recessive FIG4 hypomorphic mutations (Nakajima et al., 2013;Campeau et al., 2013). Patients present with prenatal growth retardation and have multiple congenital anomalies, including microcephaly, cataracts, sparse pale hair, facial dysmorphisms and micrognathia (Nakajima et al., 2013;Campeau et al., 2013). Many have congenital heart defects, absent or hypoplastic clavicles, and transverse upper and lower limb anomalies, and most die in infancy (Campeau et al., 2013). Survivors have severe global developmental delay and frequently present brain malformations, including agenesis of the corpus callosum, arhinencephaly, frontal lobe atrophy and pachygyria. On autopsy, cytoplasmic vacuoles that represent endolysosomal bodies are found in the bone, muscle and brain tissue of patients (Campeau et al., 2013).
To date, one family with bilateral temporo-occipital polymicrogyria associated with FIG4 mutation has been reported (Baulac et al., 2014). The six affected individuals in this family carried a homozygous missense FIG4 mutation predicted to cause loss of protein function. All presented with complex partial seizures, with onset ranging from birth to 24 years old; two had a psychotic disorder with aggressiveness and delirium, and another two committed suicide. Brain MRI in three patients showed temporooccipital polymicrogyria. Combined with the MRI changes observed in YVS, these findings suggest that FIG4 mutations can cause a spectrum of brain malformations potentially associated with impaired neuronal migration.
Lastly, heterozygous FIG4 mutations have been reported to cause amyotrophic lateral sclerosis (ALS; MIM 612577), a progressive disorder associated with motor neuron pathology (Chow et al., 2009). Nine individuals, six with sporadic and three with familial disease, from a cohort of 473 ALS patients were found to have FIG4 mutations. The average age of disease onset was 56±14 years. Each had a single, heterozygous FIG4 mutation, two of which were stop codons, two consensus splice-site variants and five missense mutations; two of the missense mutations were likely pathogenic (Chow et al., 2009). In addition, two FIG4 missense mutations have been found in a patient with an aggressive form of ALS (Bertolin et al., 2018). However, FIG4 mutations relatively rarely cause ALS. Other groups did not identify pathogenic mutations in FIG4 after screening 80 Italian ALS patients (Verdiani et al., 2013), 15 Taiwanese patients with familial ALS (Tsai et al., 2011) and eight ALS families from southeastern China .
Genotype-phenotype correlations have not been explored in depth for FIG4-related disorders, largely because mutations in FIG4 are rare, particularly outside of CMT4J. One observation is the association between the amount of protein expression and severity, as seen with YVS (caused by biallelic nonsense mutations and thus predicted to result in little or no protein expression) versus CMT4J (in which one mutated allele is typically a missense mutation). Further work is needed to establish genotype-phenotype relationships and, in addition, to define the impact on protein function of missense mutations associated with non-CMT/YVS phenotypes.
Neurological disorders associated with PtdIns(4,5)P 2 metabolism PtdIns(4,5)P 2 is thought to regulate numerous subcellular processes and signaling cascades. It has been extensively characterized as part of the canonical PtdIns pathway as a substrate for phospholipase C (PLC), producing the second messengers DAG and IP 3 (Payrastre et al., 2001;Di Paolo and De Camilli, 2006). PtdIns(4,5)P 2 is also a second messenger, regulating actin filament dynamics and thus important in cell migration, cell-cell adhesion and cytokinesis (van den Bout and Divecha, 2009). It also modulates cell survival, as well as nuclear processes such as cell-cycle progression and splicing (van den Bout and Divecha, 2009). Given its importance, it is perhaps unsurprising that several disorders are associated with abnormalities in PtdIns(4,5)P 2 regulation.
Lethal congenital contracture syndromes
Lethal congenital contracture syndromes (LCCSs) are rare conditions associated with severe joint contractures (such as arthrogryposis multiplex congenita), reduced or absent limb movements, and lethality at or soon after birth (Narkis et al., 2007;Koutsopoulos et al., 2013). LCCS3 (MIM 611369) is associated with mutations in PIP5K1C. PIP5K1C encodes a type I kinase, PIPK1γ, responsible for the synthesis of PtdIns(4,5)P 2 from PtdIns4P (van den Bout and Divecha, 2009;Lacalle et al., 2015). Narkis and colleagues described two families with LCCS3: one large consanguineous family with nine affected individuals and another family with one affected male (Narkis et al., 2007). Patients had multiple joint contractures with severe muscle wasting and atrophy, but they did not exhibit features seen in other LCCSs, such as hydrops, fractures or multiple pterygia (Narkis et al., 2007). The specific mutation described in both families was the biallelic substitution p. D253N in the PIPK1γ kinase domain that abrogates kinase activity (Narkis et al., 2007).
The hypothesized pathogenic mechanism of PIP5K1C mutations concern the known role of PtdIns(4,5)P 2 in synaptic vesicle trafficking and in clathrin coat dynamics (Di Paolo et al., 2004;Narkis et al., 2007). PtdIns(4,5)P 2 directly participates in vesicle budding and endocytosis, likely in a dynamin-1-dependent manner (Saheki and De Camilli, 2012). This is in keeping with the phenotype of Pip5k1c KO mice, which have decreased PtdIns(4,5)P 2 levels in the brain and impaired synaptic vesicle dynamics, including slower rates of vesicle endocytosis and recycling (Di Paolo et al., 2004).
Another Pip5k1c KO mouse manifests neural tube closure defects, cardiac abnormalities and perinatal lethality within the first 24 h after birth and shows reduced ability to produce PtdIns(4,5)P 2 levels in the brain (Wang et al., 2007).
Interestingly, LCCS5 is caused by recessive mutations in the large GTPase DNM2 (MIM 615368) (Koutsopoulos et al., 2013). This observation might suggest that DNM2 acts in the same pathway as PIPK1γ, given that the pleckstrin homology (PH) domains in dynamin GTPases bind to PtdIns(4,5)P 2 (Barylko et al., 1998) and dynamins are required for fission of clathrin-coated vesicles (Saheki and De Camilli, 2012). It is plausible that DNM2 uses PtdIns(4,5)P 2 generated by PIPK1γ to finalize the fission step of clathrin-coated vesicles during synaptic vesicle formation and, when either gene is mutated, synaptic vesicles fail to form and/or recycle efficiently. Future studies should consider systems-based approaches to learn how these genes (and other genes unrelated to PIPs) interact in models of LCCS.
Neurological ciliopathies related to PIP2 metabolism Dent disease type 2 and Lowe syndrome (oculocerebrorenal syndrome) are allelic conditions caused by mutations in the X-linked gene OCRL, which encodes a PtdIns(4,5)P 2 5phosphatase. Lowe syndrome is characterized by ophthalmological findings, including congenital cataract, glaucoma or microphthalmia, and renal abnormalities, such as proximal tubulopathy, rickets and renal Fanconi syndrome (Lewis et al., 2001). Patients with Lowe syndrome have neonatal hypotonia, and most develop epilepsy and mental retardation (Lewis et al., 2001). The milder form of Lowe syndrome is Dent disease, which manifests primarily with renal findings, including low molecular weight proteinuria and hypercalciuria; patients also occasionally have cognitive and behavioral problems (Lewis et al., 2001;Hichri et al., 2011;Hoopes et al., 2005). Dent disease mutations generally result in a truncated protein that retains some catalytic activity (Shrimpton et al., 2009), whereas Lowe syndrome mutations are frequently characterized by loss of protein expression and/or catalytic activity. Modifier genes might also influence disease severity (Bökenkamp et al., 2009).
Much of our knowledge of OCRL function, as well as of the pathogenic mechanisms caused by OCRL mutations, has come from cell and animal models. OCRL localizes primarily to endosomes and the Golgi apparatus (Ungewickell et al., 2004;Vicinanza et al., 2011) as well as to primary cilia in retinal pigment epithelial cells and kidney tubular cells (Luo et al., 2012). Loss of OCRL function is associated with structural and functional abnormalities in the early endosome and with the impaired recycling of multiple receptors (Ungewickell et al., 2004;Vicinanza et al., 2011). Loss of OCRL also results in accumulation of PtdIns(4,5)P 2 (Zhang et al., 1998), specifically in the transition zone of primary cilia (Prosseda et al., 2017) and in endolysosomes (Festa et al., 2018). PtdIns(4,5)P 2 accumulation is hypothesized to impair endosomal trafficking to cause disease by increasing endosomal filamentous (F-) actin or by altering F-actin dynamics, which then affects multiple cellular processes (Vicinanza et al., 2011), including cilia formation and function. A transgenic Ocrl y/− ;Inpp5b −/− double KO mouse expressing human INPP5B was developed as a model of Lowe syndrome and Dent disease (Bothwell et al., 2011) and was used to confirm accumulated PtdIns(4,5)P 2 in endolysosomes, hyper-polymerization of F-actin and impaired receptor trafficking (Festa et al., 2018).
Interestingly, this transgenic model was developed because the first Ocrl KO mouse generated had no obvious phenotype because of functional compensation by INPP5B, another PtdIns(4,5)P 2 5phosphatase (Jänne et al., 1998). This was confirmed via the generation of Ocrl/Inpp5b double KO mice, which have a lethal embryonic phenotype (Jänne et al., 1998). Such compensation appears to depend on tissue-specific expression patterns of OCRL and INPP5B (Bernard and Nussbaum, 2010). Indeed, Luo and colleagues observed that OCRL and INPP5B have differential coexpression patterns between human and mouse eye tissue (Luo et al., 2013). Furthermore, they show that ocrl and inpp5b act synergistically in primary ciliogenesis, suggesting genetic compensation in zebrafish (Luo et al., 2013). Collectively, these data illustrate the possibility that OCRL-related disorders could be rescued by increasing levels of INPP5B, perhaps via AAV gene delivery or via drugs known to upregulate INPP5B expression in Lowe syndrome-affected tissues.
In zebrafish, mutation of ocrl causes a phenotype with many of the human disease features, including cystic brain lesions, seizures and kidney abnormalities (Ramirez et al., 2012). This model was subsequently found to have mild defects in cilia formation and impaired endocytic recycling (Oltrabella et al., 2015). Intriguingly, the ocrl mutants accumulate PtdIns(4,5)P 2 , which can be rescued through morpholino knockdown of the PIP 5-kinase pip5k1ab, suggesting that PIP5K could also be a therapeutic target for Lowe and Dent syndromes (Oltrabella et al., 2015).
The INPP5E mutations that cause Joubert syndrome diminish the 5-phosphatase activity of the protein (Bielas et al., 2009), whereas MORM syndrome mutations result in a protein with phosphatase activity but that abnormally localizes to cilia (Jacoby et al., 2009). Inpp5e homozygous null mice are perinatal lethal and display typical signs of a ciliopathy, including bilateral anophthalmos, postaxial hexadactyly and polycystic kidneys, as well as cerebral developmental defects, including anencephaly and exencephaly (Jacoby et al., 2009). Primary cilia from the renal cysts of these KO mice are abnormal in size and morphology (Jacoby et al., 2009).
When Inpp5e is conditionally inactivated in the mouse kidney epithelium, its loss causes severe polycystic kidney disease (Hakim et al., 2016) due to the loss of INPP5E 5-phosphatase-mediated hydrolysis of PtdIns(3,4,5)P 3 , which results in increased pAkt and the activation of the mTOR growth pathway. When mTOR was inhibited in this study with everolimus, the size and weight of kidneys in the Inpp5e conditional KO mouse were reduced. This corresponded with a 50% improvement in renal function compared to placebo, indicating that the combination of mTOR inhibition and PI3 kinase inhibition (to reduce pAkt) could be the most effective means of rescuing this phenotype.
Consistent with these findings, Dyson and colleagues recently demonstrated that expression of constitutively active smoothened (Smo) re-activated the SHH pathway in Inpp5e KO mice, rescuing many phenotypes . Conduit and colleagues expanded on these findings, showing that INPP5E hydrolyzes PtdIns(3,4,5)P 3 in the transition zone (TZ) of cilia, thereby decreasing pAkt inhibition of pGSKβ, promoting both the ciliogenesis and SHH signaling pathways. When INPP5E 5phosphatase activity is lost, PtdIns(3,4,5)P 3 accumulates in the TZ, resulting in increased pAKT inhibition of pGSKβ, leading to cilia loss. Given that cilia are required for proper SHH signaling, INPP5E loss disrupts the progression of SHH-dependent medulloblastoma . Importantly, PI3K inhibition prevents PtdIns(3,4,5)P 3 accumulation and restores TZ components to the cilia . Thus, although PI3K inhibition is a potential therapeutic target for ciliopathy in Joubert syndrome, it could have a detrimental effect in the context of SHH-driven medulloblastomas by sustaining ciliogenesis and driving proliferation.
It is important to note that synaptojanin is highly expressed in the brain, as compared to OCRL, which is ubiquitously expressed in all tissues. Synaptojanin 1 also contains a polyphosphatase SAC1-like domain that is essential for synaptic vesicle endocytosis in neurons and is thus essential for the maintenance of neuronal transmission (Clayton et al., 2013). Synj1 KO mice die shortly after birth and exhibit numerous neurological defects, including severe weakness, ataxia and convulsions (Cremona et al., 1999;Clayton et al., 2013). Recently, Fasano and colleagues showed that SYNJ1 regulates the morphology of early endosomes (EE) and plays an important role in trafficking and recycling of proteins through the EE compartment of neuronal cells (Fasano et al., 2018). The authors propose that defects in the endocytic compartment provide insight into the mechanisms governing altered neuronal plasticity and/or the loss of neuronal viability seen in Parkinson's disease type 20.
Interestingly, mutations in SYNJ1 also result in a subtype of early infantile epileptic encephalopathy (EIEE; MIM 617389). The mouse phenotype thus appears to recapitulate the parkinsonian and epileptic features of human disease. In addition, in mouse models of Down syndrome, SYNJ1 overexpression contributed to enlarged EE and brain dysfunction (Cossec et al., 2012). This finding makes SYNJ1 an interesting potential target for the treatment of Down syndrome (Cossec et al., 2012).
Congenital muscular dystrophy and early-onset cataracts
Several reports have linked biallelic mutations in INPP5K to congenital muscular dystrophy with cataracts and intellectual disability (MIM 617404). INPP5K is a PtdIns(4,5)P 2 / PtdIns(3,4,5)P 3 5-phosphatase, preferring PtdIns(4,5)P 2 as a substrate (Dong et al., 2018). Wiessner and colleagues reported on eight families with 12 affected individuals with hypotonia during infancy, cataract onset between six months and six years of age, and delayed motor development (Wiessner et al., 2017). These patients also had a mild intellectual disability but were without facial or oculomotor weakness or cardiac involvement. Creatine kinase levels were high in affected individuals, between four to fourteen times the upper limit of normal. And nine patients had non-specific muscle pathology, showing variable degrees of dystrophic features. In three patients, muscle biopsies also showed vacuolated muscle fibers. The four INPP5K mutations reported in these eight families were all missense mutations that did not appear to affect protein level upon overexpression studies in COS-7 cells; however, phosphatase activity was reduced (Wiessner et al., 2017).
Osborn and colleagues independently reported on four additional families with five affected individuals with biallelic INPP5K mutations (Osborn et al., 2017). They all presented during infancy or in early childhood with muscle weakness and motor delay, and exhibited moderate to severe intellectual disability. Cataracts were noted in three patients, and brain MRI showed normal cerebellum and white matter. Creatine kinase was consistently high (>1000 U/L). The reported mutations included four missense that were confirmed to reduce the phosphatase catalytic activity of INPP5K. However, the C-terminal frameshift deletion p. Asn417Lysfs*26, which is located outside of the phosphatase domain and reported in one patient, did not significantly affect the catalytic activity; protein levels in patients or cell models were not investigated (Osborn et al., 2017).
INPP5K dysfunction has been modeled in zebrafish using morpholino-mediated knockdown of one or both zebrafish paralogs (inpp5ka and inpp5kb) (Osborn et al., 2017). The resulting inpp5ka morphants showed microphthalmia, microcephaly, curled and shortened bodies, severely impaired swimming and an abnormal touch-evoked escape response. Histological findings in skeletal muscle showed abnormal myosepta, gross myofiber disorganization and reduced neuromuscular junction arborization, suggestive of poor synaptic formation. Electron microscopy revealed less compact myofibrils, short sarcomeres, as well as undefined A-and I-bands, and reduced triad size. A recent study has uncovered a role for INPP5K in the regulation of ER morphology whereupon loss of INPP5K results in expansion of ER sheets (Dong et al., 2018). It remains to be determined whether this mechanism is involved in the onset of this form of congenital muscular dystrophy; however, it is an intriguing possibility given the importance of specialized ER, i.e. sarcoplasmic reticulum, in skeletal muscle function. As for the intellectual disabilities seen in patients, Dong and colleagues point out that these are consistent with the function of INPP5K in neurons (Dong et al., 2018).
Disorders of PIP acyl chain composition
An example of neurologic disease related indirectly to PtdIns(4,5)P 2 comes from MBOAT7 mutations, which cause autosomal recessive mental retardation 57 (MIM 617188). MBOAT7 encodes a lysophosphatidylinositol acyltransferase (LPIAT) that remodels the fatty acid composition of PtdIns in a process called the Lands cycle (Lee et al., 2008. Mboat7 null mice die prematurely and display a smaller cerebral cortex and hippocampus with lamination defects . This study further showed that loss of LPIAT activity reduces the amount of polyunsaturated fatty acid, such as arachidonic acid, that is incorporated into the sn-2 position of PtdIns, increasing the saturated fatty acid composition in phosphorylated derivatives, such as PtdIns4P and PtdIns(4,5)P 2 . This study highlights that fatty acid composition might influence PIP localization and/or their interaction with PIPbinding proteins. An improved understanding of how changing the fatty acid composition of PIPs can modify models of PIP disorders will further our understanding of PIP biology and perhaps lead to the discovery of novel targets for therapeutic intervention.
PI4K-related disorders
Dysregulation of the mono-phosphorylated PIP PtdIns4P is also implicated in neurologic disease. Mutations in the PI4KA gene, which encodes a PI4 kinase type IIIα, cause PMGYCHA (MIM 616531), a disorder characterized by bilateral perisylvian polymicrogyria, cerebellar hypoplasia, arthrogryposis and variable contractures (Pagnamenta et al., 2015). PI4KA generates PtdIns4P from PtdIns, and PtdIns4P serves as a precursor to maintain plasma membrane PtdIns(4,5)P 2 levels (Bojjireddy et al., 2014;Tan et al., 2014). A recent report has provided insight into the pathogenic mechanisms underlying the PI4KA-related central nervous system abnormalities. When Alvarez-Prats and colleagues specifically deleted Pi4ka from Schwann cells in mice, the resulting mutants had severe myelination defects (Alvarez-Prats et al., 2018). It also resulted in the depletion of plasma membrane PtdIns4P, but no decrease in PtdIns(4,5)P 2 . However, plasma membrane phosphatidylserine, phosphatidylethanolamine and sphingomyelin were significantly decreased as well, suggesting that the regulation of PtdIns4P by PI4KA is important for maintaining levels of other lipid species at the plasma membrane, which are essential for the proper myelination of Schwann cells. It remains unclear, however, how impaired PI4KA alters brain development and neuronal migration in order to cause polymicrogyria.
Interestingly, a mouse KO for type II PI4K, Pi4k2a, develops late-onset features resembling hereditary spastic paraplegia (HSP), with severe axonal degeneration in the spinal cord that is associated with tremors, high-frequency nodding, spastic gait, limb weakness and urinary incontinence (Simons et al., 2009). Several genes associated with HSP are related to axon transport, and the phenotype of Pi4k2a mice is consistent with defective axon transport. Indeed, the inhibition of PI4K activity disrupts the retrograde axonal transport of neurotrophins, and the acute effects of PI4KIIa knockdown or inhibition include disrupted intracellular trafficking and phosphoinositide signaling (Bartlett et al., 2002;Minogue, 2006). Given the importance of PIPs in membrane trafficking, their involvement in HSP-like pathology is unsurprising. The effects of Pi4k2a deficiency in mice may result from a direct deficit of PtdIns4P or its PtdIns(4,5)P 2 derivative. Nevertheless, as yet, no PI4K2A mutations have been found in patients with HSP, so the relevance of these findings to human disease is unknown.
Neurological disorders of genes with PIP-binding domains
All PIP metabolic disorders can be considered, at least in part, to be disorders of PIP-binding proteins, given that the loss of PIP kinases and phosphatases is associated with changes in absolute and local PIP levels, thereby disrupting the localization and functions of myriad PIP-binding effectors. There are as many as 70 neurological disorders that result from mutations in genes containing a PIPbinding domain (Table S1); here, we present a few notable examples that provide insights into the function of specific PIPs in disease.
Loss-of-function mutations in C2 domain-containing phospholipase C beta 1 (PLCB1), which cleaves PtdIns(4,5)P 2 into DAG and IP 3 (McLaughlin et al., 2002), and in PH domain-containing dynamin 1 (DNM1), which binds to PtdIns(4,5)P 2 during membrane fission events (Barylko et al., 1998), cause two similar subtypes of EIEE (MIM 613722 and MIM 616346, respectively) (Kurian et al., 2010;EuroEPINOMICS-RES Consortium et al., 2014). The EEIE phenotype is also associated with mutations in SYNJ1, as discussed earlier. The similar presentations might reflect a common pathway involving these genes in the etiology of EIEE. Indeed, a large-scale survey of causative mutations in EIEE suggested synaptic dysregulation as a common underlying mechanism (EuroEPINOMICS- RES Consortium et al., 2014); however, this hypothesis has yet to be functionally validated for the majority of these genes. This disorder might also provide a unique example of how a disease of a PIPmetabolizing enzyme causes abnormal PIP-binding effector activity, or vice versa. It is worth mentioning here that another PH domaincontaining protein that binds PtdIns3P, collybistin (ARHGEF9), is also mutated in EIEE (Alber et al., 2017;Wang et al., 2018). However, patients with missense mutations in the PH domain do not develop seizures, whereas patients with missense mutations affecting other regions of the gene have epilepsy (Alber et al., 2017;Wang et al., 2018), suggesting that PIP binding may not have a critical role in the manifestation of this subtype of EIEE.
Protrudin localizes to the ER, where it mediates ER-to-lysosome and late endosome (LyLE) contact sites by binding to PtdIns3P on LyLEs (Raiborg et al., 2015;Hong et al., 2017). This contact allows transfer of kinesin-1 motor proteins to the LyLEs, which translocates mTOR-and synaptotagmin VII-containing LyLEs along microtubules to the cell periphery (Raiborg et al., 2015;Hong et al., 2017). This maintains mTOR in its activated state to attenuate autophagy and also promotes fusion with the plasma membrane to promote neurite outgrowth (Raiborg et al., 2015;Hong et al., 2017). Interestingly, mutations in kinesin-1 (KIF5A) also cause a spastic paraplegia (MIM 604187) (Ebbing et al., 2008), as do mutations in the gene encoding the microtubule-severing protein spastin (SPAST) (MIM 182601) (Svenson et al., 2001;Roll-Mecak and Vale, 2008), consistent with the proposed mechanism of action for protrudin by Raiborg and colleagues (Raiborg et al., 2015). This is a potential example of how a common pathway can mechanistically unify genetically heterogeneous disorders. It will thus be interesting to explore potential crosstalk between spastizin and the protrudin pathway given that they regulate similar autophagic processes and may involve a common pool of PtdIns3P.
In recent years, there has been increasing recognition of the role of somatic mutations in certain neurodevelopmental syndromes. Somatic mutations occur in several diseases associated with epilepsy, autism spectrum disorders and intellectual disability (Poduri et al., 2013). Recent advances in the genomic sequencing and bioinformatic analyses of these variants has allowed us to more accurately estimate the contributions of such phenomena to human neurological disorders (Evrony et al., 2016).
Mutations in the PI3K/Akt pathway are known to cause somatic neurologic syndromes. Postzygotic activating mosaic mutations of PIK3CA have been associated with overgrowth syndromes, including MCAP (megalencephaly-capillary malformationpolymicrogyria syndrome; MIM 603387) (Riviere et al., 2012) and CLOVES syndrome (congenital lipomatous overgrowth, vascular malformations and epidermal nevi; MIM 612918) (Kurek et al., 2012). Both conditions are characterized by increased birth weight and height, asymmetric limb hemihypertrophy and vascular malformations. Patients with MCAP also manifest megalencephaly, brain malformation, significant developmental delay and seizures, whereas CLOVES patients have significant skeletal abnormalities, such as megaspondylodysplasia, or hyperostosis of the skull, in addition to skin manifestations of lipomas or linear epidermal nevus (Kurek et al., 2012;Lindhurst et al., 2012;Riviere et al., 2012). Recently, 19 patients with CLOVES were treated with the PIK3CA inhibitor BYL719 and all showed marked improvement in disease symptoms (Venot et al., 2018). Germline mutations of the PI3K regulatory subunit PIK3R2, as well as the predominant AKT serine/ threonine kinase isoform in the brain AKT3, also cause MCAP (Riviere et al., 2012).
Of note, somatically acquired mutations of PIK3CA are reported in many cancers, including breast (MIM 114480), colorectal (MIM 114500), gastric (MIM 613659), hepatocellular (MIM 114550) and ovarian (MIM 167000) ( Table 1). The importance of PI3K in cancer is underscored by the anticancer effect of drugs that inhibit downstream effectors of PI3K/Akt signaling, such as rapamycin, an mTOR inhibitor (Yuan and Cantley, 2008). There are many excellent reviews that discuss in detail the role of the PI3K/Akt pathway and mTOR in cancers and neurodevelopmental disorders (Fresno Vara et al., 2004;Fruman and Rommel, 2014;Mayer and Arteaga, 2016).
Conclusion
A growing number of neurological disorders are associated with mutations in genes that encode PIP-metabolizing enzymes and PIP-binding proteins, underscoring the importance of these lowabundance phospholipids for health and disease. Studying these diseases, particularly in cell and animal models, has identified crucially important roles for PIPs and the enzymes that regulate them in organismal biology, including their participation in key processes, such as synaptic vesicle recycling, skeletal muscle T-tubule biogenesis and maintenance, peripheral myelination, and cilia formation and signaling. These studies have also unveiled new therapeutic avenues and treatments for several neurological disorders.
Nevertheless, there is still much to learn about the regulation of PIPs in many neurological disorders. PIPs define organelle identity, acting as molecular signposts for many effector proteins and molecular switches, such as Rab GTPases, to coordinate membrane and protein traffic throughout the cell (Di Paolo and De Camilli, 2006;Ketel et al., 2016). Given this role in regulating distinct processes via effector proteins, PIPs can be regarded as being pleiotropic in nature. This pleiotropy poses one of the major challenges in the field of PIP disorders: determining the primary causal mechanism(s) of disease by demonstrating a causal relationship between one or multiple processes that are regulated by the PIP and the onset of the disorder. For example, we know that loss of MTM1 in XLMTM results in PtdIns3P accumulation, leading to aberrant receptor trafficking and recycling through endosomal compartments (Velichkova et al., 2010;Ribeiro et al., 2011;Sabha et al., 2016;Ketel et al., 2016), and that these changes correlate with disease severity. However, we still do not know how many PtdIns3P effector(s) behave differently in this disease, or how many protein(s) are trafficked improperly, and whether or not these changes cause the observed disease phenotype. To complicate matters, this assumes that the sole function of MTM1 is to dephosphorylate PtdIns3P, which is certainly not the case given its phosphatase-independent functions (Amoasii et al., 2012) and role in the ubiquitin-proteasome system (Gavriilidis et al., 2018).
One potentially universal strategy to elucidate PIP-dependent membrane and protein traffic processes would be to perform subcellular proteomics in inducible KO models. Subcellular proteomics can be performed in a variety of ways, including, but not limited to, techniques such as TurboID (and its predecessor BioID; Fig. 3A,B) (Branon et al., 2018;Roux et al., 2013) or subcellular fractionation profiling followed by mass spectrometry (Borner et al., 2014;Itzhak et al., 2016;Christoforou et al., 2016). Inducible KOs can also be achieved in a variety of ways, such as tamoxifen-or tetracycline-driven and heat-shock-inducible Cre/Lox recombination (Fig. 3C) (Imai et al., 2001;Bäckman et al., 2009;Le et al., 2007). Increasingly sophisticated methods of gene perturbation with modified CRISPR/Cas9 technologies allow for rapid and tunable gene editing (Senturk et al., 2017). The combination of these methods would allow researchers to determine the spatiotemporal changes in local proteomes that occur before and after a given PIP is modulated (by kinases or phosphatases; Fig. 3) or when a given PIP-binding protein is lost. In keeping with the MTM1 example, this would involve performing proteomics on specific endosomal populations, myonuclei and/or the sarcolemma, before and after MTM1 is inducibly knocked out. This should, hypothetically, provide rich information about how PtdIns3P [or PtdIns(3,5)P 2 ] effectors behave when MTM1 is lost, and a detailed timeline of how pathologic changes begin.
Alternatively, a powerful system that facilitates spatiotemporal manipulation of PIPs was first published over a decade ago by two independent groups (Varnai et al., 2006;Fili et al., 2006). This system uses rapamycin (or its analogs, rapalogs) to recruit PIP enzymes fused to the FKBP12-rapamycin binding (FRB) domain of mTOR [or the rapamycin intracellular receptor FK506 binding protein (FKBP)] to specific subcellular compartments labeled with a well-established marker fused to FKBP (or FRB) via FRB-FKBP heterodimerization (Fig. 3C) (Inobe and Nukina, 2016). This system has since been used to temporally manipulate PIPs, in particular PtdIns(4,5)P 2 and PtdIns4P, with great success (Szentpetery et al., 2010;Ufret-Vincenty et al., 2015;Gulyás et al., 2017). However, this technique still awaits widespread adoption, particularly to study PIP metabolism disorders. Future studies interested in directly linking PIP metabolism or PIP binding by an effector protein to a disease mechanism may benefit from adopting this powerful yet underutilized approach.
Applying unbiased techniques that capture a global view of the protein localization changes that occur when a PIP is modulated would undoubtedly improve our understanding of PIP biology and its role in disease. We also recommend considering more bioinformatic, systems biology approaches to identify a common pathway for genes that cause the same disorder because, as we have learned, many neurological disorders are genetically heterogeneous. Earlier, we presented an elegant example by Raiborg and colleagues in which they implicated several genes that cause spastic paraplegia in a common pathway (Raiborg et al., 2015). Similarly, the identification of DNM2 and BIN1 as genetic modulators of MTM1 Lionello et al., 2019), all of which cause a centronuclear myopathy, indicates intimate crosstalk among genes involved in the same (or similar) disorders. We recommend that future research on genetically heterogeneous disorders (e.g. Charcot-Marie-Tooth syndromes, epileptic encephalopathies, spinocerebellar ataxias) consider this theme when trying to build a mechanistic model of disease pathogenesis.
On this note, it is worthwhile mentioning how the disruption of other PtdIns derivatives results in neurological diseases. Mutations in PtdIns glycan anchor biosynthesis (PIG) genes such as PIGP (MIM 617599) and PIGA (MIM 300868) can cause early infantile epileptic encephalopathy ( (MIM 280000), PIGY mutations cause hyperphosphatasia with mental retardation syndrome 6 (MIM 616809), whereas those in post-GPI attachment to proteins 1 (PGAP1) cause mental retardation 42 (MIM 615802) ( Table 2). Most recently, Nguyen and colleagues showed individuals with mutations in GPAA1, a GPI transamidase complex protein, present with developmental delay, hypotonia, earlyonset seizures, cerebellar atrophy and osteopenia (MIM 617810) . Given that many PtdIns/PIP kinases and PIP phosphatases phenotypically overlap in these conditions, it is tempting to infer an intimate crosstalk between the phosphorylated and glycoslyated derivatives of PtdIns. Indeed, mutations in PLCB1, DNM1, ARHGEF9 and SYNJ1 cause EIEE, and mutations in MBOAT7 as well as 13 genes with PIP-binding domains (ARHGEF6, CASK, CC2D1A, CNKSR2, COL4A3BP, DLG3, EPB41L1, FGD1, IQSEC2, KIF1A, OPHN1, SYNGAP1, TRIO) cause mental retardation (Table S1). It is possible that a glycosylation defect in PtdIns impinges on PIP-related pathways given that GPI proteins anchor approximately one in every 200 mammalian proteins to the cell membrane . These examples further emphasize the global importance of PtdIns metabolism in normal brain development and represent additional areas for future research.
With the more widespread application of techniques, such as whole-exome sequencing in clinical genetics, it is unsurprising that new associations between human diseases and mutations in PIP enzymes are emerging. Still, as many as 636 PIP-modulating or -binding proteins have not, as of yet, been linked with human genetic disorders (Table S1), indicating that they are either indispensable for early embryonic development or functionally redundant. Lastly, because of their enzymatic properties, kinase and phosphatase regulators of PIP metabolism are attractive therapeutic targets for neurological disease. This has been demonstrated for human cancers, for which PI3 kinase inhibitors have long been considered as potential treatments. In the future, inhibitors or activators of the PIP enzymes and PIP-binding proteins associated with neurological disease are likely to emerge as viable therapeutics. Fig. 3. Potential experimental approaches to interrogate pleiotropy in phosphoinositide biology. PIPs and their effectors have pleiotropic functions. Elucidating these will require methods that capture global changes that occur when PIPs are altered. This figure shows the complex protein networks that interact with PIPs and illustrates the possible combination of BioID screening with inducible protein expression/ localization methods to investigate how PIP kinase mutants or PIP-binding protein mutants acquire differential subcellular protein interactomes. (A) A Rab GTPase can be used as an organelle marker that can be fused with the biotin protein ligase BirA, which biotinylates nearby endogenous PIP-interacting proteins, to provide subcellular specificity to a differential interactome. This interactome, mapped based on biotinylation (yellow circle), is different between wild-type cells (upper panel) and cells that do not express, for example, a PIP kinase (lower panel). (B) If BirA is fused to a PIP-binding protein, mutations in a PIP-binding protein (red circle) alter the interactome. This results in a BioID screening output showing biotinylation of PIPindependent interactors, whereas the PIP-containing compartment will lose or have reduced biotinylation. Although useful as an organelle marker, fusing Rab to BirA may result in BioID screens identifying non-specific transient interactions, such as those with the Rab GDP dissociation inhibitor GDI. These methods could be extended upon with artificial regulation of protein complex formation, such as the rapamycin-or rapalog-induced dimerization between FKBP-and FRB-fused proteins (Inobe and Nukina, 2016), and with inducible gene editing to increase spatiotemporal control. (C) Concomitant heat-shockinduced expression of the Rab-FRB fusion, the BirA-FKBP fusion and Cre recombinase, and addition of a rapalog to induce FKBP-FRB dimerization, allows for simultaneous control over protein biotinylation and Cre-induced deletion of a PIP kinase. This would allow for identification of differential subcellular protein interactomes before and shortly after knockout of a gene (in this example, a PIP kinase).
==
Domain: Biology Medicine
|
[{"added": "2019-08-16T13:04:04.344Z", "created": "2019-08-01T00:00:00.000", "id": "199662818", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://dmm.biologists.org/content/dmm/12/8/dmm038174.full.pdf", "pdf_hash": "ce531d6b47aeed50e75f0eebf3105be35d8ebd3b", "pdf_src": "Highwire", "provenance": "peS2o-0075.json.gz:32153", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "603e2e8b5bfd1184b96e0445b1f37b15c82e4fd6", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Replication and Predictive Value of SNPs Associated with Melanoma and Pigmentation Traits in a Southern European Case-Control Study
Background Genetic association studies have revealed numerous polymorphisms conferring susceptibility to melanoma. We aimed to replicate previously discovered melanoma-associated single-nucleotide polymorphisms (SNPs) in a Greek case-control population, and examine their predictive value. Methods Based on a field synopsis of genetic variants of melanoma (MelGene), we genotyped 284 patients and 284 controls at 34 melanoma-associated SNPs of which 19 derived from GWAS. We tested each one of the 33 SNPs passing quality control for association with melanoma both with and without accounting for the presence of well-established phenotypic risk factors. We compared the risk allele frequencies between the Greek population and the HapMap CEU sample. Finally, we evaluated the predictive ability of the replicated SNPs. Results Risk allele frequencies were significantly lower compared to the HapMap CEU for eight SNPs (rs16891982 – SLC45A2, rs12203592 – IRF4, rs258322 – CDK10, rs1805007 – MC1R, rs1805008 - MC1R, rs910873 - PIGU, rs17305573- PIGU, and rs1885120 - MTAP) and higher for one SNP (rs6001027 – PLA2G6) indicating a different profile of genetic susceptibility in the studied population. Previously identified effect estimates modestly correlated with those found in our population (r = 0.72, P<0.0001). The strongest associations were observed for rs401681-T in CLPTM1L (odds ratio [OR] 1.60, 95% CI 1.22–2.10; P = 0.001), rs16891982-C in SCL45A2 (OR 0.51, 95% CI 0.34–0.76; P = 0.001), and rs1805007-T in MC1R (OR 4.38, 95% CI 2.03–9.43; P = 2×10−5). Nominally statistically significant associations were seen also for another 5 variants (rs258322-T in CDK10, rs1805005-T in MC1R, rs1885120-C in MYH7B, rs2218220-T in MTAP and rs4911442-G in the ASIP region). The addition of all SNPs with nominal significance to a clinical non-genetic model did not substantially improve melanoma risk prediction (AUC for clinical model 83.3% versus 83.9%, p = 0.66). Conclusion Overall, our study has validated genetic variants that are likely to contribute to melanoma susceptibility in the Greek population.
Introduction
Plethora of studies has shown that ultra-violet (UV) light exposure and certain phenotypic traits, i.e. red or blonde hair, light-colored eyes, fair skin complexion, and prominent mole pattern are major risk factors for the development of cutaneous melanoma (CM) [1][2][3][4][5][6]. A strong genetic background has been supported by twin studies showing a 55% contribution of genetic effects in melanoma variation liability [7].
High-penetrance germline mutations in CDKN2A and CDK4 genes are rare (0.2-1.2%) in sporadic CM, but they are encountered in approximately 5% of families with only two members with CM, and in 30-40% of families with 3 or more affected members [8][9][10]. The advent of high-throughput genotyping technologies and their utilization in population-based studies has discovered a considerable number of rare and common genetic variants at different genetic loci associated with melanoma. The most prevalent low penetrance locus is the melanocortin 1 receptor gene (MC1R), whose variants have been associated both with melanoma [11][12][13][14][15][16][17][18] as well as with related traits [11,[19][20][21]. Apart from MC1R, a significant number of low penetrance genes involved in various cellular pathways, such as pigmentation, cell cycle control, DNA repair, oxidation stress, apoptosis, senescence and melanocyte differentiation and migration have been implicated in melanoma susceptibility [22]. A detailed synopsis and meta-analysis of reported melanoma-associated variants is available in MelGene, an on-line database ( [URL]) [23]. In addition to common variants, a rare germline variant in MITF (rs14917956 -E318K) that alters MITF transcriptional activity was recently found to be associated with melanoma and renal cell cancer [24][25].
Most genetic association studies on CM have been performed in populations with fair skin and, hence, the effect of melanomaassociated variants in relatively darker skin populations residing in areas of higher ambient UV-exposure is less well known. Being a southern European country, Greece is characterized by a high degree of sun exposure year-round, a population of relatively darker skin complexion compared to northern European countries and the lowest incidence of melanoma (4-5 per 100,000 personyears) among European countries [26][27][28]. Mutational analyses performed by our group in Greek patients with sporadic and genetically enriched melanoma, found a higher prevalence of CDKN2A/CDK4 mutations than previously reported, suggesting a more prominent role of genetic susceptibility to melanoma in regions with a relatively low incidence of melanoma [28][29]. In the present study, we sought to replicate the most prominent results of MelGene and other findings from genome-wide association studies (GWAS) in a Greek case-control study. Our research replicates a number of variants that are associated with melanoma risk in the Greek population and the relevant pathogenetic pathways that are involved; it also highlights differences in risk allele frequencies among the Greek population and the HapMap European sample concerning mainly pigmentation-related risk loci. Finally, it provides insights about the predictive value of identified genetic risk factors compared to wellestablished clinical ones.
Study Population
The study population consisted of Greek melanoma cases and control subjects, above 18 years of age. The case sample consisted of patients diagnosed with non-familial, histologically confirmed invasive melanoma at A. Sygros Hospital, a large referral center of melanoma in Athens, and participating melanoma centers, from 2003 to 2009. The control sample included blood donors from a blood donation center and individuals with minor skin diseases attending A. Sygros Hospital. Controls were matched 1:1 on age (+/22 years) and gender to the cases. Individuals with a history of melanoma, other types of skin cancer, or any non-dermatological malignancy were excluded from the control arm of the study.
Each subject was interviewed and examined by a dermatologist or trained physician and information was retrieved on demographic variables (age, sex), pigmentation traits (eye, hair, and skin color), phototype, and sun exposure variables (sunburns, tanning). The Declaration of Helsinki protocols were followed and the Scientific and Ethics Committee of Andreas Sygros Hospital has reviewed and approved the research protocol; all participating individuals gave written informed consent.
SNP selection and Genotyping
All variants included in this study were selected from the last update of the MelGene field synopsis (October 2011), a large online database that was created with the purpose of comprehensively collecting and meta-analyzing all published genetic associations of melanoma ( [URL]) [23]. More specifically, the 34 selected variants from MelGene were distinguished in two groups: 1) all variants associated with melanoma at a level of p,0.05 following meta-analysis of relevant data from at least 3 independent case-control datasets (28 variants) and 2) additional biologically plausible variants representing potential causal pathways and selected from GWAS (3 variants) and candidate gene studies (3 variants) with genome-wide (p,10 27 ) or nominally significant (p,0.05) associations. These variants were also included in MelGene but not necessarily metaanalyzed due to insufficient number of available datasets. In all, of the 34 variants, 19 had reached genome-wide significance in a previous GWAS or in MelGene, and the other 15 had not.
DNA isolation, Genotyping and Quality control
Genomic DNA was isolated from peripheral blood using the QIAamp DNA blood mini kit (Qiagen). DNA concentration was quantified in samples prior to genotyping by using Quant-iT dsDNA HS Assay kit (Invitrogen). The concentration of the DNA was adjusted to 5 ng/ml.
A total of 50 ng from each DNA sample were used to genotype the selected 34 SNPs using the Sequenom iPLEX assay (Sequenom, Hamburg, Germany). Allele detection in this assay was performed using matrix-assisted laser desorption/ionizationtime-of-flight mass spectrometry [30].
Our quality control criteria included the inclusion of SNPs with a genotype call rate of 95% or higher, as well as SNPs showing no deviation from Hardy-Weinberg equilibrium (HWE) in the controls using a chi-squared test (P.0.05).
Statistical Analysis
We examined the association of each SNP with CM by performing conditional logistic regression analyses assuming a multiplicative model of inheritance considering the minor allele as the reference allele. To control for the effect of the other covariates/risk factors on CM in the Greek population, each SNP was subsequently incorporated into multivariable logistic regression models using a stepwise variable selection approach. The covariates considered were eye color (light: blue, green/gray and light brown or dark: dark brown and black), hair color (light: blond/red and light brown or dark: dark brown and black), skin color (light: fair/pale and light brown or dark: dark brown), phototype (type I, II, III or IV, according to the Fitzpatrick scale), tanning ability (burn, minimal tan, burn then tan or deep tan), and sunburn (presence or absence). We estimated odds ratios and 95% confidence intervals (95% CI) for all models. Missing values for any of the non-genetic risk factors were imputed using multiple imputation methods. Variables where all the required information was available were used for the construction of the models for the estimation of the imputed missing values.
Additionally, we estimated the correlation of risk allele frequencies between the HapMap CEU sample and the Greek population across all the evaluated SNPs. Moreover, we estimated the correlation of the effect sizes found in the Greek population with those found previously in the original publications or MelGene dependent on the source of SNP selection. We examined whether the direction of the effect estimates was in the same or in opposite directions. For the sample size of our study, we estimated the power G i to detect each of the previously described effects at a = 0.05 level given the observed minor allele frequency in the Greek studied population assuming a multiplicative (per-allele) genetic model. We used the QUANTO software ( [URL], accessed 30 September 2012). The sum of the power estimates corresponds to the number of variants that would be expected to replicate. Subsequently we calculated the binomial test for the expected vs. the replicated variants across all evaluated SNPs.
Finally, we created receiver operating characteristic (ROC) curves to assess the predictive ability of the CM-associated SNPs. We considered 3 models including, respectively, the phenotypic traits alone (model 1); the phenotypic traits along with the SNPs that remain statistically significant after Bonferroni correction (model 2); and the phenotypic traits along with all nominally statistically significant SNPs (model 3). In order to assess the validity of our models, we used k-fold cross-validation with k = 2 splits and 1,000 replications.
All statistical analyses were performed in STATA version 11.2 (College Station, TX, USA). All P-values are two-tailed.
Results
Our sample included 284 patients with CM matched on age and sex to 284 controls; of those, 270 (48%) were men. Median age was 44 years (range 18-85) for patients and 42 years (range 18-81) for controls. Demographics and phenotypic traits are shown in Table S1. Missing values in phenotypic characteristics were due to the fact that blood samples and questionnaires in one participating center were collected in the early phase of this study, and the corresponding individuals could not be found in order to retrieve these data. A total of 34 variants were selected for genotype analysis ( Table 1). All of them were successfully genotyped with call rates of 95% or above. Deviation from HWE in the control population was noticed for one singlenucleotide polymorphism (SNP) (rs4636294), which was subsequently excluded from further statistical analyses.
From the selected variants, four SNPs are found in the 39-UTRs and one in the 59-UTR of the respective gene loci; 13 are located in introns; and 10 are within exons. The remaining 6 variants are found in intergenic positions. We found evidence for strong pairwise LD (r 2 .0.85) between rs2218220 and rs4636294 (r 2 = 0.95), which deviated from HWE; rs10757257 and rs1335510 (r 2 = 0.96); rs1393350 and rs1126809 (r 2 = 0.94); and rs1885120, rs910873 and rs17305573 (r 2 = 0.90). For the remaining, moderate LD was observed (r 2 ,0.60). Table 1 shows the 33 analyzed SNPs, their effect sizes, minor and major alleles and the corresponding frequencies in the Greek population. All alleles identified as minor in the Greek population were also minor alleles in the CEU HapMap sample with one exception (rs6001027 whose minor allele was T in the Greek population but C in HapMap CEU). Figure 1 shows the correlation between the ORs identified for the 33 eligible SNPs in the Greek population and in the original source where these were selected. We noticed overall modestly high correlation of the respective effect estimates (r = 0.72, P,0.0001). No differences in ORs between the Greek population and the original source were beyond chance (i.e. 95% CI between the two populations showed overlap for each SNP). Overall, no nominally significant difference in ORs was noticed across all SNPs in the two populations (P = 0.411 for Mann-Whitney U). Reference source = Melgene: nominal association with melanoma after meta-analysis of data for this variant derived from at least 3 datasets (MelGene is an online database of all reported genetic associations of melanoma which includes a systematic meta-analysis of melanoma-associated variants from published datasets and grading of this associations for strength of epidemiogical evidence) [23], or data derived from original study (variants not metaanalyzed in Melgene). 2 Showed deviation from HWE, and was therefore not included in the analyses: N/A for MAF & OR in the Greek sample. 3 All individuals were homozygous for the major allele. Abbreviations: MAF, minor allele frequency; CI, confidence interval; OR, odds ratio. When limited to SNPs that had previously reached genomewide significance in either Melgene or a previous GWAS, the correlation of effect sizes was r = 0.83 (P,0.0001) and the correlation of risk allele frequencies was r = 0.98 (P,0.0001).
Association of variants with CM risk
Conversely, for the 14 SNPs that had not previously reached genome-wide significance, the respective correlation coefficients were r = 0.24 (P = 0.43) and r = 0.72 (P = 0.003). MelGene status = Data from MelGene, an online database of reported genetic associations of melanoma including a systematic meta-analysis of melanoma-associated variants from published datasets and grading of these associations for strength of epidemiogical evidence [23]. OR (95% CI) and p value correspond to nominal association with melanoma after meta-analysis of data for each variant.
Five SNPs were significantly associated with CM in the multivariable analyses after controlling also for hair color, skin color, eye color, phototype, sunburn and tanning ( Table 2).
Power Considerations
The power of our study to detect ORs similar to those previously found, given the allele frequencies observed in the Greek population, ranges from 5.2% for rs12203592 to 100% for rs16891982 at a = 0.05. By summing the power estimates for all SNPs to detect the respective ORs seen previously, we estimated that if ORs were identical in the Greek population, our study would be expected to have found 8 nominally statistically significant associations among the 33 tested. Among the 18 variants that had been previously identified with genome-wide significance and did not show deviation from HWE, our study would be expected to have found 6 nominally statistically significant associations and 7 were indeed nominally significant.
Comparison of risk allele frequencies between Greek sample and HapMap CEU
For 20 SNPs, the respective minor alleles were the risk alleles for melanoma. Table 3 shows risk alleles in the Greek sample and their frequency in both the Greek sample and HapMap CEU. The risk alleles in the Greek population had a median frequency of 20% (IQR, 4-60%), while their median frequency in HapMap CEU was 32% (IQR, 12-62%) (P = 0.243 for Mann-Whitney U). The correlation between the two populations was very high (r = 0.95, P,0.0001) (Fig. 2).
The risk allele frequencies of nine SNPs (rs6001027-C, rs16891982-G, rs12203592-T, rs258322-T, rs1805007-T, rs1805008-T, rs910873-A, rs17305573-C, and rs1885120-C) were different beyond chance between the Greek sample and HapMap CEU (i.e. 95% CI of risk allele frequencies in the Greek population and the HapMap sample did not overlap). All these variants (except for rs6001027, a nevi-related SNP in PLA2G6) had significantly lower frequencies of risk alleles in the Greek population compared to HapMap CEU, while six of those are variants of genes with well-established role in the genetic control of pigmentation (rs16891982 in SCL45A2, rs12203592 in IRF4, rs258322 in CDK10, rs1885120 in MYH7B, rs1805007 and rs1805008 both in MC1R).
Predictive value of predisposing SNPs in melanomaassociated risk factor models Figure 3 shows the areas under the curve (AUC) for 3 models considering different levels of genetic information. Compared to the phenotypic traits alone, models including the CM-associated SNPs only slightly improved the AUC. The AUC for the model that included only the nominally significant phenotypic traits (i.e. eye color, skin color, sunburn, phototype and tanning) (model 1) was 83.3%, whereas for the model that included these traits along with the 3 SNPs that remained significant after Bonferroni correction in the univariable analysis (model 2) was 83.7%, and
Discussion
We have replicated SNP-melanoma associations, with MAFs ranging from 2% to 41%. Eight associations were nominally statistically significant in the Greek population, the majority of which (87%) had previously reached genome wide significance. The replication of variants deriving from GWAS-discovered loci in our cohort, such as 20q11.2 (ASIP region), 9p21 (MTAP region), 16q24 (MC1R region) and 5p13 (CLPTM1L region), underscores the important contribution of the agnostic approach of GWAS in revealing genuine associations of genetic factors in complex diseases. For 8 SNPs the risk alleles had significantly lower frequencies in the Greek population compared to the HapMap CEU sample, while for 1 SNP the risk allele in the Greek population was higher than HapMap. The genetic models containing the SNPs that confer risk for melanoma improved the AUC compared with the model including only the phenotypic risk factors, but the improvement was of small magnitude.
The aim of our study was to validate a selected panel of SNPs in a case-control cohort of Greek descent, given our recent findings of a higher than expected genetic contribution of CDKN2A/CDK4 Table 3. List of genotyped SNPs, risk alleles in the Greek sample, and risk allele frequency in the Greek sample and HapMap CEU. mutations in a sizable cohort of sporadic and familial cases of our population [28]. Recent GWAS employing a higher density SNP tagging in large patient datasets has revealed a number of variants in genes involved in cell cycle regulation, telomere maintenance and DNA damage response, such as MITF, ATM, PARP-1, TERT, CASP8, CCND1, as well as polymorphisms in MX2, SETDB1 and ARNT/LASS2/ANXA9 region [31][32][33][34]. Although this study was based on earlier GWAS findings and certain candidate gene studies, our findings underscore the role of genes controlling pigmentary traits and DNA damage response in melanoma susceptibility in our population. This may reflect the importance of these pathways in melanoma development in a darker-skin population residing at an area of high year-round UV-influx. Most of the SNPs with significantly lower risk allele frequencies compared to HapMap CEU are found in loci implicated in pigmentation (SCL45A2, IRF4, CDK10, MYH7B, MC1R) and all but 2 (rs16891982, rs258322, rs1805007, and rs1885120) were replicated in the Greek population according to univariable analysis. These findings imply that there might be some differences in the genetic background underlying the phenotypical differences between the Greek and other European populations, and could partially explain the lower melanoma incidence in a population of darker skin complexion residing in a country with intense yearround UV exposure. In addition, our results may underscore the role of natural selection which tends to eliminate the prevalence of predisposing alleles in a population with high sun exposure and increase the frequency of protective alleles which also act through the protective pathways of pigmentation, However, Greeks harboring certain pigmentation-related risk alleles are at risk of developing melanoma.
In the case of melanocytic nevi, the comparison of allele frequencies between nevi-related variants in our cohort and the HapMap were less conclusive, with one variant (rs6001027) showing a higher allele frequency in our population. Only one (rs2218220 in the MTAP region, chrom. 9p21) of the previously nevi-associated SNPs was found to be positively associated with melanoma in our analysis. Given that nevi have been shown to be a strong risk factor of melanoma in the Greek population [35], it is likely that our study was not powered enough to detect smaller effect sizes conferred by these variants. In addition, other nevusassociated variants, yet uncovered, may play a role in melanoma risk.
Among the three top variants of our analysis, the most prominent locus was located within the cleft lip and palate transmembrane 1-like (CLPTM1L) gene and the telomerase reverse transcriptase (TERT) gene. The major C allele of rs401681 has been repeatedly reported to confer risk for BCC and protection against melanoma [36][37][38][39], and was recently replicated at a GWAS of 2,981 melanoma patients and 1,982 controls [31]. In addition, a meta-analysis including data from an Australian case-control study showed that TERT-CLPTM1L variants do influence melanoma risk, albeit with a relatively small effect size [32]. The ''red hair'' variant rs1805007 of the MC1R gene has been consistently linked to melanoma risk in relevant studies. In meta-analyses, rs1805007 showed the highest attributable risk for melanoma among MC1R variants [13,40] with effect estimates similar to those found in this study and a previous Greek case-control study [16]. rs16891982 of the SLC45A2, influences skin pigmentation and exhibits substantially different frequencies among populations, thus determined as an ancestry informative marker. The ancestral Leu allele (rs16891982-C) has been associated with dark skin, eye, and hair color in whites [41], while exhibiting a protective effect against melanoma [39,[42][43][44].
The variants selected for this study were based on the results of a large field synopsis and on-line database that scrutinized all published data on the genetic association of melanoma and subjected them to systematic meta-analyses. All but one (rs1805005) nominally significant associations in our selected set of SNPs came from a subgroup of variants which had p values of 10 27 and are likely to represent genuine associations [45]. We were also able to assess the predictive value of genetic factors in models incorporating various phenotypical and genetic risk factors. In the examined models, the predictive value of AUC did not substantially improve by the addition of genetic variants, compared with the model that involved only the clinical risk factors. Although these genetic models do not seem to contribute substantially to melanoma risk prediction, they are nevertheless suggestive of the contribution of low-penetrance gene variants to melanoma risk. Failure of models relying on common gene variants to improve substantially the predictive discrimination of traditional risk factors is a common problem encountered in complex diseases. Much larger effect sizes and a very large number of genetic variants are needed to improve perceptively the predictive value of genetic models [46]. Moreover, our findings show that statistical significance of a risk model does not guarantee clinical utility highlighting the distinction between the statistical and clinical perspectives of genetic risk models [47].
The current study has some limitations. First, the sample size is modest resulting probably in limited power to detect small or even moderate effects for additional SNPs. Second, no data were recorded on the number of nevi, a well-known melanoma risk factor for melanoma. Nevertheless, only one (rs2218220 in MTAP) of the eight SNPs associated with melanoma has been reported to be also associated with nevus count [48]. It is possible that rs2218220 would lose its significance as a melanoma-associated variant if the number of nevi were included in the multivariate analyses. Third, failure to replicate candidate loci in pigmentationassociated genes other than MC1R, SLC45A2, CDK10, MYH7B and ASIP region could derive from a lack of sufficient statistical power. Fourth, we selected our SNPs from the last update (October 2011) of the MelGene database. However, in the meantime between updates new SNPs might have been discovered in new GWAS, which are likely not to have been included in the accumulated evidence reported in MelGene because of the practical issues of intervals between database updates. This limitation may have a limited impact since some of the newest GWAS, which are not included in this paper, i.e., Barrett et al 2011 [31], provide estimates for established genetic risk factors on expanded datasets of previous GWAS, i.e., Bishop et al 2009 [49], the results of which are included in this study.
In conclusion, our research validated a number of variants that contribute to melanoma susceptibility in Greek population. The assessment of genetic input in a population with one of the lowest incidence of the disease could highlight the variation of genetic risk factors that are in-play in different environmental and population settings from those used in the majority of previous studies. Further validation of newly described variants and a better understanding of the gene-environment interaction may provide valuable insight in the variation of melanoma risk among white populations of different ancestry.
Supporting Information
Table S1 Demographic characteristics and pigmentary phenotype of melanoma cases and control subjects.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T01:12:39.957Z", "created": "2013-02-05T00:00:00.000", "id": "14900188", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0055712&type=printable", "pdf_hash": "1a5bb56b53ce44fb187c5f3d2c4e929aae3740ee", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:36515", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "1a5bb56b53ce44fb187c5f3d2c4e929aae3740ee", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Protein-tyrosine Phosphatase PTPN9 Negatively Regulates ErbB2 and Epidermal Growth Factor Receptor Signaling in Breast Cancer Cells*
ErbB family of the receptor protein-tyrosine kinase plays an important role in the progression of human cancers including breast cancer. Finding protein-tyrosine phosphatase (PTPs) that can specifically regulate the function of ErbB should help design novel therapies for treatment. By performing a small interfering RNA screen against 43 human PTPs, we find that knockdown of protein-tyrosine phosphatase PTPN9 significantly increases ErbB2 tyrosyl phosphorylation in the SKBR3 breast cancer cell line. In addition, knockdown of PTPN9 expression also enhances tyrosyl phosphorylation of the ErbB1/epidermal growth factor receptor (EGFR) in the MDA-MB-231 breast cancer cell line. Conversely, increasing expression of PTPN9 wild type (WT) inhibits tyrosyl phosphorylation of ErbB2 and EGFR. To test whether ErbB2 and EGFR are substrates of PTPN9, PTPN9 WT, and a substrate trapping mutant (PTPN9 DA) are overexpressed in SKBR3 and MDA-MB-231 cells. Compared with vector control, expression of PTPN9 WT significantly inhibits whereas expression of PTPN9 DA dramatically enhances tyrosyl phosphorylation of ErbB2 and EGFR, respectively. In contrast, expression of PTPN9 WT or DA mutant does not affect tyrosyl phosphorylation of ErbB3 and Shc. Importantly, coimmunoprecipitation and glutathione S-transferase fusion protein pulldown experiments show that tyrosol-phosphorylated ErbB2 or EGFR is preferentially associated with PTPN9 DA compared with PTPN9 WT, indicating that ErbB2 and EGFR are substrates of PTPN9. Furthermore, PTPN9 WT expression specifically impairs EGF-induced STAT3 and STAT5 activation, and inhibits the cell growth in soft agar. Last, PTPN9 WT expression also reduces invasion and MMP2 expression of MDA-MB-231 cells. Our data suggest PTPN9 as a negative regulator of breast cancer cells by targeting ErbB2 and EGFR and inhibiting STAT activation.
The ErbB family of receptor protein-tyrosine kinase (PTK), 2 which include EGFR/ErbB1, ErbB2, ErbB3, and ErbB4, plays an important role in the development of various types of cancers, including breast cancer (1). ErbB1/EGFR is usually overexpressed in a significant fraction of triple negative breast tumors that are negative in estrogen and progesterone receptors, and ErbB2/Her2 expression with a worse outcome (2). ErbB2/Her2 is overexpressed in ϳ25% of breast cancer patients who usually have a poor prognosis (3). ErbB3 is frequently expressed in breast cancers with ErbB2 overexpression (4) and is required for ErbB2-induced breast cancer cell growth (5). ErbB members are activated upon ligand binding, inducing homodimerization (i.e. EGFR) or heterodimerization of EGFR or ErbB3 with ErbB2. ErbB dimerization initiates phosphorylation on various tyrosine residues in the cytoplasmic tail of ErbB, which serve to recruit and activate multiple signaling pathways including Ras/ ERK, phosphatidylinositol 3-kinase/Akt, Src, and STAT that drive the growth, migration, and invasion of cancer cells (1). Although great efforts have been made to develop drugs to down-regulate cell surface expression (by monoclonal antibodies) and kinase activity (by small molecule kinase inhibitors) of EGFR and ErbB2, many breast cancer patients with overexpression of ErbB2 and/or EGFR still do not respond to or develop resistance to these drug treatments. Clearly, further research is needed to uncover new ways to inhibit ErbB initiated signaling in breast cancer cells.
Protein-tyrosine phosphatases (PTPs), which include membrane-associated receptor and cytoplasmic types, are enzymes that remove phosphates from phosphorylated tyrosine residues in proteins (6). Therefore, PTPs are thought to antagonize the action of PTKs that add phosphates on tyrosine residues in proteins. PTP that specifically dephosphorylates tyrosine phosphorylation of the cytoplasmic tails of ErbBs should in principal be able to inhibit oncogenic growth and invasion of EGFR/ ErbB2/ErbB3 expressing breast cancer cells. No specific ErbB3 PTP has been identified so far. Published reports indicate that PTP1B (7)(8)(9) and PTPN6/Shp-1 (10) can dephosphorylate EGFR. In addition, PTPN13 was reported to negatively regulate ErbB2 signaling through direct dephosphorylation (11). However, the role of these PTPs in breast cancer cells is still not clear. In fact, PTP1B expression promotes ErbB2-evoked breast carcinogenesis both in vitro (12) and in mice (13,14).
PTPN9, also called PTP-MEG2, is a cytoplasmic PTP. PTPN9 plays an important role in promoting intracellular secretory vesicle fusion in hematopoietic cells (15). It is required for embryonic development (16) and growth and expansion of erythroid cells (17). The role of PTPN9 in receptor PTK signaling is less well known. Only one report shows that PTPN9 can antagonize insulin signaling by reducing insulin receptor phosphorylation and Akt activation in insulin responsive cells (18). However, it is not clear whether PTPN9 inhibits insulin signaling by direct dephosphorylation of the insulin receptor. In this article, we show that PTPN9 inhibits EGFevoked signaling and STAT3 and STAT5 by direct dephosphorylation of EGFR and ErbB2. Overexpression of PTPN9 impairs oncogenic growth and invasion of breast cancer cells overexpressing ErbB2 and/or EGFR.
Plasmids and Retrovirus Production-Retroviral MSCV-IRES-GFP (pMIG) plasmid expressing human PTPN9 and its substrate trapping mutant D470A (DA) cDNAs were as described (17). pCMV plasmid expressing N-terminal FLAGtagged PTPN9 WT and PTPN9 DA were generated by PCR. Details of these constructs are available upon request. pCDNA3 expressing the rat oncogenic (activated) form of ErbB2/NeuNT was as described (19). cDNAs of the human Shp-1 wild type (WT) and phosphatase-dead substrate trapping mutant Cys 3 Ser (CS) (20) were inserted into the EcoRI site of the pLNCX2 retroviral vector. EGFP-EGFR plasmid, which does not express GFP, was a kind gift from Dr. A. Sorkin (University of Colorado, Denver). For vesicular stomatitis virus (VSV)-G pseudotyped retrovirus production, 293T cells were co-transfected with retroviral plasmids together with the packaging plasmids pCMV-VSV-G, JK3, and pCMV-TAT2 as described (21).
Retroviral Infection of Cells, Transfection of Cells with siRNA Oligo, and Cell Growth Assays-SKBR3 and MDA-MB-231 cells were infected with retroviruses and stable pools were used for experiments. SKBR3 or MDA-MB-231 cells were incubated with pMIG retroviruses or pLNCX2 retroviruses for 24 h. For the pMIG-infected cells, GFP positive cells were isolated by fluorescence-activated cell sorting 2 days later. For the pLNCX2-infected cells, stable pools of cells were selected in the presence of 200 g/ml of G418 (Invitrogen) for 6 days. For each construct, three different stable pools of cells from different retroviral infections were used in our experiments, which gave the same results. To knockdown PTPN9 expression, SKBR3 and MDA-MB-231 cells were transiently transfected with 100 nM non-target control siRNA oligo or PTPN9 siRNA oligo using transfection reagent 1 (Dharmacon), and lysed for biochemical analyses 3 days after transfection. 293T cells were transiently transfected with the indicated amount of DNA using FuGENE transfection reagent (Roche Applied Science), and lysed for biochemical analyses 3 days post-transfection. Cell proliferation was analyzed by MTS assay according to the manufacturer's instruction (Promega) in 96-well plates. For soft agar assay, cells were resuspended in culture medium containing 0.5% (MDA-MB-231) or 0.25% (SKBR3) agar (Fisher) and added to the top of the 0.5% base agar in 6-well plates. After 2-4 weeks, the colonies were stained with 0.01% crystal violet and counted.
Expression and Purification of GST-PTPN9 Fusion Proteins-PTPN9 WT and PTPN9 DA cDNAs were cloned into the pGEX4T-2 plasmid, and transformed into BL21 Escherichia coli (kindly provided by Dr. C. K. Kassenbrock). GST and GST-PTPN9 fusion protein expression were induced under room temperature overnight with 10 M isopropyl -D-1-thiogalactopyranoside, and purified on glutathione-agarose beads as described previously (20). Glutathione-agarose beads containing equal amounts (15 g) of GST or GST-PTPN9 fusion proteins were used to incubate with equal amounts of cell lysates from MDA-MB-231 or 293T-transfected cells for 2 h at 4°C.
Matrigel Invasion Assay-Matrigel invasion assay was performed in Matrigel-coated transwells (BD Biosciences Invasion chambers). Equal numbers (10 5 ) of cells resuspended in DMEM only were added to the transwells, which were placed into the 24-well plate containing 0.5 ml of DMEM with 10% FBS. Eighteen hours later, the transwells were washed and fixed with 3.7% formaldehyde, and stained with 0.05% crystal violet. The non-invaded cells on the upper surface were removed with a cotton swab and the invaded cells on the lower surface were counted from 8 random fields under a ϫ10 objective lens of a phasecontrast microscopy.
Gelatin Zymography-Equal numbers of MDA-MB-231 vector control and PTPN9 WT expressing cells were plated in a 6-well plate and 24 h later changed to DMEM without FBS. Equal volumes of the culture supernatants were harvested 24 h later, mixed with SDS sample buffer without reducing agents, and resolved in 10% SDS-PAGE containing 2 mg/ml of gelatin. The gel was washed with 2.5% Triton X-100, incubated in 15 mM CaCl 2 overnight, stained with Coomassie Blue, and destained.
RESULTS
Changing the Expression of PTPN9 Affects Tyrosyl Phosphorylation of ErbB2 and EGFR in Breast Cancer Cells-We performed a siRNA screen against 43 human PTPs using the On-Target Plus pool of siRNA (Dharmacon) in SKBR3 cells, which overexpress ErbB2, to look for protein-tyrosine phosphatases that can specifically dephosphorylate ErbB2. Individual wells, in a 96-well plate, containing a pool of siRNAs against each PTP or scramble control RNA was incubated with cells in the presence of transfection reagent (Dharmacon). Three days later, transfected cells were harvested for immunoblotting with antibody against phospho-ErbB2 followed by reprobing with antibody against ErbB2. From three independent screens, we found that only the siRNA against PTPN9 significantly increases (ϳ2fold) ErbB2 tyrosine phosphorylation in SKBR3 cells compared with scramble control RNA (data not shown).
To verify the validity of the siRNA screening result, we knocked down PTPN9 expression by transfecting SKBR3 cells with the Non-Target control siRNA oligo (NT) and an On-Target Plus siRNA oligo against human PTPN9 (Fig. 1A). We decided to use one PTPN9 siRNA oligo (number 5) because we found that this siRNA oligo inhibits 70 -80% PTPN9 expression, as shown in Figs. 1, A and B, and 3E, whereas the other three PTPN9 siRNA oligos in the On-Target Plus pool did not reduce PTPN9 expression by more than 50% (supplemental Fig. S1). Compared with cells transfected the NT oligo, cells transfected with the PTPN9 siRNA oligo (number 5) showed a significant increase (ϳ2-fold) in tyrosyl phosphorylation of ErbB2. Reprobing the same blot with anti-ErbB2 antibody indicated that PTPN9 siRNA does not affect expression of the ErbB2 protein (Fig. 1A). We also examined whether knockdown of PTPN9 affects tyrosyl phosphorylation of EGFR in MDA-MB-231 cells, which resemble triple negative breast cancer cells, expressing a higher level of EGFR but not ErbB2 (Fig. 1B). In response to EGF stimulation, EGFR becomes tyrosyl phosphorylated in cells transfected with the NT oligo. In contrast, EGF-stimulated EGFR tyrosyl phosphorylation is further enhanced in cells transfected with PTPN9 siRNA. Reprobing the same blot with anti-EGFR antibodies indicates that PTPN9 siRNA does not affect the expression level of the EGFR protein (Fig. 1B). We also examined the effects of knocking down PTPN9 expression on tyrosyl phosphorylation of ErbB2 and EGFR using the PTPN9 siRNA oligo (number 5) in BT-474 (ErbB2 positive) and BT-20 (ErbB2 negative and EGFR positive) breast cancer cell lines. Despite various transfection conditions tested, the PTPN9 siRNA (number 5) can only knockdown PTPN9 expression by ϳ50% in BT-474 and BT-20 cells, which leads to a smaller increase in ErbB2 (supplemental Fig. S2A) (ϳ30%) and EGFR (supplemental Fig. S2B) (ϳ20 -30%) phosphorylation, respectively. These data indicate that PTPN9 contributes to tyrosyl dephosphorylation of ErbB2 and EGFR in breast cancer cells.
Conversely, we also examined the effects of PTPN9 overexpression on phosphorylation of ErbB2 and EGFR. PTPN9 WT and PTPN9 DA mutant were stably expressed in SKBR3 (Fig. 1C) and MDA-MB-231 cells (Fig. 1D) by retroviral infection. The PTPN9 DA mutant has minimal phosphatase activity due to the Asp to Ala mutation in the WPD loop of the PTP domain, and can still bind ("trap") its potential protein substrates (9) (17). Thus, expression of the PTPN9 DA mutant in cells results in enhanced tyrosyl phosphorylation of its protein substrates.
The levels of PTPN9 WT and DA mutant overexpression are about 20-fold over the endogenous PTPN9 in the stable pools of SKBR3 (Fig. 1C, left panel). In randomly grown SKBR3 cells, expression of PTPN9 WT inhibits whereas expression of PTPN9 DA enhances ErbB2 tyrosyl phosphorylation (Fig. 1C, left panel). In response to EGF stimulation, tyrosyl phosphorylation of ErbB2 and EGFR was detected robustly in vector control cells but was inhibited in PTPN9 WT-expressing cells. In contrast, EGF-induced tyrosyl phosphorylation of ErbB2 and EGFR were enhanced and prolonged in PTPN9 DA expressing cells compared with vector control cells (Fig. 1C, right panel). Overexpression of PTPN9 WT or the DA mutant does not significantly affect expression of the ErbB2 protein.
The levels of PTPN9 WT and DA mutant overexpression are also about 20-fold over the endogenous PTPN9 in stable pools of MDA-MB-231 cells (Fig. 1D). MDA-MB-231 cells express a high level of EGFR but not ErbB2. Immunoblotting with anti-pEGFR antibodies revealed that EGF-invoked tyrosyl phosphorylation (Tyr 845 , Tyr 998 , and Tyr 1068 ) of EGFR was reduced greatly in PTPN9 WT-expressing cells and was enhanced and prolonged significantly in PTPN9 DA-expressing cells compared with vector control cells (Fig. 1D). Expression of PTPN9 WT and DA does not affect protein expression of EGFR (Fig. 1D). These data (Fig. 1, C and D) strongly suggest that ErbB2 and EGFR are potential substrates of PTPN9. To test whether a lower level of PTPN9 WT overexpression also inhibits ErbB2 and EGFR phosphorylation, 293T cells, which express low levels of endogenous ErbB2 and EGFR, were transiently cotransfected with the activated ErbB2 (NeuNT) or EGFR together with increasing amounts of PTPN9 WT. Although NeuNT (Fig. 1E) or EGF-induced EGFR (Fig. 1F) are robustly tyrosyl phosphorylated in cells cotransfected with vector alone (lane 1), increasing the expression of PTPN9 WT gradually reduces tyrosyl phosphorylation of NeuNT (Fig. 1E) or EGFR (Fig. 1F). Importantly, tyrosyl phosphorylation on NeuNT and EGFR were reduced by more than 95 (Fig. 1E) and 80% (Fig. 1F), respectively, when PTPN9 WT was only overexpressed ϳ5-fold over the endogenous PTPN9 (compare lanes 4 and 1), which is similar to the effects of PTPN9 WT overexpression in SKBR3 and MDA-MB-231 cells (Fig. 1, C and D). These results indicate that inhibition of ErbB2 or EGFR phosphorylation by PTPN9 overexpression is not due to the overexpression artifact.
We also examined tyrosyl phosphorylation of ErbB3, a member of the ErbB family, and Shc, an adapter protein that becomes tyrosyl phosphorylated upon ErbB activation, in cells expressing PTPN9 WT and DA mutant. ErbB3 was immunoprecipitated by an ErbB3 antibody from lysates of SKBR3 cells expressing vector alone, PTPN9 WT, and PTPN9 DA mutant followed by immunoblotting with an anti-phosphotyrosine antibody (4G10) (Fig. 1G). Compared with vector control, expression of PTPN9 WT or DA mutant has minimal effects on total tyrosyl phosphorylation of ErbB3 (Fig. 1G) and ErbB3 Tyr 1289 phosphorylation (Data not shown). Likewise, EGF-induced tyrosyl phosphorylation of p52 Shc (Tyr 239/240 ) was also not affected by expression of PTPN9 WT or DA mutant in SKBR3 (Fig. 1H, left panel) and MDA-MB-231 cells (Fig. 1H, right panel) compared with vector control cells. The p66 and p46 Shc isoforms are expressed at a low level (Ͻ10% of p52 Shc) in these two cell lines, which are not included here. Proteintyrosine phosphatase Shp-1 is expressed in breast cancer cells (22). However, we found that overexpression of Shp-1 WT and Shp-1 CS mutant (a phosphatase-dead substrate trapping mutant) has minimal effects on EGF-evoked tyrosyl phosphorylation of EGFR (Tyr 1068 ) compared with vector control in MDA-MB-231 cells (Fig. 1I). These results indicate that overexpression of PTPN9 specifically affects tyrosyl phosphorylation of ErbB2 and EGFR, rather than nonspecifically dephosphorylating tyrosyl-phosphorylated proteins in breast cancer cells.
ErbB2 and EGFR Are Potential Substrates of PTPN9-To further address whether PTPN9 expression inhibits ErbB2 and EGFR phosphorylation by directly dephosphorylating tyrosine residues in ErbB2 and EGFR in vivo, we took advantage of the observation that PTPs with a aspartic acid 3 alanine (DA) mutation in the WPD loop of the PTP catalytic domain have little phosphatase activity, but still maintain the ability to bind (trap) potential substrates (9). Expression of the DA mutant behaves as a biochemical dominant-negative in cells by protecting dephosphorylation of its substrates, resulting in enhanced tyrosyl phosphorylation of the potential protein substrates and association of the DA mutant with its substrates. ErbB2 association with PTPN9 WT and DA mutant in stable pools of SKBR3 cells was investigated by immunoprecipitation ( Fig. 2A). In both vector and PTPN9 WT overexpressing cells, no tyrosyl-phosphorylated proteins were detected in the PTPN9 immunoprecipitates. In contrast, tyrosyl-phosphorylated (Tyr(P)) proteins with molecular masses of 175, 100, and 65 kDa were specifically associated with the PTPN9 DA mutant ( Fig. 2A, lane 3). Reprobing the PTPN9 immunoprecipitates with anti-PTPN9 antibodies showed that the 65-kDa Tyr(P) band was comigrated with PTPN9 DA, strongly suggesting that the 65-kDa band is the tyrosyl-phosphorylated PTPN9 DA mutant. The same amounts of PTPN9 WT and PTPN9 DA proteins were immunoprecipitated ( Fig. 2A). One likely possibility is that PTPN9 DA is hyperphosphorylated. PTPN9 WT is normally phosphorylated by a tyrosine kinase in the cells, and rapidly dephosphorylated by itself. PTPN9 DA trapping mutant is not only phosphatase-dead, but also protects the DA mutant from dephosphorylation by trapping the phosphorylated tyrosine residues in PTPN9 DA, thus resulting in hyperphosphorylation of the PTPN9 DA mutant. Hyperphosphorylation was also seen in the substrate trapping mutants of other PTPs including Shp-1 (20).
We do not know the nature of the 100-kDa Tyr(P) protein. Due to the lower sensitivity of detection, reprobing the immunoblot with anti-ErbB2 antibody failed to detect the presence of ErbB2 comigrated with the 175-kDa Tyr(P) band in the PTPN9 DA complex although it is very likely that the 175-kDa band is the tyrosyl-phosphorylated ErbB2.
To provide evidence that the PTPN9 DA trapping mutant can coimmunoprecipitate with the phosphorylated ErbB2, 1 (lane 6). G, overexpression of PTPN9 WT does not affect tyrosyl phosphorylation of ErbB3. The same SKBR3 cells used in C were starved, and stimulated with 5 ng/ml of heregulin (HRG1). Equal amounts of lysates were immunoprecipitated (IP) with anti-ErbB3 antibodies, immunoblotted with anti-phosphotyrosine antibody (4G10), and reprobed with anti-ErbB3 antibodies. H, overexpression of PTPN9 WT does not affect tyrosyl phosphorylation of Shc. The same SKBR3 cells used in C were starved and stimulated with EGF. Equal amounts of lysates were immunoblotted with anti-pShc (Y239/240) antibodies and reprobed with anti-Shc antibodies. I, overexpression of Shp-1 does not affect tyrosyl phosphorylation of EGFR. MDA-MB-231 cells stably expressing pLNCX2 vector alone, and pLNCX2 expressing Shp-1 WT and CS mutant by retroviral infection were starved and stimulated with 25 ng/ml of EGF. Equal amounts of lysates were immunoblotted with the pEGFR (Y1068) antibodies and reprobed with anti-EGFR antibodies. Densitometry was used to quantify bands in Western blots (Fig. 1, A-D and I). Numbers under the pErbB2 or pEGFR blots show the ratios of pErbB2/total ErbB2 or pEGFR/total EGFR. Numbers under the PTPN9 blots show the ratios of PTPN9/-actin. Data shown in this figure are representative from at least three independent experiments.
293T cells were transiently cotransfected with plasmid expressing NeuNT, the activated form of rat ErbB2, together with the FLAG-tagged PTPN9 constructs (Fig. 2B). Transfected 293T cells were lysed and incubated with anti-FLAG antibody. Although tyrosyl-phosphorylated ErbB2 was not associated with FLAG-PTPN9 WT, a 175-kDa Tyr(P) protein that comigrated with ErbB2 was associated with FLAG-PTPN9 DA. Immunoblotting with anti-ErbB2 antibodies showed robust association of ErbB2 with the FLAG-PTPN9 DA (Fig. 2B, lane 3) and minimal association of ErbB2 with FLAG-PTPN9 WT. This result indicates that only the PTPN9 DA trapping mutant, not PTPN9 WT, is associated with tyrosyl-phosphorylated ErbB2, strongly suggesting that ErbB2 is a substrate for PTPN9.
To provide evidence that EGFR is a substrate for PTPN9, 293T cells were cotransfected with plasmid-expressing EGFR together with the FLAG-tagged PTPN9 constructs. Transfected cells were stimulated with EGF before being lysed and subjected to immunoprecipitation with anti-FLAG antibody (Fig. 2C). Although the phosphorylated EGFR was not found associated with FLAG-PTPN9 WT, a 170-kDa Tyr(P) protein was associated with FLAG-PTPN9 DA. Immunoblotting with anti-EGFR antibodies revealed that EGFR was comigrated with the 170-kDa Tyr(P) protein in the FLAG-PTPN9 DA immunoprecipitates (Fig. 2C, lane 3), strongly suggesting that EGFR is also a substrate for PTPN9.
It is possible that expression of the PTPN9 DA mutant in cells could affect expression or activity of another protein that may mediate the interaction between the PTPN9 DA mutant and the hyperphosphorylated ErbB2 or EGFR. To exclude this possibility, equal amounts of GST alone, GST-PTPN9 WT, and GST-PTPN9 DA fusion proteins were expressed in bacteria, purified by glutathione-agarose beads (Fig. 2D, left panel), and incubated with cell lysates from 293T cells transiently transfected with NeuNT (Fig. 2D, middle panel) or MDA-MB-231 cells (EGF stimulated) (Fig. 2D, right panel). Consistent with the results from our co-immunoprecipitation experiments (Fig. 2, A-C), GST or GST-PTPN9 WT does not pulldown phosphorylated ErbB2 or EGFR. In contrast, GST-PTPN9 DA pulls down tyrosyl-phosphorylated ErbB2 (Fig. 2D, middle panel) or EGFR (Fig. 2D, right panel), further supporting that PTPN9 DA traps the phosphorylated ErbB2 and EGFR directly. Overall, our data support that ErbB2 and EGFR are direct substrates for PTPN9.
PTPN9 Overexpression Inhibits EGF-induced STAT Activation-Because PTPN9 overexpression specifically dephosphorylates EGFR and ErbB2 (Figs. 1 and 2), we expect that EGFinduced downstream signaling pathways should also be reduced. EGF-induced activation of ERK and Akt in SKBR3 and MDA-MB-231 cells overexpressing PTPN9 was examined by immunoblotting using phosphospecific antibodies against ERK and Akt. To our surprise, EGF-invoked Akt and ERK phosphorylation was not impaired in PTPN9 WT and DA overexpressing cells compared with vector control cells (Fig. 3A). Similarly, in MDA-MB-231 cells, EGF-invoked Akt and ERK phosphorylation was not affected in PTPN9 WT and DA overexpressing cells compared with vector control cells (Fig. 3B). The observation that PTPN9 overexpression has no effect on ERK activation is consistent with the data that PTPN9 WT overexpression has no effect on Shc tyrosyl phosphorylation in SKBR3 and MDA-MB-231 cells (Fig. 1H).
EGF stimulation also leads to activation of STAT3 (23) and STAT5 (24,25) in breast cancer cells. EGF-induced STAT3 and STAT5 activation was examined in SKBR3 and MDA-MB-231 cells overexpressing PTPN9 by immunoblotting using phosphospecific antibodies against STAT3 and STAT5 (Fig. 3, C and D). In contrast to ERK and Akt activation, EGF-induced STAT3 and STAT5 phosphorylation was dramatically inhibited in PTPN9 WT expressing SKBR3 (Fig. 3C) and MDA-MB-231 cells (Fig. 3D) compared with vector control cells. Interestingly, whereas EGF-induced STAT5 activation was not affected in cells expressing PTPN9 DA compared with vector control, EGF-induced STAT3 activation was significantly reduced in cells expressing PTPN9 DA compared with vector control cells (Fig. 3, C and D). Consistent with the inhibition of STAT3 phosphorylation by PTPN9 overexpression, we found that knockdown of PTPN9 expression by PTPN9 siRNA resulted in an increase of EGF-induced STAT3 phosphorylation in SKBR3 (left panel) and MDA-MB-231 (right panel) cells (Fig. 3E).
Overexpression of PTPN9 WT Inhibits the Oncogenic Growth and Invasion of Breast Cancer Cells-ErbB2 and EGFR play important roles in driving the growth, invasion, and metastasis of breast cancer cells in part through activation of STAT3 and STAT5 (26). We investigated whether expression of PTPN9 affects the growth and invasion of SKBR3 and MDA-MB-231 cells. First, we examined the growth of PTPN9 overexpressing cells in tissue culture plates by MTS, which measures viable cells. On tissue culture plastic plates, for MDA-MB-231 (Fig. 4A) and SKBR3 (Fig. 4B) cells, cell growth in serum containing medium was the same in vector control and PTPN9 overexpressing cells. Similar results were obtained when cells were grown in serumfree medium (data not shown). Next, we examined the effect of PTPN9 WT overexpression on cell growth in soft agar-anchorage independent growth, which is one of the hallmarks of cancer cells. Although there was robust colony formation in the vector control of MDA-MB-231 (Fig. 4C) and SKBR3 (Fig. 4D) cells, the number of colonies were decreased by ϳ70% both in PTPN9 WT expressing MDA-MB-231 and SKBR3 cells. These results indicate that increased expression of PTPN9 impairs the oncogenic growth of breast cancer cells.
We also examined whether overexpression of PTPN9 affects the invasion of breast cancer cells. MDA-MB-231 cells expressing vector and PTPN9 WT were subjected to the Matrigel invasion assay. Vector control cells showed robust invasion through the Matrigel. In contrast, PTPN9 WT expressing cells displayed a 30% decrease in invasion (Fig. 5A). We cannot assess the effect of PTPN9 expression on the invasion of SKBR3 cells because both the vector control and PTPN9 WT-expressing cells could not invade through the Matrigel in our assay. We also examined the effects of PTPN9 expression on cell migration by performing the Fig. 1, C and D) were starved, stimulated with 25 ng/ml of EGF for the indicated times, lysed, immunoblotted with antibodies against pERK1/2, pAkt (Ser 473 ), and reprobed with antibodies against ERK1/2, Akt, and PTPN9. C and D, expression of PTPN9 WT inhibits EGF-induced STAT3, and STAT5 in SKBR3 and MDA-MB-231 cells. The same SKBR3 (C) and MDA-231 (D) cells used in A and B, respectively, were starved, stimulated with 25 ng/ml of EGF for the indicated times, lysed, and immunoblotted with antibodies against the indicated pSTAT3 (Y705) and pSTAT5 (Y694), and reprobed with antibodies against STAT3 and STAT5. Shown (A-D) are the representative results from three independent experiments. E, PTPN9 knockdown enhances EGF-induced STAT3 activation. SKBR3 (left) and MDA-MB-231 (right) cells were transfected with control (NT) and si-PTPN9 oligos, starved, and stimulated with 25 ng/ml of EGF for the indicated times. Equal amounts of lysates were immunoblotted with anti-pSTAT3 antibodies and reprobed with anti-STAT3 and anti-PTPN9 antibodies. Shown here is the representative result from two independent experiments. wound healing assay. However, we did not find that overexpression of PTPN9 WT affects cell migration in SKBR3 and MDA-MB-231 cells (data not shown).
Cancer cells invade through its extracellular matrix environment through secretion of matrix degrading proteases. MDA-MB-231 cells are known to express matrix metalloproteinase 2 (MMP2) (27). Thus, we ask whether PTPN9 expression affects MMP2 expression. Gelatin zymography was used to examine the expression of MMP2 and MMP9 proteins from MDA-MB-231 cells expressing vector and PTPN9 WT (Fig. 5B). MMP2 protein was highly expressed in vector control cells. In contrast, MMP2 expression was dramatically reduced in PTPN9-expressing cells. We could not detect any MMP9 protein expression both in vector control and PTPN9 expressing MDA-MB-231 cells.
Quantitative PCR was used to assess MMP2 mRNA expression (Fig. 5C). Consistent with zymography data (Fig. 5B), PTPN9 WT overexpression resulted in reduced MMP2 mRNA expression. In addition, MMP9 mRNA was found to be expressed at a very low level in MDA-MB-231 cells. These data suggest that PTPN9 expression inhibits the invasion of MDA-MB-231 cells likely through the suppression of MMP2 gene expression.
DISCUSSION
EGFR and ErbB2 are involved in various types of human cancers including breast cancer. Thus, identification of protein-tyrosine phosphates that can inhibit EGFR and ErbB2 function by direct dephos- phorylation should provide new information to design novel cancer therapies. In this article, for the first time, we provide evidence that PTPN9 is a protein-tyrosine phosphatase specific for EGFR and ErbB2 and that negatively regulates EGFR and ErbB2 signaling.
Our study is the first report that PTPN9 inhibits receptor PTK activation by direct dephosphorylation of EGFR and ErbB2. First, knockdown of PTPN9 expression by siRNA results in an increase of tyrosyl phosphorylation of ErbB2 (Fig. 1A) and EGFR (Fig. 1B). Second, overexpression of PTPN9 WT inhibits tyrosyl phosphorylation of EGFR and ErbB2 (Fig. 1, C-F) without affecting EGFR and ErbB2 protein expression. Importantly, PTPN9 WT overexpression does not affect tyrosyl phosphorylation of ErbB3 (Fig. 1G), Shc (Fig. 1H), and activation of Src PTK (data not shown). In contrast, expression of the PTPN9 substrate trapping mutant DA enhances tyrosyl phosphorylation of EGFR and ErbB2 in these same assays (Fig. 1, C and D). Third, only the substrate trapping mutant of PTPN9 DA, not PTPN9 WT, is coimmunoprecipitated with (Fig. 2, A-C) and can pulldown (Fig. 2D) the phosphorylated EGFR or ErbB2. These results strongly suggest that the PTPN9 DA mutant is able to form a stable complex with the phosphorylated EGFR and ErbB2 because the DA phosphatase domain likely binds the phosphotyrosine residues in EGFR or ErbB2 but cannot catalyze the dephosphorylation reaction. Currently, we cannot exclude the possibility that PTPN9 DA coimmunoprecipitates with tyrosyl-phosphorylated EGFR and ErbB2 via another tyrosine-phosphorylated protein such as the p100 Tyr(P) protein although the level of tyrosine phosphorylation on p100 is lower compared with that of ErbB2 in the PTPN9 DA complex ( Fig. 2A). Although a previous report shows that PTPN9 inhibits insulin receptor phosphorylation and insulininduced Akt activation in insulin responsive tissues, it is still not clear whether insulin receptor is a direct substrate of PTPN9 (18).
Our result also reveals PTPN9 as one of the PTPs that can dephosphorylate EGFR and ErbB2, and modulate EGFR and ErbB2 induced signaling and biological responses. EGFR has been reported to be the substrate for tyrosine phosphatase Shp-1 and PTP1B. EGFR is shown to be a potential substrate for Shp-1 only in cotransfection experiments with overexpression of Shp-1 in 293 cells (10). Shp-1 via its SH2 domain binds Tyr 1173 in EGFR and dephosphorylates other tyrosine residues in EGFR, and inhibits EGF-induced ERK activation (10). However, our results indicate that overexpression of Shp-1 WT or CS mutant does not affect tyrosyl phosphorylation of EGFR (Fig. 1I) and EGF-induced ERK activation (data not shown) in MDA-MB-231 cells. By fluorescence resonance energy transfer imaging, PTP1B has been shown to dephosphorylate EGFR on the surface of endoplasmic reticulum after endocytosis of the activated EGFR (28). ErbB2 has been reported to be a substrate for PTPN13 in HeLa and 293 cells (11). Besides its effect on STAT3 activation, PTPN13 expression also inhibits EGF-induced ERK and Akt activation (11). Our results show that PTPN9 specifically dephosphorylates EGFR and ErbB2 (Fig. 1), and inhibits EGF-induced STAT activation without affecting ERK and Akt activation (Fig. 3). Our studies using subcellular fractionation and immunofluorescent staining of PTPN9 indi-cate that overexpressed PTPN9WT and endogenous PTPN9 show similar pattern of subcellular localization, which is highest in the low density microsomes rich in transporting vesicles, medium in the high density microsomes, low in the plasma membrane, and undetectable in cytosol (supplemental Fig. S3). This is consistent with the reported localization of PTPN9 being mainly present in secretory vesicles. However, our data also reveal that a small fraction of PTPN9 is present in plasma membrane. These data suggest that PTPN9 may interact and/or dephosphorylate EGFR and ErbB2 both in the plasma membrane and intracellular membrane compartments, which will be the future focus of our study. These results indicate that PTPN9 has distinct role in regulating the activation status of EGFR and ErbB2 compared with PTP1B, Shp-1, and PTPN13.
Published reports show that EGFR can activate STAT3 and STAT5 in two ways (26). EGFR-activated Src can promote phosphorylation of STAT3 and STAT5. Alternatively, Tyr 845 in EGFR can be phosphorylated by Src upon EGF stimulation, which recruits and activates STAT3 and STAT5 (26). A recent study by quantitative proteomics indicates that phosphorylated Tyr 998 in EGFR can also recruit STAT5 directly (29). We found that PTPN9 expression did not affect EGF-induced Src activation in SKBR3 and MDA-MB-231 cells (data not shown). However, phosphorylation of Tyr 845 and Tyr 998 in EGFR was inhibited by PTPN9 expression in MDA-MB-231 cells (Fig. 1D). Therefore, we favor the working model that PTPN9 expression results in impaired phosphorylation on tyrosine residues (Tyr 845 and Tyr 998 ) in EGFR that recruit and activate STAT3 and STAT5 in MDA-MB-231 cells. In SKBR3 cells that overexpress ErbB2, EGF stimulation induces heterodimerization of EGFR and ErbB2. Because PTPN9 can also dephosphorylate and inhibits ErbB2 activation, PTPN9 expression could directly dephosphorylate Tyr 845 and Tyr 998 in EGFR or indirectly by inhibiting ErbB2 activation in SKBR3 cells. In any event, our data do not support that PTPN9 directly dephosphorylates STAT3 and STAT5. Supporting this notion, we found that EGF-induced STAT3 and STAT5 phosphorylation is reduced in cells expressing PTPN9 DA trapping mutant compared with vector control cells (Fig. 3, C and D). If STAT3 and STAT5 were substrates of PTPN9, we would find enhanced STAT3 and STAT5 phosphorylation in PTPN9 DA-expressing cells compared with vector control cells because expression of the substrate trapping mutant of PTPN9 should protect its substrates from dephosphorylation such as in the case of hyperphosphorylated ErbB2 and EGFR in the presence of PTPN9 DA expression (Fig. 1, C and D). These data are consistent with the idea that PTPN9 WT dephosphorylates the phosphorylated tyrosine residues in EGFR that recruit and activate STAT3 and STAT5.
Activation of ERK and phosphatidylinositol 3-kinase/Akt pathways play important roles in mediating EGFR/ErbB2 evoked biological responses such as cell growth, migration, and invasion in breast cancer cells (1). However, recent studies indicate that EGFR can activate STAT3 and STAT5, which turn on gene expression critical for the proliferation and survival of breast cancer cells (26). Because PTPN9 expression specifically inhibits EGF-induced STAT3 and STAT5 activation without affecting ERK and Akt activation in SKBR3 and MDA-MB-231 cells (Fig. 3), it is very likely that impaired STAT3 and STAT5 activation contributes to the reduced cell growth in soft agar when PTPN9 was overexpressed (Fig. 4). In addition, our data also indicate that PTPN9 overexpression reduces invasion and decreases MMP2 gene expression in MDA-MB-231 cells. It was reported that STAT3 activation promotes invasion, metastasis, and MMP2 gene expression in melanoma cells. STAT3 activates MMP2 gene transcription by binding to the high affinity STAT3 binding site in the promoter region of the MMP2 gene (30). Because PTPN9 expression inhibits STAT3 activation (Fig. 3), MMP2 gene expression (Fig. 5C), and invasion (Fig. 5A), it is likely that PTPN9 expression impairs cell invasion through down-regulation of the STAT3-MMP2 pathway in MDA-MB-231 cells. A recent study indicates that subclones of MDA-MB-231 cells, specifically metastasis to the brain and lung in the xenograft model, overexpress the EGFR ligand HB-EGF (31), indicating the importance of EGFR signaling in metastasis of breast cancer cells. Our result suggests that the EGFR-induced STAT3-MMP2 pathway may contribute to brain and lung metastasis of MDA-MB-231 cells, and the level of PTPN9 expression may affect breast cancer metastasis in vivo.
In this study, we find that expression of PTPN9 inhibits the oncogenic growth and invasion of breast cancer cells by specifically dephosphorylating EGFR and ErbB2 and impairing EGFR and ErbB2 signaling. It will be important to examine whether the PTPN9 expression level can predict the disease outcome in ErbB2/Her2 positive and triple negative breast cancers in a future study. Furthermore, it will be also interesting to explore the role of PTPN9 in other human cancers such as lung and brain cancers where EGFR is the critical driver for the malignant progression. Last, therapeutic approaches that can increase PTPN9 expression may be useful in treating breast cancer patients with EGFR and/or ErbB2 overexpression.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T02:57:51.422Z", "created": "2010-03-24T00:00:00.000", "id": "35714188", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "http://www.jbc.org/content/285/20/14861.full.pdf", "pdf_hash": "740ae61203130da1e4ad2572d9e3b6a9d40e33a7", "pdf_src": "Highwire", "provenance": "peS2o-0075.json.gz:36607", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "36f45b25e56282be6d60841324d5eeb401ee4198", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Targeted Depletion of TDP-43 Expression in the Spinal Cord Motor Neurons Leads to the Development of Amyotrophic Lateral Sclerosis-like Phenotypes in Mice*
Background: Most amyotrophic lateral sclerosis (ALS) cases are characterized with TDP-43(+), ubiquitin(+) inclusions in their diseased spinal cord motor neurons. Results: Mice with targeted depletion of TDP-43 expression in the spinal cord motor neurons developed a range of ALS-like phenotypes. Conclusion: TDP-43 is essential for the survival and functioning of mammalian spinal cord motor neurons. Significance: Loss of TDP-43 function could be one major cause for neurodegeneration in ALS with TDP-43 proteinopathies. ALS, or amyotrophic lateral sclerosis, is a progressive and fatal motor neuron disease with no effective medicine. Importantly, the majority of the ALS cases are with TDP-43 proteinopathies characterized with TDP-43-positive, ubiquitin-positive inclusions (UBIs) in the cytosol. However, the role of the mismetabolism of TDP-43 in the pathogenesis of ALS with TDP-43 proteinopathies is unclear. Using the conditional mouse gene targeting approach, we show that mice with inactivation of the Tardbp gene in the spinal cord motor neurons (HB9:Cre-Tardbplx/−) exhibit progressive and male-dominant development of ALS-related phenotypes including kyphosis, motor dysfunctions, muscle weakness/atrophy, motor neuron loss, and astrocytosis in the spinal cord. Significantly, ubiquitinated proteins accumulate in the TDP-43-depleted motor neurons of the spinal cords of HB9:Cre–Tardbplx/− mice with the ALS phenotypes. This study not only establishes an important role of TDP-43 in the long term survival and functioning of the mammalian spinal cord motor neurons, but also establishes that loss of TDP-43 function could be one major cause for neurodegeneration in ALS with TDP-43 proteinopathies.
ALS are sporadic but ϳ10% of patients have a familial history (known as familial ALS; fALS). In both sALS and fALS, there are progressive manifestations of dysfunction of the lower motor neurons and cortical motor neurons but without sensory symptoms (3). Age and gender are documented sALS risk factors (5) with a male-to-female ratio of 3:2 among patients. Among the mutations associated with ALS, those in the copper-zinc superoxide dismutase (SOD1) gene have long been thought to cause ALS through a toxic gain of function rather than impairment of the antioxidant function of the SOD1 enzyme (6). Other genes with mutations associated with the fALS include alsin (ALS2), senataxin (ALS4), vesicle-associated membrane protein (VAPB, ALS8), Angiogenin, and the p150 subunit of dynactin (DCTN1) (3). More than 30 mutations in the TDP-43-coding region of Tardbp have also been identified in ALS patients with or without apparent family history, corresponding to ϳ4% of fALS and less than 1% of sALS (7). Most patients with the TDP-43 mutation(s) develop a classical ALS phenotype without cognitive deficit suggesting an important role of TDP-43 in the development of ALS (7)(8)(9). TDP-43, or TAR DNA-binding protein-43 (10), is an ubiquitously expressed nuclear protein encoded by one of the mRNA isoforms from the highly conserved Tardbp gene (11). It is a RNA-binding protein involved in transcriptional repression, pre-mRNA splicing, and translation (12,13). TDP-43 has also been identified as the major pathological signature protein of the intracellular inclusions typical for disease cells of a range of neurodegenerative diseases, including the frontotemporal lobar degeneration with ubiquitin-positive, Tau-and ␣-synuclein-negative inclusions (FTLD-U) and ALS (13)(14)(15)(16)(17). TDP-43 molecules in the diseased cells of the patient brains or spinal cords are characterized by abnormal ubiquitination, hyperphosphorylation, and partial cleavage to generate ϳ25-kDa and 35-kDa C-terminal fragment(s). Furthermore, TDP-43 is partially or completely cleared from the nuclei of either neuronal or glial cells containing the TDP-43(ϩ) and ubiquitin(ϩ) aggregates/inclusions, or UBIs, in the cytoplasm (18).
Several mouse models have been established for ALS disease including the strains of rodents that are transgenic with SOD1, ALS2 knock-out mice, and mice with genetically engineered genes coding for the neurofilament subunits (reviewed in Refs. 19 and 20). Among these, the mutant human SOD1 (hSOD1) transgenic mouse model is currently the most widely used because it shares several clinical phenotypes with the ALS patients. The first symptom of the hSOD1 mice is a fine "jittering/tremor" in one or more of the limbs, which appears at ϳ90 to 100 days of age (21,22). At later stages, the cytopathological features of the hSOD1 transgenic mice include motor neuron loss with astrocytosis, the presence of SOD1-positive inclusions including Lewy body-like hyaline inclusions/astrocyte hyline inclusions, and vacuole formation (23).
Overexpression of TDP-43 in transgenic rodents could also lead to development of motor neuron disease-like symptoms. These transgenic rodents develop one or more of several symptoms, which include motor neuron dysfunction, muscle defectrelated pathology, and neuronal loss (24 -27). The lifespans of some of these transgenic mouse lines are short, likely due to the cytotoxicity of the overexpressed TDP-43 as well as the relatively lower motor neuron specificity of the promoters, e.g. Thy1, prion, etc., used to express the transgenes (24, 26 -28). Finally, the appearance of cells with cytoplasmic TDP-43(ϩ) UBIs and TDP-43-depleted nuclei at later stages of pathogenesis of the TDP-43 transgenic mice (25,26) suggest that the disease phenotypes in the TDP-43 transgenic mice may result in part from loss of function of TDP-43. However, the pathological phenotypes of the mice could also be caused entirely by gain of toxicity from overexpression of the exogenous TDP-43.
Thus, the relative contributions of loss of function and gain of cytotoxicity to the neurodegeneration in FTLD-U and ALS with TDP-43(ϩ) UBIs are not clear (reviewed in Refs. 16 and 29 -31). Also, regardless of its currently known biochemical and structural properties, the physiological functions of TDP-43 in different mammalian tissues are also unknown. Previously, we have shown, by gene targeting approaches in mice, that TDP-43 is important for mouse early embryonic development (32). As described in the following, we have taken advantage of the Tardbp lx mouse line from that study and generated mice with spinal cord motor neuron-specific knock-out of TDP-43 expression. Furthermore, these mice exhibit pathological phenotypes in striking similarity to ALS.
Generation of HB9 Promotor-driven Conditional Tardbp Gene
Knock-out Mice-The Tardbp allele of the C57BL/6j mice was knocked out specifically in the postmitotic motor neurons in the spinal cord by crossing mice carrying the Tardbp conditional allele (Tardbp lx ) with mice carrying a Cre-recombinase transgene driven by the HB9-promoter (HB9:Cre)(see "Results" for more details). The viabilities and weights of the mice were monitored regularly. Genotyping of the mice was performed by PCR of genomic DNAs from the tail biopsies in comparison to the nail samples.
Tests of Motor Neuron Functions-For the limb-clasping test, the mice were suspended by pulling their tails. For the rotarod test, the mice were trained before testing to exclude differences in motivation and motor learning. Mice were first trained for four consecutive days. In the testing phase, they were put on the rod rotating constantly at 2.5 rpm, and the speed was gradually increased to 25 rpm over a 3-min period. The timer was stopped when the mice fell from the rod or when they gripped the rod and started to rotate with it.
Tissue Preparation and Immunostaining Analysis-The mice were sacrificed under deep anesthesia, and perfused transcardially with 4% paraformaldehyde in PBS (pH 7.4). The spinal cords were dissected and the lumber segments (L3-L6) were identified using the ribs and vertebrae as the guide. The segments were processed for cryoprotection, and 100 serial cross-sections were made of the lumbar spinal cords at a thickness of 10 m. Every 12th section (12 sections total from each animal) was stained with 1% cresyl violet (Sigma). Each section was visualized with an Axioimage-Z1 light microscope at ϫ20 magnification, and the cells were counted manually by tracing the perimeter of each motor neuron. The cell counts were made within an area demarcated by a horizontal line drawn through the central canal and encompassing the ventral horn of the gray matter to include layers 7-9. Initially, we identified the motor neurons using the following criteria: 1) the presence of a large nucleolus located within the nucleus surrounded by light bluestaining cytoplasm; and 2) a cell soma area over 100 m 2 . The ␥ motor neurons range in their soma areas from 100 to 250 m 2 , whereas the soma areas of the larger ␣ motor neurons range from 250 to 1100 m 2 .
In parallel sets of the sections, the motor neurons were also analyzed by immunofluorescence staining with single or combined use of several different antibodies: a goat antibody directed against the choline acetyltransferase (ChAT; Chemicon), a rabbit anti-TDP-43 (Proteintech), a rabbit anti-GFAP (Milopore), a rabbit anti-Iba1 (Wako), a mouse anti-MAP2 (Sigma), a mouse antiubiquitin (Zymed Laboratories Inc.), a mouse anti-neuronal nuclei (NeuN), and a mouse anti-human ERR␥ (PPMX). The sections were preincubated in PBS solution containing 10% (v/v) normal donkey serum and 0.2% Triton X-100 for 1 h, and then incubated overnight with one or more of the antibodies. Antibody against the rabbit/mouse IgG (Jackson ImmunoResearch) was used for secondary detection. The images were taken from the ZEISS-LSM710 and LSM780 confocal microscope.
Immunoblotting Analysis-Total extracts of the spinal cords were prepared by homogenization of the tissue in the RIPA buffer (0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 5 mM EDTA, 150 mM NaCl, 50 mM Tris-HCl, pH 8.0) containing protease and phosphatase inhibitors. The RIPA-soluble and urea-soluble fractions of the proteins from the spinal cords were obtained by centrifugation of the tissue homogenate at 15,000 ϫ g at 4°C for 30 min. The supernatant was collected as the "RIPA-soluble" fraction. The pellet was washed 3 times with RIPA buffer and then solublized in the urea buffer (7 M urea, 2 M thiourea, 4% CHAPS, 30 mM Tris-HCl, pH 8.5) to give the "urea-soluble" fraction. 4 g of RIPA extract or ϫ4 equivalent volumes of urea extract per lane were separated on a 10% Tris glycine SDS-PAGE gel. Immunoblotting analysis of the RIPAsoluble and urea-soluble fractions of the spinal cord extracts followed standard procedures with use of appropriate antibodies. For comparing the protein levels of WT, Tardbp lx/Ϫ mice, and HB9:Cre-Tardbp lx/Ϫ mice, quantification of band intensity was performed using Image J software (NIH). For TDP-43/ ChAT expression quantification, protein levels relative to wild type (WT) mice was determined using TDP-43/ChAT immunoblots within the linear range of band intensity after normalization to tubulin expression.
Generation of Mice with Conditional Knock-out of Tardbp Expression in the Spinal Cord Motor Neurons-The Tardbp
gene was inactivated in the motor neurons with use of the loxP/ cre recombination system. We first generated a conditional allele of the Tardbp locus by flanking exons 2 and 3 with loxP sites (Tardbp lx allele), which was suitable for Cre-mediated Tardbp-floxed gene inactivation ( Fig. 1A; Ref. 32). We then crossed the mice with motor neuron-specific Mnx1 (HB9):Cre mice to generate the HB9:Cre-Tardbp lx/Ϫ mice (Fig. 1).
HB9 is a homeodomain transcription factor that is expressed selectively in motor neurons in the developing spinal cord (E9.5) and is essential for differentiation of the postmitotic motor neurons (33). Also, HB9:Cre mice express the cre gene specifically in the spinal cord motor neurons and have been used previously to manipulate gene expression in the motor neurons (33). The motor neuron specificity of the HB9-driven Cre expression have also been verified by crossing the HB9:Cre mouse line to a ROSA26:LacZ reporter mice in several studies (33)(34)(35)(36)(37)(38). Examination of the GFP signals in HB9:GFP transgenic mice have indicated that expression of HB9 is restricted in the motor neurons, but not in the sensory neurons (39).
On the other hand, it should be emphasized here that HB9 expression in cells other than the spinal cord motor neurons could not be completely ruled out. For instance, expression of HB9:Cre has also been observed in cells other than the ventral motor neurons, which could be due to HB9 promoter activity in the visceral motor neurons or in motor neuron progenitors (34). Other studies have shown that Cre is expressed in motor neurons as well as in some ventral interneurons at the embryonic stage (35,36). In any case, as shown and described below, the use of HB9:Cre mice eliminated TDP-43 expression mainly in ChAT(ϩ) motor neurons in the spinal cords of our conditional knock-out mice.
Development of ALS-like Morphological and Behavioral Phenotypes in HB9:Cre-Tardbp lx/Ϫ Mice-Unlike the EIIa:Cre-Tardbp lx/lx mice we generated previously (32), which were lethal at the peri-implantation stage, HB9:Cre-Tardbp lx/Ϫ mice were viable and phenotypically normal at birth, suggesting that TDP-43 was not essential for normal development of spinal cord motor neurons. However, although the average weight of HB9:Cre-Tardbp lx/Ϫ mice was only slightly lower than the controls at early births, the difference became more prominent afterward ( Fig. 2A). The body weight was shown previously to be a simple and reliable measure for the disease "onset" and progression in the ALS mouse model with transgenic expression of the mutant hSOD1 (G93A) (40), with the inflection point of the weight curve providing a simple definition of the disease onset. Interestingly, similar to the mutant hSOD1 transgenic mice, HB9:Cre-Tardbp lx/Ϫ mice also showed a peak of weight gain during 90 to 100 days (ϳ13 weeks, Fig. 2A). Soon after that, the mice started to show significant weight loss ( Fig. 2A), abnormal hind limb clasping (Fig. 2B), and deficiency in rotarod test (Fig. 2C). It should be noted here that throughout our study, Tardbp lx/Ϫ mice were used as the control in comparison to the HB9:Cre-Tardbp lx/Ϫ mice. Like in the study of EIIa: Cre-Tardbp lx/ϩ mice (32), the levels of TDP-43 expression in the spinal cords of Tardbp lx/Ϫ and the wild-type mice, either 10-or 20-week-old, were similar (supplemental Fig. S1A).
Although all of the above clinical signs were similar between the HB9:Cre-Tardbp lx/Ϫ mice and the mutant hSOD1 transgenic mice, the disease progression after the symptom onset was somewhat slower in HB9:Cre-Tardbp lx/Ϫ mice. Significantly, HB9:Cre-Tardbp lx/Ϫ mice exhibited a kyphosis phenotype beginning at 20 weeks of age and it became severe at 24 weeks, as exemplified in Fig. 2D. The kyphosis likely resulted from weakening of the extensor thoracic paraspinal muscles, as observed in ALS patients (41) as well as in the mutant hSOD1 (G93A) transgenic mice (42). The apparent muscle weakness of HB9:Cre-Tardbp lx/Ϫ mice was accompanied with muscle atrophy as revealed by morphological analysis of the gastrocnemius and soleus muscles of the mice (supplemental Fig. S2). Finally, similar to ALS (3), the development of ALS-like phenotypes in HB9:Cre-Tardbp lx/Ϫ mice were also male dominant, with a male/female ratio of 3:1. The average lifespan of the HB9:Cre-Tardbp lx/Ϫ mice showing ALS-like phenotypes was 10 months.
Loss of Motor Neurons and Enhancement of Astrocytosis in the Spinal Cord of HB9:Cre-Tardbp lx/Ϫ Mice-ALS was defined as a neurodegenerative disorder characterized by progressive muscle weakness indicating the degeneration of motor neurons in the primary motor cortex, brainstem, and spinal cord. Progressive atrophy of the muscle fibers, which were denervated as their corresponding motor neurons degenerated, then led to weakening of the affected muscles. We reasoned that the restricted population of the spinal cord motor neurons in HB9: Cre-Tardbp lx/Ϫ mice might be affected by the depletion of TDP-43. Indeed, immunofluorescence staining and histology analysis of the spinal cords showed a decrease of ChAT-positive motor neurons among the ventral horn cells of 20-week-old HB9: Cre-Tardbp lx/Ϫ mice (Fig. 3A). Furthermore, among the remaining ChAT-positive motor neurons of HB9:Cre-Tardbp lx/Ϫ mice, 62% lost the immunostaining signals of TDP-43, as exemplified in the confocal images ( Fig. 3A and supplemental Fig. S3) and statistically presented in the histogram of Fig. 3A.
The amounts of TDP-43 and ChAT in the spinal cords of the mutant and control mice were also analyzed by immunoblotting. As shown in supplemental Fig. S1B, the difference between the amounts of TDP-43 of the spinal cords of Tardbp lx/Ϫ and HB9: Cre-Tardbp lx/Ϫ mice was barely detectable (supplemental Fig. S1B) because HB9-expressing cells only accounted for 5-10% of the total cell population of the spinal cord (33,34). On the other hand, the level of the ChAT protein was significantly reduced in the HB9:Cre-Tardbp lx/Ϫ spinal cord (supplemental Fig. S1B), consistent with loss of the ChAT(ϩ) spinal cord motor neurons in the mutant mice as described for Fig. 3A.
The numbers of motor neurons in the ventral horn of the lumbar (L3-L5) level of the spinal cords of mice were quantified with respect to the ␣and ␥-types. In the mammalian spinal cords, ␣ and ␥ motor neurons could be distinguished both by the cell body sizes, for which the ␥ motor neurons were significantly smaller than the ␣ motor neurons (43), and by the expression of specific molecular markers, e.g. NeuN for the ␣ motor neurons (44,45). As seen in Fig. 3B, Nissl staining showed that the numbers of both the ␣ and ␥ motor neurons, as defined by their relative soma sizes, in the lumbar regions of the spinal cords of 20-week-old HB9:Cre-Tardbp lx/Ϫ mice were decreased significantly when compared with the control mice.
In view of the reported changes of the cell body sizes of neurons from ALS or with TDP-43 proteinopathies (for example, Refs. 46 and 47), we also carried out immunofluorescence staining experiments in which both ChAT and NeuN were used as markers to identify the total, ␣-, and ␥-type motor neurons, respectively, in the spinal cords of 10-week-old (Fig. 4, upper panel) as well as 20-week-old mice (Fig. 4, lower panel). As seen, a significant loss of the motor neurons, either the ␣or ␥-type, was also observed in the symptomatic 20-week-old HB9:Cre-Tardbp lx/Ϫ mice (Fig. 4, lower panel), although the overall Nissl staining patterns (Fig. 3B) or NeuN staining patterns of the spinal cord sections were similar between the HB9:Cre-Tardbp lx/Ϫ and control mice. Notably, consistent with the report that larger ␣ motor neurons in ALS were more vulnerable to degeneration (48), the % of reduction of the ␣ motor neurons (ϳ46%) in the 20-week-old HB9:Cre-Tardbp lx/Ϫ mice was larger than that of the ␥ motor neurons (ϳ25%), as estimated with use of the molecular markers (Fig. 4, lower panel). Finally, a small decrease (ϳ10%) of spinal cord motor neurons could also be detected at the presymptomatic age of 10 weeks (Fig. 4, upper panel). This progressive loss of motor neurons during ALS pathogenesis of our mouse model was very similar to that observed before in the ALS mouse model with transgenic overexpression of hSOD1 (21). Altogether, the data of Fig. 4 suggested that the pathological symptoms of HB9:Cre-Tardbp lx/Ϫ mice were associated mainly with the loss of spinal cord motor neurons at the symptomatic on-set stage (20 weeks old).
Because reactive astrocytosis and microglia activation were often taken as indications of neuronal toxicity or neuronal death (49), we first quantified the level of the glial fibrillary acidic protein (GFAP) expression by immunofluorescence staining. As seen in Fig. 5, enhanced GFAP immunoreactivity was detected in sections of the lateroventral lumbar spinal cords of HB9:Cre-Tardbp lx/Ϫ mice when compared with the controls. Quantification indicated an ϳ2-fold increase of GFAP-positive cells in the spinal cords of HB9:Cre-Tardbp lx/Ϫ mice. Interestingly, microglia activation was also observed, as indicated by the enhanced immunoreactivity of ionized cal-cium binding adaptor molecule 1 (Iba1) in sections of lateroventral lumbar spinal cords of the HB9:Cre-Tardbp lx/Ϫ mice as compared with controls (supplemental Fig. S4).
Accumulation of Ubiquitinated Proteins in the Spinal Cord Motor Neurons with Elimination of TDP-43 Expression-One
hallmark of the diseased cells of TDP-43 proteinopathies was the accumulation of polyubiquitinated proteins, including TDP-43 itself, in the cytoplasm (29). To see if polyubiquitinated proteins would also accumulate, in the absence of TDP-43 in the spinal cord motor neurons of HB9:Cre-Tardbp lx/Ϫ mice, we carried out immunofluorescence staining experiments. When anti-ChAT and anti-ubiquitin were used for co-staining, strong anti-ubiquitin signals were observed in the lumbar spinal cord cells of the HB9:Cre-Tardbp lx/Ϫ mice but not in the control mice, as exemplified in Fig. 6A (Fig. 6A, compare the lower 2 rows to the upper 2 rows). Furthermore, when co-staining experiments were carried out with anti-ChAT, anti-TDP-43, and anti-ubiquitin, it was found that the cytoplasmic ubiquitinpositive deposits detected in spinal cords of 20-week-old HB9: Cre-Tardbp lx/Ϫ mice were mostly present in ChAT(ϩ)/TDP-43(Ϫ) cells (Fig. 6A), suggesting the accumulation of polyubiquitinated proteins in the spinal cord motor neurons with elimination of TDP-43. On the other hand, the accumulation of the immunofluorescent ubiquitin(ϩ) signals was not observed in the spinal cord motor neurons of 10-week-old mice, either the control or the HB9:Cre-Tardbp lx/Ϫ mice (data not shown).
Although ubiquitin(ϩ) cells in the spinal cords of the mutant mice (Fig. 6A) only accounted for a few % of the whole spinal cord cell population, we carried out immunoblotting to see if the increased ubiquitin signals came from polyubiquitinated proteins (Fig. 6B). Interestingly, most of the polyubiquitinated proteins isolated from the spinal cords of either the Tardbp lx/Ϫ mice (Fig. 6B, lanes 1 and 3) or the HB9:Cre-Tardbp lx/Ϫ mice (Fig. 6B, lanes 2 and 4) were in the urea-soluble fraction (Fig. 6B, lanes 3 and 4). Furthermore, the amount of polyubiquitinated proteins in the spinal cords of the HB9:Cre-Tardbp lx/Ϫ mice was severalfold higher than that of the Tardbp lx/Ϫ mice (compare Fig. 6B, lanes 3 and 4). The data of Fig. 6B suggested that most of the accumulated ubiquitin signals detected in spinal cord motor neurons of HB9:Cre-Tardbp lx/Ϫ mice (Fig. 6A) came from relatively insoluble, polyubiquitinated proteins. The data of Fig. 6 demonstrated that HB9-directed elimination of TDP-43 expression in mouse spinal cords could lead to the accumulation of polyubiquitinated proteins at the symptom onset stage but not the pre-symptom stage.
DISCUSSION
The results described above altogether indicated that the spinal cords, in particular the ventral horn of the lumbar spinal cord, of mutant HB9:Cre-Tardbp lx/Ϫ mice underwent significant motor neuron loss, reactive astrocytosis, microglia activation, and accumulation of polyubiquitinated proteins, all of which were characteristic of ALS with TDP-43(ϩ) UBIs. Because these HB9:Cre-Tardbp lx/Ϫ mice are the first mouse , and vice versa. Furthermore, separate analysis of the immunofluorescence staining data showed that whereas ϳ6% of the total cell population in the spinal cord of the HB9:Cre-Tardbp lx/Ϫ mice were Ub(ϩ), only 0.4% of the control Tardbp lx/Ϫ spinal cord cells were Ub(ϩ). B, representative immunoblots of polyubiquitinated proteins in fractionated spinal cord extracts. The total protein extracts were isolated from the spinal cords of Tardbp lx/Ϫ control and HB9:Cre-Tardbp lx/Ϫ mice, fractionated into the RIPA-and urea-soluble fractions, and analyzed by immunoblotting with use of anti-ubiquitin (top) and anti-actin (bottom). Note that the increase of high molecular weight ubiquitinated protein species (***) are in the urea-soluble fraction of the HB9:Cre-Tardbp lx/Ϫ sample in comparison to the control sample. The arrows point to the two nonspecific bands on the anti-ubiquitin blots. model with targeted depletion of TDP-43 expression in the spinal cord motor neurons and 90% of the non-SOD1, non-FUSinclusion type ALS patients are signatures with TDP-43(ϩ) UBIs, this mouse model we have generated should be valuable for research of ALS with TDP-43 proteinopathies.
Like other neurodegenerative diseases with inclusion formation during pathogenesis, e.g. Alzheimer disease, Huntington disease, Parkinson disease, etc. (50), neurodegeneration in patients with the TDP-43(ϩ) UBIs has also been suggested to be the result of loss of functions of TDP-43, cytotoxicity of the TDP-43(ϩ) UBIs per se, or a combination of both (29,31). In vivo evidence supporting the first scenario includes observations of depletion of the nuclear TDP-43 in diseased cells of ALS and FTLD-U patients with TDP-43(ϩ) UBIs (14,15) and the loss of motor function (47,51) or learning ability (47) of Drosophila with knockdown of TDP-43. On the other hand, current evidence supporting the second scenario is mainly by association, which includes the cytotoxicity observed in cell cultures upon overexpression of the carboxyl-terminal 25-kDa fragment to form cytoplasmic TDP-43(ϩ) UBIs (52) and development of FTLD-U-like symptoms or motor-neuron degeneration in transgenic mice with overexpression of TDP-43 accompanied with formation of TDP-43(ϩ) UBI in vivo (24 -26, 28, 53-55). With respective to the above, the effects of the restricted knock-out of TDP-43 expression in spinal cord motor neurons of HB9:Cre-Tardbp lx/Ϫ mice on the development of a range of ALS-like phenotypes in these mice has demonstrated an essential role of TDP-43 in the survival and functioning of the mammalian spinal cord motor neurons. Furthermore, it also provides direct evidence that loss of functions of TDP-43 in spinal cord motor neurons, and perhaps in a minor fraction of other types of cells as well, is sufficient and thus could be one major cause leading to the pathogenesis of ALS with TDP-43(ϩ) UBIs.
The molecular and cellular basis of how loss of function of TDP-43 would lead to the development of TDP-43 proteinopathies is currently unknown and may take awhile to be fully understood in view of the multifunction nature of TDP-43. Notably, loss of the spinal cord motor neurons, albeit small, already could be detected at the presymptomatic stage of the HB9:Cre-Tardbp lx/Ϫ mice (Fig. 4, upper panel). Because HB9directed expression of Cre was as early as E9.5, this result suggested that TDP-43 is not required for biogenesis of the spinal cord motor neurons during early development. However, the maintenance/survival of the spinal cord motor neurons of the adult mice does involve TDP-43, the long term lack of which would lead to the progressive loss of these neurons. Finally, it is especially interesting to note that accumulation of ubiquitinated proteins could still occur in the cytosol of the spinal cord motor neurons with elimination of the TDP-43 expression of the HB9:Cre-Tardbp lx/Ϫ mice (Fig. 6, A and B).
These ubiquitin signals, however, are not present as inclusions or aggregates as seen in the diseased samples of ALS (14,15). Although, they do appear to be elevated from polyubiquitinated proteins in immunoblotting analysis (Fig. 6B). In any case, we suggest that accumulation of polyubiquitinated proteins in the diseased cells of TDP-43 proteinopathies might be independent of the formation of TDP-43(ϩ) UBIs. It is likely that the initial loss of function of TDP-43, as the result of mutations in some of the ALS-associated disease genes, including TDP-43 itself, may cause impairment of the proteasome system and/or autophagy thus leading to accumulation of polyubiquitinated proteins in the diseased cells. Interestingly, this scenario is supported by a recent study showing that TDP-43 is required for the maintenance of a functional autophagy in rodent cells (56). Upon depletion of the mouse TDP-43 and the consequent destabilization of the ATG7 mRNA, the autophagy is impaired and ubiquitinated proteins accumulate in the cytosol (56), in a strikingly similar pattern as observed in Fig. 6. In the presence of TDP-43, these polyubiquitinated proteins then could form the TDP-43(ϩ) UBIs with TDP-43 (57,58). To what extent and how generally applicable this proposed scenario is to the ALS cases with TDP-43(ϩ) proteinopathies awaits to be validated in the future.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T02:31:33.096Z", "created": "2012-06-20T00:00:00.000", "id": "33770863", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "http://www.jbc.org/content/287/33/27335.full.pdf", "pdf_hash": "8dc9df4f645f10ccb1bbf12fdc5f3424fee1fbfd", "pdf_src": "Highwire", "provenance": "peS2o-0075.json.gz:32996", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "c18ee116abff6c16bc37f21822aa16cb5e89c531", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The Extract of Ginkgo biloba EGb 761 Reactivates a Juvenile Profile in the Skeletal Muscle of Sarcopenic Rats by Transcriptional Reprogramming
Background Sarcopenia is a major public health problem in industrialized nations, placing an increasing burden on public healthcare systems because the loss of skeletal muscle mass and strength that characterizes this affection increases the dependence and the risk of injury caused by sudden falls in elderly people. Albeit exercise and caloric restriction improve sarcopenia-associated decline of the muscular performances, a more suitable and focused pharmacological treatment is still lacking. Methodology/Principal Findings In order to evaluate such a possible treatment, we investigated the effects of EGb 761, a Ginkgo biloba extract used in chronic age-dependent neurological disorders, on the function of the soleus muscle in aged rats. EGb 761 induced a gain in muscular mass that was associated with an improvement of the muscular performances as assessed by biochemical and electrophysiological tests. DNA microarray analysis shows that these modifications are accompanied by the transcriptional reprogramming of genes related to myogenesis through the TGFβ signaling pathway and to energy production via fatty acids and glucose oxidation. EGb 761 restored a more juvenile gene expression pattern by regenerating the aged muscle and reversing the age-related metabolic shift from lipids to glucose utilization. Conclusions/Significance Thus, EGb 761 may represent a novel treatment for sarcopenia both more manageable and less cumbersome than exercise and caloric restriction.
Introduction
Sarcopenia, a term coined by Rosenberg [1], is the major feature of the age-related decline in neuromuscular performances and is characterized by the loss of skeletal muscle mass as well as strength [2]. These morphological and functional modifications result from intrinsic events, such as changes in the muscle fiber type composition, mitochondrial dysfunction and oxidative damage, and from extrinsic factors including reduced physical activity and excessive and/or unbalanced nutritional intake [3]. Exercise and caloric restriction (CR), known to slow down the impairment of the aged muscle performances [4], are difficult to manage while a more specific pharmacological treatment for sarcopenia that would be more suitable for elderly people is still lacking.
Muscle wasting and weakness associated with sarcopenia may benefit from drugs that target neurodegenerative diseases, since sarcopenia shares several features with different age-related neuromuscular disorders [5]. In this perspective, one of such drugs is EGb 761, a Ginkgo biloba extract, which has been shown to be an effective treatment in chronic age-dependent neurological disorders such as, for example, Alzheimer's disease [6,7]. Therefore, we treated with EGb 761 aged sarcopenic rats in order to evaluate whether this compound, similarly to exercise and CR, improves age-impaired muscular functions. Biochemical and electrophysiological analyses show that EGb 761 has a protective effect on muscular wasting and improves soleus muscle isometric contractile force. These effects are associated, as shown by DNA microarray analysis, with the restoration of a more juvenile transcriptional activity of genes relevant for regeneration and energy production processes in the soleus of aged rats. This is the first time, to our knowledge, that EGb 761 has been shown to benefit the aged muscle in mammals suggesting that it may represent a novel treatment for sarcopenia.
EGb 761 Affects Total Body and Muscle Weights
At the beginning of the experiment, 22 month-old rats were randomly separated in two groups and were weighted at regular intervals. The first group, Aged Control (AC), did not receive any treatment, while the second group, Aged Treated (AT), received EGb 761 in the drinking water. After 5 weeks of this regimen, the mean body weight of the AC group was increased by 2.2% while the rats in the AT group lost 4% (p = 0.02) of their initial body weight, indicating that EGb 761 counteracts the weight gain naturally occurring in aged laboratory rats (Table 1).
In order to ascertain whether this effect of EGb 761 was correlated with differences in muscle anatomic-functional features, we proceeded to further analyses and focused our study on the soleus muscle, a slow twitch muscle (type I fiber) with predominant oxidative metabolism. The animals were euthanized and the soleus muscles dissected and weighted. The soleus mean weight was increased by 6.5% in the AT compared to the AC group of rats. Moreover, the ratio of the soleus weight reported to the total body weight was increased by 13% (Table 1).
This result suggests that EGb 761 affects body and muscle weight by increasing the muscular mass and, thus, reversing toward the juvenile values the age-related decrease of the ratio of muscle v/s body weight. In order to evaluate whether this phenomenon was also associated with variations of a biochemical index of age-related muscle wasting processes, we assessed the effect of the EGb 761 treatment on the circulating creatine kinase.
EGb 761 Decreases Serum Creatine Kinase
We observed a striking decrease in the serum creatine kinase (CK) levels in AT rats (3716223 U/L), compared to AC rats (7376178 U/L). Interestingly, the CK value obtained in the AT rats was closely similar to the value observed in a group of young (Y) 4 month-old rats (370653 U/L). Since CK serum levels act as a peripheral marker of muscle damage, these findings further suggest that EGb 761 has a protective effect against the damages that affect the aging muscle. In order to ascertain whether this protective effect was also associated with increased muscular performances, we compared the contractile force of the soleus muscle under different conditions in the AC and AT groups of rats.
EGb 761 Enhances the Isometric Contractile Force
We measured the maximal force developed by the soleus muscles of AC and AT rats, either during a single intermittent isometric twitch or during tetanic contractions, evoked by direct muscle stimulation and through nerve stimulation (Fig. 1, a-d). The isometric twitch and tetanus force developed by individual AT muscles, evoked by direct muscle stimulation, were compared to the mean values for the AC muscles, normalized to the weight of the muscle (Fig. 1, e, f). The mean ratio between AT versus AC was 2.1 for twitch muscle force evoked by direct stimulation (Fig. 1e) and 1.46 for tetanic contractions at 40 Hz (Fig. 1f), indicating an improvement of muscle force. For nerve-evoked contractions, similar values were obtained with a mean ratio of 2.3 for the single twitch (Fig. 1e) and 1.4 for tetanic contraction at 40 Hz (Fig. 1f).
Furthermore, analyses of the peak force-frequency relationship during direct tetanic stimulation (40-100 Hz) revealed that AT soleus muscles develop significantly higher peak force at 40 and 60 Hz when compared to AC muscles (Fig. 2a). Also, a significant change on nerve-evoked tetanus peak force was detected at 40 Hz in AT muscles when compared to controls (Fig. 2b). Taken together, these results indicate that isometric force developed by soleus muscle from EGb 761-treated aged rats is improved with respect to the same-age controls, with an enhancement dependent on the stimulation frequency since no significant improvement was detected at the higher stimulation frequencies studied.
Thereafter, in order to gain insight on the molecular mechanisms underlying these anatomical, biochemical and physiological modifications elicited by EGb 761, we analyzed the global gene expression in the rat soleus muscle by DNA microarray analysis. The results already obtained point out to some precise phenomena based on regeneration or energy production amelioration for the improvement of muscular functions elicited by EGb 761. Therefore, the analysis of DNA microarray experiments investigating molecular events associated with aging and with EGb 761 treatment on global gene transcription was focused on genes implicated in these processes on a limited number of rats for each group.
EGb 761 Treatment Elicits Transcriptional Reprogramming in the Soleus Muscle of the Aged Rat
The DNA microarray analysis identified a total of 1518 genes, out of 9906, modulated either by aging or EGb 761 or both. Aging affected the expression of 1015 genes, 2/3 of them, for a total of 626, were down-regulated, while the remaining 389 were up-regulated, as assessed by comparing the old sarcopenic (AC) versus the young group (Y) of rats (Table S1). EGb 761 affected the expression of 618 genes in nearly equivalent proportions; 333 genes were up-regulated and 284 genes were down-regulated when comparing the treated (AT) versus the AC group of aged rats (Table S2). Among the genes modulated by EGb 761, a total of 115 were Values represent the mean6S. E. M. Weights are expressed in g for total body and in mg for the soleus muscle. The ratio represent the muscle weight (mg) divided for the total body weight (in g) for the different rat groups.doi:10.1371/journal.pone.0007998.t001also modified by aging (Fig. 3). The DNA microarray data has been loaded into the ArrayExpress database (ArrayExpress accession: E-MEXP-1937, release date: 2009-06-30).
The microarray data was further investigated using hierarchical clustering analysis based on the overall similarity of the variations in the genes expression patterns. Permutation analysis was conducted on the three different groups of rats altogether. About 2/3 of the genes modulated by EGb 761 or Aging showed more than 30% of expression variability in each of the 3 groups. These genes were classified as non exploitable (NE) and were not used for the clustering analysis (see Methods) (Fig. 3). Using this very stringent approach, four different clusters (A to D) were identified by Pearson correlation analysis, corresponding to the different types of gene expression behavior in response to aging and/or treatment (Fig. 3). The first cluster assembles genes that are differentially expressed during aging and, then, reverted toward the young expression level by EGb 761 (Cluster A, Table S3). The second cluster contains genes that display an additive effect of aging and EGb 761 on their expression, that is genes downregulated or up-regulated during aging and whose expression was further decreased or increased, respectively, by EGb 761 (Cluster B, Table S4). The two remaining clusters assemble genes whose expression is modified by the treatment but not by aging (cluster C, Table S5), and genes modulated by aging but not by the treatment (cluster D, Table S6). Since we were interested in the effect of EGb 761 on aging, we focused our analysis on the cluster A and B, while the clusters C and D were not further investigated.
Thus, according to the results obtained by clustering analysis, EGb 761 modified the expression of 30% (115 out of 378) of the genes affected by aging. This result stresses the impact of the Ginkgo biloba extract on aging since these genes represent almost 60% (115 out of 198) of the genes regulated by EGb 761.
Validation of the DNA Microarray Results by qPCR on Selected Genes
In order to validate the DNA microarray results before further analysis, we analyzed the m-RNA expression levels of 13 selected genes having different levels of expression by real-time quantitative RT-PCR (qPCR) in the soleus muscle, including those used for the microarray.
The comparison of the pattern of expression in the three different groups of rats between the DNA microarray and the qPCR analysis established that all the genes displaying variations in their expression pattern in the transcriptomic analysis were also showing concomitant variations by qPCR analysis (Fig. 4). Moreover, among these genes Acvr1, AdipoQ, Bambi, Col5A, Fasn, Fst and Gmfg, had a similar pattern of variation by DNA microarray and qPCR analysis both when comparing the young with the aged group of rats as well as when comparing the aged with the aged-treated group of rats. Among the remaining genes, Cd36, Id1 and Ucp3, displayed a similar pattern when comparing the young with the aged group of rats and Areg, Id3 and Tgf-b, when comparing the aged with the aged-treated group of rats (Fig. 4). Therefore, a majority of 20 out of 26 (77%) of the comparison made with the 3 groups of rats for these 13 genes were confirmed by qPCR, validating for further analysis the DNA microarray data.
The EGb 761 Affects Genes Implicated in Regeneration and Energy Production Functions
We classified all the genes belonging to clusters A and B according to their biological functions, using the Ingenuity pathway analysis software and the gene ontology database. The most relevant of these functions, according to the number of genes modified both by aging and EGb 761 treatment, are related to cell cycle, cell death, growth and proliferation, as well as metabolism (Table 2). Among the most relevant genes, there are cell division control protein 42 (Cdc42) and transforming growth factor b2 (Tgfb2), which are ubiquitous factors that intervene also in the development and regeneration of the skeletal, nervous and connective tissues. Moreover, the EGb 761 treatment was also associated with the upregulation by 50 to 70% of genes specifically implicated in skeletal muscle development such as Follistatin (Fst), Follistatin related protein (Fstl1), BMP and activin membrane-bound inhibitor (Bambi), a2-macroglobulin (A2m), Activin receptor type I (Acvr1), embryonic Myosin heavy chain (Myh3) and Ryanodine receptor 3 (Ryr3). Since several of these genes participates also in the TGFb signaling pathway, which plays a relevant role in the muscle satellite cell proliferation and differentiation [8], these results suggest that the EGb 761 modifies gene expression in a manner consistent with the remodeling of the sarcopenic muscle via the modulation of the myogenesis (Fig. 5).
The other most relevant effect of EGb 761 is related to the transcriptional reprogramming of genes implicated in metabolic processes in mitochondria. Namely, EGb 761 induced the downregulation of the transcripts for the Fatty acid synthase (Fasn), responsible for fatty acid synthesis, the Acetyl-Coenzyme A acyltransferase2 (Acaa2), involved in cholesterol and lipoprotein synthesis and fatty acid elongation, the Hexokinase 1 (Hk1) and the Phosphofructokinase 1 (Pfk1), both implicated in glucose metabolism. The treatment induced as well the up-regulation of the transcripts for the Pyruvate dehydrogenase kinase 4 (Pdk4) and the Fatty acid translocase (Cd36) involved in lipid metabolism and transport, respectively, and the Uncoupling proteins 1 and 3 (Ucp1 and Ucp3), relevant for energy production (Table 2).
These findings are consistent with a previous observation that the EGb 761 modifies the mitochondrial functions in several tissues [9] and suggest that EGb 761 coordinates the expression of genes that enhance mitochondrial energy production by increasing lipid beta oxidation and inhibiting glucose oxidation (Fig. 6).
Discussion
Sarcopenia is a major public health problem in industrialized nations, placing an increasing burden on public healthcare systems. The progressive loss of muscle mass associated with advancing age that is characterized by the slowing of movement and a gradual decline in muscle strength, increase dependence and the risk of injury from sudden falls in elderly people. Although caloric restriction and physical exercise have been shown to be effective treatments for sarcopenia, their effect is limited to slowing down the progressive loss of muscle mass and functions and their implementation is problematic in elderly people. Therefore, drug treatments appear to be more appropriate for sarcopenia, albeit, until now, no pharmacological therapy has been reported apart from the suggestion that leucine-rich dietary components may help the elderly to preserve muscle mass [10]. However, since the ageing process manifests itself with similar features in different tissues, it has been suggested that drugs for age-related neurodegenerative diseases may also be beneficial for sarcopenia [5]. In this context, EGb 761(IPSEN), an extract of Ginkgo biloba, has been shown to extend life span of caenorhabditis elegans [11] and to enhance learning and longevity in rats [12]. Also, in vivo and in vitro experiments have demonstrated the efficacy of EGb 761 in protecting against age-related processes such as increase of oxidative stress, brain mitochondrial dysfunction [9] and chronic age-dependent neurological disorders [6,7]. Here, we demonstrate that EGb 761 improves muscle performances, as assessed by electrophysiological studies, and, quite remarkably, restores a ''juvenile'' profile in the soleus muscle of old rats by modulating the transcriptional expression of genes implicated in development/ regeneration and energy production pathways.
We have focused our study and the DNA microarray analysis on the soleus since this muscle is, in the rat and humans, almost homogeneously composed by slow twitch type I fibers. Thus, the age related changes are easier to analyze as they do not result from fiber type modifications as, for example, in the gastrocnemius and/or other mixed fiber type muscles. The transcriptomic analysis revealed that EGb 761 up-regulates TGFb and several other genes such as Fst, Fstl1, Bambi and Acvr1 that participate to the modulation of the TGFb signaling pathway during myogenesis. Tgfb2 contributes, with other factors of the TGF super-family such as BMP, Activin and Myostatin, to myogenesis [13] (Fig. 5) and may participate to the regeneration process that is impaired by aging, resulting in the incomplete structural and functional recovery after muscle injuries in elderly people [14]. It is noteworthy that TGF-b2 plays a contextual role in myogenesis, since it inhibits proliferation and differentiation but, at the same time, promotes the fusion of the satellite cells depending on the presence or absence of other factors [15].
Fst, one of the genes up-regulated by EGb 761 and associated to the TGFb signaling pathway, acts as a positive regulator of the muscle differentiation [16], since, by binding to myostatin, it prevents the myostatin-induced inhibition of proliferation and differentiation [17]. Fst expression increases with age as a protective mechanism in response to the parallel elevation of myostatin, albeit this compensatory effect is ineffective in preventing sarcopenia [18]. Accordingly, our data show an important increase (3 fold) in the expression of Fst in the aged rats. However, after EGb 761 the over-expression of this transcript is further boosted by 2 fold (see progression cluster). Moreover, according to our results, others transcripts such as Bambi, Fstl1 and A2m, all inhibitors of BMPs and Activin [19], are also up-regulated by EGb 761 but not by aging, suggesting that the concomitant action of all these factors on the TGFb pathway mediates the regenerative effects of EGb 761.
The transcriptomic analysis highlights another major effect of EGb 761 upon transcripts, such as Fasn, Acaa2, Cd36, Hk1 and Pdk4, coding for key mitochondrial enzymes associated to lipid and glucose metabolism and involved in Krebs cycle, b2-oxidation, fatty acid transport and oxidative stress (Table 2). During aging the mitochondria shift from lipids to glucose as substrate for energy production, which results in increased intramuscular fat accumulation [20,21]. EGb 761 decreased the expression of Fasn, involved in FA synthesis, and Acaa2, implicated in cholesterol and lipoprotein synthesis and FA elongation, and increased the expression of Cd36, a FA transporter present in the sarcolemma, which promotes FA entry in the muscle cell [22]. Since Fasn increases blood triglycerides and, with Acaa2, favors the accumulation of intramuscular adipose tissue, the down-regulation of these genes could instead facilitate mitochondrial lipid oxidation (Fig. 6).
The striking four-fold increase of Ucp3 transcripts observed by the DNA microarray data analysis further stresses the effect of EGb 761 in optimizing FA utilization. UCP3 is an uncoupling protein, preferentially expressed in skeletal muscle, whose overexpression increases the activity of several key mitochondrial enzymes associated with FA oxidation, and decreases intramuscular triglyceride content [23]. Also, UCP3 lowers mitochondrial membrane potential, notably through fatty acid anion export from the matrix, and, consequently, decreases reactive oxygen species The up-regulated genes are presented in bold characters and the down-regulated genes are in plain characters.doi:10.1371/journal.pone.0007998.t002 (ROS) and prevents mitochondrial damage [23]. Interestingly, the increase of UCPs activity correlates with extended lifespan in longliving avian species [24,25] and in mice [26].
As a corollary to the facilitation of FA oxidation, EGb 761 also inhibited glucose oxidation in mitochondria by down-regulating Hk1, associated with glycolysis, and by up-regulating the glucokinase regulatory protein, which inhibits HK1, and Pdk4, which inhibits the pyruvate dehydrogenase complex (Fig. 6). Taken together these modifications contribute to the reprogramming of the ''glucose-fatty acid cycle'', in which free FA compete with glucose for mitochondrial oxidation [27], toward an increased utilization of FA, resulting in the increase of energy production and the reduction of mitochondrial ROS generation and of intramuscular fat storage (Fig. 6). Other studies on gene expression that used caloric restriction (CR) or exercise training for preventing sarcopenia have reported that CR either stimulated glycolysis, by the up-regulation of enzymes such glucose-6phosphate isomerase, and FA synthesis, by over-expressing Fasn [28][29][30], or, in agreement with our findings, induced a transcriptional switch toward an increase of FA metabolism and a reduction of lipid biosynthesis in the muscle [31]. Exercise training as well, similarly to EGb 761, improves mitochondrial FA uptake and oxidation, by reducing the lipid stores in the muscle [32,33]. Thus, EGb 761 produces effects comparable to CR or exercise training and represents a more manageable option for the prevention and treatment of sarcopenia.
Besides these major effects, EGb 761 elicited also the overexpression of transcripts that are usually present during regeneration processes such as Myh3 [34] and Ryr3. The Ryr3, a minor form of ryanodine receptor, regulates the resting free Ca 2+ concentration and, thus, may favor a sustained contraction with increased force in the skeletal muscle [35]. In addition, the upregulation of the transcripts for Myh10 and Mybph also contribute to the increase of the muscle strength.
The two-fold increase of Tenascin C (Tnc) gives another example of how EGb 761 may modulate transcripts that are relevant for muscle strength. Tnc, in the adult muscles, is an extracellular matrix (ECM) molecule that localizes in the tendon along with other components of the ECM such as the collagen I and III, playing an important role in the force transmission and tissue structure maintenance. The collagen constitutes 60-85% of the tendon dry weight and is predominantly type I (,60%) arranged in tensile-resistant fibers, types III (#10%), IV (,2%) [36,37], V, and VI [38]. In particular, the type I collagen, carrying fibrous tensile strength, allows the transmission of force generated by skeletal muscle into movement [39]. Interestingly, the collagen I, V, VI expression is increased between 50% and 100% by EGb 761. Therefore, these components of the ECM may reorganize the structure of the tendon and participate to the improvement of muscle strength and body stability, whose impairment represents the main cause of falls in elderly people.
Finally, mitochondrial function improvement represents the main factor for the enhancement of muscular strength after EGb 761 treatment. Mitochondria play a central role in the production of cellular energy in the form of ATP, but are also key regulators of apoptosis, which contributes to the aging processes underlying sarcopenia. The age-related reduction in ATP production is responsible of the decrease in contractile force observed during aging. EGb 761 contributes to a better ATP production and availability by improving FA transport and utilization.
In conclusion, EGb 761 may represent a novel therapeutic intervention, endowed with a rejuvenation effect mediated by transcriptional reprogramming of a majority of genes implicated in aging, and with several advantages over CR and exercise to prevent sarcopenia in elderly people. Among these advantages there is a more easily implementation and therefore a better compliance with the treatment. Moreover, the effects of EGb 761 are mediated both by the restoration of the expression of genes that are pathologically down-regulated by aging and the boosting of the expression of genes that are physiologically up-regulated to counteract the effects of the aging process. This is another advantage over exercise and CR whose action is limited to slowing down the muscular ageing progression. It is also noteworthy that this effect may be unique to Ginkgo biloba extract. Indeed, Ginkgo biloba is one of the oldest life forms and contains many molecules (like ginkgolides) that do not exist in other organisms and are extremely difficult to synthesize. Thus, the natural extract of such a living fossil, shaped by natural selection during several millions of years, may contain several unique molecules that are less disruptive than synthetic products and better suited for fitting with biological systems [40].
EGb 761 Treatment of Rats
The Ginkgo Biloba extract (EGb 761), provided by IPSEN laboratories (France), was administered in the water at a dose of 75 mg/Kg body weight for two months. During the 60 days of the experiment, the animals were weighed each week and the volume of drinking solution was measured. The AC and AT groups were sacrificed at 23 months of age, young rats were sacrificed at 4 months.
Determination of Serum Creatine Kinase
Blood samples were taken from the hearts of anaesthetized rats, immediately before sacrifice. Activities of serum CK were determined using a Biomerieux kit (enzyline CK NAC optimized 10).
Measurement of Soleus Contractility
Rats were anesthetized with Isoflurane (AErraneH, Baxter S. A., Lessines, Belgium) inhalation before being sacrificed by section of the carotid arteries.
Soleus muscles with their attached nerves were carefully removed, mounted in a silicone-lined bath filled with Krebs-Ringer solution continuously perfused with 95% O 2 and 5% CO 2 throughout the experiment. The Krebs-Ringer solution was maintained at 2260.5uC instead of 37uC in order to optimize the experimental conditions, since at a temperature of 37uC the risk of nerve conduction failure is greatly increased. One of soleus tendons was securely anchored onto the silicone-coated bath, the other tendon was attached to an FT03 isometric force transducer (Grass Instruments, West Warwick, USA).
Muscle twitches and tetanic contractions were evoked either by stimulating the motor nerve via a suction microelectrode, with supramaximal current pulses of 0.15 ms duration, at frequencies indicated in the text, or by direct electrical stimulation. The resting tension was adjusted for each preparation investigated with a mobile micrometer stage (to allow incremental adjustments of muscle length) in order to obtain maximal contractile responses. Signals from the isometric transducer were amplified, collected, and digitized with the aid of a computer equipped with a DT2821 analogue to digital interface board (Data Translation, Marlboro, USA), as previously reported [41]. Data acquisition and analysis were performed with a program kindly provided by Dr. John Dempster (University of Strathclyde, Scotland). All data are expressed as the mean6SEM. The statistical significance of differences between controls and test values was assessed with Student's t-Test (two-tailed). Differences were considered significant if /P/,0.05.
Tissue Processing
For microarrays and qPCR experiments the soleus muscles from each animal were excised, snap-frozen in liquid nitrogen and stored at 280uC.
RNA Extraction
Total RNA was isolated from an aliquot of muscle thawed on ice; and extraction was performed by a silica gel-based purification method (RNeasy Mini Kit, Qiagen). For qPCR, total RNA was subjected to DNAse treatment (Qiagen) 30 min at room temperature. Total RNA yield was measured by the OD 260 , and the quality was evaluated on nanochips with the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA).
Microarrays and analysis
Gene expression levels were measured by using rat oligonucleotide arrays as described previously [42]. The microarrays contain 9906 different genes (CodeLink system, Uniset rat I, GE Healthcare Europe GmbH, Freiburg, Germany). Each experimental group, Aged Treated (AT), Aged Control (AC) and Young (Y) rats, was formed by 2 rats (biological replicates) and the total RNA isolated from the soleus muscle was hybridized in duplicate (technical replicates). Thus 4 microarrays were utilized for each group of rats.
After hybridization and washing, the slides were scanned using a Genepix 4000B scanner (Axon, Union City, USA) and Genepix software. The scanned image files were analyzed using CodeLink expression software, version 4.0 (GE Healthcare).
CodeLink software was used to normalize the raw hybridization signal. The threshold of detection was calculated using the normalized signal intensity of the 100 negative controls, spots with lower signal intensities were termed ''absent''.
ESTs were analyzed by BLAST research on the NCBI nucleotides databases. The clusters were generated using Pearson correlation. Statistical comparison and filtering were performed using Genespring software 7.0 (Agilent).
A first selection of genes was performed by pair-wise comparisons between AT and AC rats (Egb761 effect) or Y and AT rats (Aging effect). Each sample from one group was compared with each sample from the other group. This permutation analysis comparing 3 different group allows for strengthening the relevance of the findings when a given differences is found across different groups. To identify significant changes, the following criteria were used: 1) p-value less than 0.05; 2) a fold change of 1.3 was used to define a set of significantly up-and down regulated genes; 3) the over 1.3 fold change should be reproducible across all comparisons. Therefore a given mRNA transcript was considered as differentially expressed in the comparison of any two samples if the difference gave a P value of #0.05 in the Welch ANOVA parametric test, using a multiple test correction (Benjamini and Hochberg False Discovery Rate). Moreover, to avoid false positive genes, in particular for genes with small variation between groups, we removed from the two lists above genes with variability $30% inside each of the three groups (Y, A and AT), those genes were classified as non exploitable (NE). Thus, a gene was considered for the clustering analysis only if it met the above criteria in all pairwise comparisons and if the detected signal was above the background for at least one of the compared groups, thereby carrying a statistically significant absolute call 'present' or 'marginal' in all samples.
qPCR and Analysis
The qPCRs were performed as described previously [42] with some minor modifications. The primers were purchased from Eurogentec (Seraing, Belgium). All primers had Tms between 59 and 61uC and all products were 100 to 150 bp long. GAPDH and cyclophilin are used as internal standards to control amplification variations due to differences in the starting mRNA concentration.
Figure 1 .
Figure 1. Contractility studies. Representative examples of isometric contractile force in isolated soleus muscle from aged controls (a, c), and EGb 761-treated rats (b, d). In a and b the recordings were obtained by direct muscle stimulation and in c and d by nerve stimulation. In each recording (a to d) are shown evoked twitches (0.2 Hz stimulation) and tetanic contractions (40 Hz, 600 ms stimulation). The force is expressed in Newton (N). Histograms represent mean values6S. E. M. of the force (expressed in N/g) for the single twitch (e) and for tetanic stimulation (f). Mean ratio values between treated and controls are indicated in respective boxes with number of values (n) used for calculations.* Significantly different from controls p,0.05.doi:10.1371/journal.pone.0007998.g001
Figure 2 .
Figure 2. Force-frequency relationship. Peak force during isometric tetanic responses in isolated soleus muscles from aged controls (#) and EGb 761-treated rats (N) obtained by direct muscle stimulation (A) and by nerve stimulation (B) at frequencies indicated (abscissa) during trains of 600 ms duration. An interval of 30 s was used between trains. Each circle represent the mean6S. E. M. (n = 3).* Significant different from controls p,0.05.doi:10.1371/journal.pone.0007998.g002
Figure 3 .
Figure 3. Microarray data analysis and clustering. Comparison of gene expression profiles of soleus muscle in young rats (Y), aged control rats (AC) and aged treated rats (AT). Genes are clustered using Pearson's correlation. Down-regulated genes are represented in green, up-regulated genes, in red, number of genes is reported in parentheses.doi:10.1371/journal.pone.0007998.g003
Figure 4 .
Figure 4. Graphic representation of gene expression variations. Comparison between variations of gene expression in the different groups of rats according either to DNA microarray analysis (blue lines) or to qPCR (red lines). Values of the AC (aged control) group are standardized to 1 for each gene.doi:10.1371/journal.pone.0007998.g004
Figure 5 .
Figure 5. Schematic representation of the interactions between the genes regulated by EGb 761 associated to myogenesis/ regeneration in the soleus muscle of aged rats. Regulated genes are included in frames.8 indicates up-regulation of gene expression by EGb 761.ˆindicates down-regulation of gene expression by EGb 761. R = activation.--I = inhibition. Activins and BMPs may be diverted into alternative pathways through interaction with soluble and membrane-bound binding proteins Activin, myostatin and BMPs signal via type II activin receptor. The access of these TGF-b2 superfamily members to their type II receptors is blocked by extracellular binding proteins (follistatine, Fstl1 and a2macroglubulin) and membrane-bound pseudoreceptors (BAMBI for activin signaling).doi:10.1371/journal.pone.0007998.g005 All animal procedures were performed under authorizations of the ''Direction des Services Ve ´te ´rinaires de Paris'' of the ''Prefecture de Police de Paris'' and strictly followed the guidelines for the ethical treatment of the animals set forth by French law (decree Nu 87-848 dated 11/19/1987 and modified by the decree Nu 2001-464 dated 5/29/2001). The study was conducted with male Wistar rats on a group of young, 4-months old, rats (Young group = Y), and two groups of aged, 23-month old, rats. The aged rats either received (Aged Treated group = AT) or did not receive (Aged Control group = AC) EGb 761 treatment.
Figure 6 .
Figure 6. Schematic representation of the interactions between the genes regulated by EGb 761 associated with regulation of energy metabolism in mitochondria in the soleus muscle of aged rats. Regulated genes are included in frames.8 indicates upregulation of gene expression by EGb 761.ˆindicates down-regulation of gene expression by EGb 761. R = activation.----I = inhibition. FA (Free Fatty Acids) are taken up by the fatty acid transporter protein Cd36 and activated to fatty acyl-coenzyme A (FA-CoA). FA-CoA oxidation would increase the ratios of acetyl-CoA/CoA and of NADH/NAD+, which inhibit the pyruvate dehydogenase (PDH) complex by activating the pyruvate dehydrogenase kinase 4 (Pdk4). Increase FA oxidation products as citrate should further inhibit phosphofructokinase (Pfk) and hexokinase. These changes would slow down oxidation of glucose and pyruvate and glycogen stock is maintained. In the same time, fatty acid synthase (Fasn) and mitochondrial 3-oxoacyl-Coenzyme A thiolase (Acaa2) are inhibited resulting in the lipid synthesis inhibition and a preferential use of these lipids for ATP synthesis.doi:10.1371/journal.pone.0007998.g006
Table 1 .
Body and soleus muscle weight.
Table 2 .
List of the genes modulated both during aging and after EGb-treatment classified according to the Gene Ontology database.
==
Domain: Biology Medicine
|
[{"added": "2017-04-19T03:04:01.852Z", "created": "2009-11-24T00:00:00.000", "id": "585973", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0007998&type=printable", "pdf_hash": "1b30bbd8a8648c644e9d4fd461f1ec15142b23a9", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:31308", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "1b30bbd8a8648c644e9d4fd461f1ec15142b23a9", "year": 2009}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
IL-1β Signaling Promotes CNS-Intrinsic Immune Control of West Nile Virus Infection
West Nile virus (WNV) is an emerging flavivirus capable of infecting the central nervous system (CNS) and mediating neuronal cell death and tissue destruction. The processes that promote inflammation and encephalitis within the CNS are important for control of WNV disease but, how inflammatory signaling pathways operate to control CNS infection is not defined. Here, we identify IL-1β signaling and the NLRP3 inflammasome as key host restriction factors involved in viral control and CNS disease associated with WNV infection. Individuals presenting with acute WNV infection displayed elevated levels of IL-1β in their plasma over the course of infection, suggesting a role for IL-1β in WNV immunity. Indeed, we found that in a mouse model of infection, WNV induced the acute production of IL-1β in vivo, and that animals lacking the IL-1 receptor or components involved in inflammasome signaling complex exhibited increased susceptibility to WNV pathogenesis. This outcome associated with increased accumulation of virus within the CNS but not peripheral tissues and was further associated with altered kinetics and magnitude of inflammation, reduced quality of the effector CD8+ T cell response and reduced anti-viral activity within the CNS. Importantly, we found that WNV infection triggers production of IL-1β from cortical neurons. Furthermore, we found that IL-1β signaling synergizes with type I IFN to suppress WNV replication in neurons, thus implicating antiviral activity of IL-1β within neurons and control of virus replication within the CNS. Our studies thus define the NLRP3 inflammasome pathway and IL-1β signaling as key features controlling WNV infection and immunity in the CNS, and reveal a novel role for IL-1β in antiviral action that restricts virus replication in neurons.
Introduction
West Nile virus (WNV) is a single stranded, positive sense RNA virus of the flaviviridae family, and is a prototypical Flavivirus related to Yellow fever virus, Tick borne encephalitis virus, Japanese encephalitis virus (JEV) and Dengue virus [1][2], all of which are major public health threats. Among these viruses, WNV has emerged into the Western hemisphere and continues its spread through into North America [3]. WNV is normally maintained in mosquito and avian reservoirs, with infection of human and other animals occurring through contact with infected mosquitoes [4][5]. Infection is largely controlled acutely; however WNV can spread to the central nervous system (CNS), leading to encephalitic disease and death [6][7]. Overall however, the immune processes within the CNS that serve to control WNV infection and pathogenesis are not well defined.
WNV pathogenesis has been studied in murine models of infection to show that the virus initially replicates in epithelial cells and skin Langerhans dendritic cells (DCs) at the site of mosquito inoculation [8]. The virus then traffics to the draining lymph node, leading to secondary viremia and infection of the spleen where it can replicate in macrophage and DC subsets [5,9]. After amplification in peripheral tissues, WNV spreads to the CNS, where it replicates in neurons, causes neuronal destruction, and imparts inflammation leading to encephalitis that is comparable to human disease [5][6][7]. Both innate and adaptive immune defenses serve to control tissue tropism and initial spread of virus into to the CNS [10][11][12][13], while T lymphocyte responses are involved in mediating clearance of virus following CNS entry [14][15]. In particular, CD8 + T cells are thought to be the main contributors to late CNS clearance of WNV through mechanisms involving IFNc, TNF-a and perforin [13][14]16].
The inflammatory response is a key component in protective immunity against WNV infection. However, this response must be kept in check to limit bystander destruction of both peripheral and CNS tissues. For example encephalitis, which is marked by inflammatory cell recruitment to the CNS, has been shown to be both protective as well as destructive to CNS tissue during WNV infection [17]. Recruitment of populations of peripheral CD45 + leukocytes into the CNS has been shown to be important for limiting WNV pathogenesis [18][19][20]. In contrast, in other studies CD45 + leukocytes were shown to enhance susceptibility to WNV, likely due to increased immune-pathology associated with inflammatory cell-mediated destruction of CNS tissue. Thus, while inflammation is required for clearance of WNV in the CNS, the timing and magnitude of this inflammatory response must be tightly regulated to avoid off target pathology.
IL-1 signaling is involved in multiple aspects of the immune response to infection including immune regulation of inflammation, modulation of adaptive immune programs and direct antiviral control of pathogens [21][22][23][24]. These processes are driven by IL-1a and IL-1b which signal through the IL-1R1 (IL-1Ra) and MyD88 to drive downstream NF-kB activation and subsequent expression of genes whose products regulate the immune response to infection [21,[25][26]. Activation of IL-1 requires two distinct signals, ''signal 1'' which drives mRNA expression in an NF-kB-dependent manner and ''signal 2'' which processes the cytokine to its functional form [27]. ''Signal 2'' processing of IL-1b to its active form is mediated by inflammasomes, signaling structures comprised of NOD-like receptors (NLRs), adaptor molecules such as ASC and the effector, Caspase-1 [27][28]. In contrast, IL-1a is not processed by the inflammasome and is instead thought to processed and activated by other host protease pathways [29] suggesting that the majority of immune responses driven by inflammasome activation are mediated by IL-1b and not IL-1a.
Several distinct inflammasomes have been described based upon their inclusion of a specific NLR or signaling-initiator molecule, including the NLRP1, NLRC4, NLRP3, RIG-I and AIM2 [30]. Of these, the NLRP3 inflammasome has been the best characterized to mediate IL-1b secretion in response to RNA viruses in vitro and in vivo [24,31]. Inflammasome activation and IL-1b signaling are important for immunity against several viruses including Influenza A, hepatitis B, Sendai and vesicular stomatitis virus (VSV) [24] and drive host responses that regulate cellular infiltration to sites of infection [32][33][34][35], adaptive immunity [23,36] and direct viral control in combination with other host factors such as IFN-a/b (type I IFN), IFN-c and TNF-a [21][22]37]. In the context of CNS infection, IL-1 signaling has been associated with both protection and enhancement of disease. The synergistic effects of IL-1b and TNF-a were associated with protection against encephalitis by the neurotropic virus HSV-1 [35] while IL-1b 2/2 animals showed increased pathogenesis and lethality to Sindbis virus [38][39]. Thus, IL-1 signaling likely functions in a context-dependent manner to control or exacerbate disease. Little is known about IL-1 signaling in immunity to flavivirus infection. Recently, JEV was shown to trigger IL-1b secretion from astrocytes and microglia [40], and this was subsequently shown to be dependent on the NLRP3 inflammasome [41]. In addition, IL-1b expression was shown to be important for recruitment of DCs to the lymph node after WNV infection [8]. Thus, inflammasome signaling may integrate with multiple immune pathways to participate in the control of Flavivirus infection.
In this study, we systemically examined the role of IL-1 signaling in WNV infection to show that IL-1b signaling driven by the NLRP3 inflammasome acts to mediate protective immunity against infection. We reveal a novel role for IL-1b in specifically limiting viral replication within the CNS. Mechanistically, IL-1b synergizes with type I IFN to control virus replication. Moreover, we found that the lack of viral control in Il-1r 2/2 mice correlated with defects in CNS inflammation, T lymphocyte effector activity, and neuropathology. Our observations link the production of IL-1b in the CNS to restriction of WNV replication in neurons and limitation of CNS disease and thus we conclude that the NLRP3 inflammasome and IL-1 signaling are important determinants of immune regulation that impart protective CNS inflammation and control of WNV infection.
IL-1b production associates with human WNV infection
In order to assess whether IL-1 was associated with human WNV infection, we examined cytokine expression in the plasma from individuals infected with WNV. Blood donors, testing positive for WNV RNA after routine blood screening were enrolled in a follow-up study to assess their plasma cytokine levels over a six-month period after their initial blood donation (index). We observed that the levels of IFN-c and TNF-a, immune factors important in immunity to WNV, [13,18] were enhanced in individuals infected with WNV and the expression of these factors was maintained for long periods of time in these individuals ( Figure S1A,B).
In addition to these known host restriction factors, the levels of IL-1b were enhanced in the plasma of WNV + individuals and displayed significant increase over time (trend analysis) at 7, 21, 42, and 180 days post-index when compared to normal controls (Cntrl) ( Figure 1A; trend analysis (p,0.0001), inset). This was similar to the expression of the IL-1r antagonist (IL-1ra), a natural regulator of IL-1b [25], which is known to track with expression of the cytokine ( Figure 1B; trend analysis (p = 0.003), inset). In contrast, levels of IL-1a in the plasma were not altered by WNV infection at any time-point tested ( Figure S1C). These data are consistent with the expression of IFN-c and TNF-a, and is in line with recent data demonstrating that WNV RNA persists at low levels out to 200 days post-index in whole blood of infected humans [42] and in vivo in peripheral tissues of mice [43] for extended periods of time.
Author Summary
Since its introduction into North America in 1999, West Nile virus (WNV) has emerged as a leading cause of viral encephalitic disease in the United States. While low level inflammation is important for clearance of WNV, high levels of inflammation are associated with increased disease. The goal of this study was to identify host signaling pathways that control the balance of inflammation and protective immunity to WNV. Using a mouse model of infection, we identified a central nervous system (CNS)-intrinsic requirement for the NLRP3 inflammasome and IL-1b signaling in limiting WNV associated disease within the CNS. First, IL-1b signaling was essential for regulating the magnitude and kinetics of inflammation within CNS. Secondly, the absence of IL-1b signaling disrupted the quality of the effector T lymphocyte response against the virus. Finally, these dysregulated immune responses were linked to a direct ability for IL-1b signaling to synergize with type I IFN signaling and limit virus replication within cortical neurons, key target cells of WNV infection within the CNS. Together this study identifies the NLRP3 inflammasome and IL-1b signaling as key restriction factors that act to regulate viral load and the quality of inflammatory responses within the CNS to impart protective immunity against WNV infection.
We found that WNV RNA levels decreased from day 7 to day 21 post-index ( Figure 1C). This correlated with the induction of IL-1b at day 7, suggesting that initial viral load might drive the expression of the cytokine. In contrast, an inverse correlation was found between the levels of WNV viral load and the levels of IL-1b in plasma during the first six weeks post-index (GEE estimate = 21.28, p = 0.01). Together, these data demonstrate that IL-1b expression but not IL-1a expression is associated with WNV infection in humans and indicates that WNV infection triggers the inflammasome signaling pathway to induce IL-1b during infection.
Inflammasome and IL-1b signaling are important for immunity to West Nile virus in vivo To evaluate the role of IL-1b signaling in immunity against WNV infection, we assessed infection in a well characterized murine model of WNV subcutaneous (s.c) footpad inoculation [5,9]. C57BL/6-WT (WT) or mice deficient in their ability to respond to IL-1a or IL-1b, IL-1Ra chain 2/2 (Il-1r 2/2 ), were challenged with 10 2 plaque-forming units (PFU) of a virulent strain of WNV, WNV-TX [44]. Il-1r 2/2 mice were found to be highly susceptible to WNV infection ( Figure 2A) displaying increased mortality (73.3% mortality compared to 23% in WT mice). WT mice presented with clinical signs of disease, including mild-paresis and weight loss, by day 6 post-infection (p.i.) with WNV. These responses peaked by day 10 p.i. and the majority of mice were fully recovered by day 16 p.i. (Figures 2B,C). While Il-1r 2/2 mice displayed identical onset of clinical disease and weight loss as compared to WT mice, they presented with enhanced disease/ hind limb dysfunction (clinical scoring), weight loss and eventual death by day 10 p.i. (Figures 2A,B,C), thus demonstrating an important role for this signaling pathway in preventing late stage WNV-mediated disease.
IL-1b, but not IL-1a, requires processing via the NLRP3inflammasome for its functional secretion and activity against multiple viruses [29]. Therefore we next examined whether NLRP3 signaling was responsible for the observed phenotype in Il-1r 2/2 animals infected with WNV. We found that mice lacking NLRP3 (58.3% mortality) or its downstream effector, Caspase-1 (50% mortality), displayed increased susceptibility and clinical disease in response to WNV challenge ( Figure 2D, S2A-D). This outcome was in contrast to infection of mice lacking NLRC4, which mediates a distinct inflammasome not associated with IL-1b processing in response to viral infection [24]. In these mice, the absence of NLRC4 did not influence susceptibility ( Figure 2E) or clinical disease (Figures S2E,F) to WNV infection compared with WT controls. Thus, these observations are consistent with WNV infection triggering ''signal 2'' processing of IL-1b in vivo through the NLRP3 inflammasome to mediate immunity. In addition to inflammasome processing of IL-1b, we also examined the requirement for Myd88, the adaptor protein that propagates both IL-1 and TLR signaling [26]. Similar to our observations in Il-1r 2/2 and NLRP3-inflammasome deficient animals, Myd88 2/2 animals were highly susceptible to WNV challenge (100% mortality) ( Figure 2F). These results correlate with recent observations in mice deficient in MyD88, which display enhanced mortality to WNV due to lack of CNS-specific control of the virus [19] and likely reflects the essential role of MyD88 in signaling the response to IL-1b [25][26]. Together, these data are consistent with a model in which IL-1b activation via the NLRP3 inflammasome and signaling via MyD88 act as the major pathways involved in mediating the Il-1r 2/2 phenotype to control late stage WNV disease.
IL-1b responses control West Nile virus replication within the CNS
In order to more fully understand the contribution of IL-1 signaling in the in vivo response to WNV, we examined IL-1a and IL-1b expression in tissues known to be active sites of viral replication in WT mice [5,9]. WNV infection induced variable expression of both IL-1a and IL-1b in the draining popliteal lymph node (DLN) and spleen of infected mice ( Figures S3A,B) however neither cytokine were detected in the serum of infected WT animals at any time-point tested (data not shown). In contrast to these peripheral tissues, we found the most dramatic differences in IL-1a and IL-1b in the brains of infected animals. Here, IL-1b reached detectable levels by day 6 p.i. and was maintained at high levels through day 9 p.i. when the cytokine level peaked ( Figure 3D). This was in contrast to IL-1a which was expressed only at low levels throughout the course of infection ( Figure 3D) suggesting that IL-1b may play the predominant role in immunity against WNV. We observed no difference in the magnitude of IL-1a (data not shown) or IL-1b ( Figure S3C) levels in the CNS between WT and Il-1r 2/2 animals. We next assessed whether the tissue-specific expression of IL-1b influenced WNV replication within peripheral organs, blood, and the CNS. We observed no difference in the kinetics or magnitude of viral load detected between WT and Il-1r 2/2 deficient animals within the spleen or serum (Figures 3B,C), and this was consistent with the lack of appreciable IL-1b at the peak of viral replication in these tissues (Figures S2B, data not shown). In contrast, we observed a modest trend of increased viral load at day 2 post infection in the DLN ( Figure 3A), suggesting a role for IL-1b signaling in immunity to WNV at early time-points. Despite only minimal differences in viral load in peripheral organs, viral load in the brains of Il-1r 2/2 mice was increased when compared to WT mice as early as day 8 p.i. and this persisted through day 10 p.i. ( Figure 3E). A similar enhancement of viral load was observed in the brains of MyD88 and NLRP3-deficient animals ( Figures S4A,B), as well as, in day 10 spinal cords of all IL-1 signaling-deficient animals ( Figures S4C). The difference in viral control were not due to earlier entry of virus in the CNS as virus was first detected in the CNS at day 6 p.i. with similar magnitude of viral load in both strains ( Figure 3E). Instead, the initial differences in viral control occurred between day 7 and 8 p.i. and correlated to a time-point in which IL-1b was detected at high levels in WT mice ( Figure 3D,E). Together, these data suggest NLRP3 inflammasome activation of IL-1b and the subsequent actions of IL-1b operate to limit the replication and/or spread of WNV within the CNS.
Type I interferon (IFN) signaling has been shown to contribute to control of tissue tropism and CNS control of WNV [9,[11][12]. Therefore, we tested whether type I IFN could be involved in the lack of viral control in the DLN and CNS of inflammasome-deficient animals. We did not observe a significant difference in secretion of IFN-b in the serum or spleen of infected Il-1r 2/2 or WT animals (data not shown). In contrast, we did observe a significant reduction in IFN-b expression in the DLN at day 2 p.i. in Il-1r 2/2 mice as compared to their WT counterparts ( Figure S3D). This difference correlated to a trend of increased viral load ( Figure 3A) but did not appreciably impact overall peripheral virus replication ( Figures 3B,C). In contrast to peripheral tissues, we observed differential expression of IFN-b within the CNS between WT and Il-1r 2/2 animals. WT mice displayed early IFN-b expression within the CNS with peak levels at day 7 p.i. ( Figure 3F). This was consistent with the peak of viral load in these mice suggesting a direct correlation between virus replication and type I IFN production. In contrast, Il-1r 2/2 animals showed a delay in expression of IFN-b, with levels tracking with maximal viral load and peaking at day 9 p.i. ( Figure 3F). Importantly, the magnitude of the peak of IFN-b expression was not significantly altered between WT and Il-1r 2/2 animals despite significantly higher viral loads in the absence of IL-1b signal ( Figures 3E,F). These observations suggest that sustained IL-1b signaling in the CNS is important in maintaining efficient activity and/or early type I IFN signaling that controls WNV infection.
IL-1b signaling regulates the CNS inflammatory response during West Nile virus infection
IL-1b signaling has been associated with various immune functions including regulation of cell recruitment and inflammation, modulation of adaptive immunity and direct anti-viral activity. Therefore we assessed these activities as possible We first examined CNS infiltration of total CD45 + leukocytes as well as a subpopulation of CD45 + CD11b + cells (a population which is comprised of macrophage dendritic cell and neutrophil subsets) as these cell populations have been previously associated with immunity against WNV in the CNS [18][19][20]. Brains of WT or Il-1r 2/2 mice were harvested at day 6, 7, 8 and 10 p.i. with WNV and the kinetics of CD45 + leukocyte infiltration were assessed by flow cytometry.
We observed minimal infiltration of CD45 + leukocytes in the brains of WT or Il-1r 2/2 mice at day 6 p.i.(Figures 4C). However, by day 7 p.i., the numbers of total cellular infiltrates, ( Figure 4B), as well as the frequency and total numbers of CD45 + leukocytes ( Figures S5A, 4C) and CD11b + /CD45 + infiltrates (Figures S5B, 4A,D) was increased in WT animals. This was followed by a peak in accumulation of cell infiltrates at day 8 p.i. and subsequent reduction in cell numbers by day 10 p.i. (Figures S5A,B, 4A-C) which tracked with parallel kinetics of viral load in the CNS of WT mice ( Figure 3E). Conversely, Il-1r 2/2 mice displayed reduced numbers total cell infiltrates ( Figure 4B) as well as frequency and total numbers of CD45 + leukocytes ( Figures S5A, 4C) and CD11b + /CD45 + ( Figures S5B, 4A,D) at day 7 p.i in the CNS. Furthermore, despite reaching levels of infiltrates comparable to WT mice at day 8 p.i., Il-1r 2/2 animals showed a continued increase in immune cell infiltrates through day 10 p.i., suggesting that the magnitude of inflammation was directly correlated to viral load ( Figure 3E) in the CNS in these animals.
Resting microglia express CD11b but are distinguished from activated microglia or infiltrating myeloid cells by lower levels of CD45 expression. We observed the numbers of resting microglia, (CD11b + /CD45 lo ), were reduced in the CNS at early times (day 6-7p.i.) in WT and Il-1r 2/2 animals (Figures 4E, S5C) suggesting increased activation of a subset of microglia upon viral entry into the CNS in both strains of mice. However, while the numbers of non-activated microglial cells was restored to basal levels by day 10 p.i. in WT mice they remained predominantly active day 10 p.i. in Il-1r 2/2 animals further suggesting that lack of IL-1b signaling imposes a disruption in CNS inflammatory homeostasis during WNV infection ( Figures 4A,E, S5C).
Pro-inflammatory cytokines (TNF-a, IL-6), and chemokines (CCL5, CCL2) which are involved in the recruitment of CD45 + leukocytes, were also enhanced in the CNS of Il-1r 2/2 animals at late time-points when compared to WT controls (Figures 4F-I) implying that the increase in these inflammatory mediators likely contributes to the increased inflammation within the CNS of Il-1r 2/2 mice. In contrast, IL-6 ( Figure 4F) and CCL2 ( Figure 4I) were expressed at comparably lower levels at day 7 p.i. in the CNS of Il-1r 2/2 animals and associated with reduced infiltration of peripheral inflammatory leukocytes. Interestingly we observed increased inflammatory responses in the CNS of NLRP3-deficient animals, which displayed high CNS viral load and enhancement of IL-6, TNF-a and CCL5 levels after WNV infection (Figures S5D-F). Together, these observations indicate that the NLRP3inflammasome and IL-1b signaling act early during WNV Figure 3. IL-1b signaling is required for CNS-specific protective immunity against WNV. Examination of in vivo IL-1 cytokine expression and viral load in WT and Il-1r 2/2 animals. 6-10 wk old WT mice were infected s.c. with 100 PFU WNV-TX02 or mock infected. Viral loads were assessed from WT (closed circles) or Il-1r 2/2 mice (open squares) by plaque assay from spleen (B) and brain (E) lysates or by Taqman based qRT-PCR using specific primers and probes to WNV for DLN (A) and Serum (C). Cytokine expression for IL-1a and IL-1b was assessed by Luminex array for Brain lysates (D) or IFN-b levels by ELISA (F) Data are shown as the mean +/2 S. E. M. for n = 3-6 mice per time-point. *p,0.05, *** p,0.0005. Dashed lines represent the lower limit of detection for each assay. BLD denotes below limit of detection. doi: 10
Loss of IL-1b signaling leads to increased neuropathology in the CNS of West Nile virus infected animals
Regulated inflammation in the brain is an important component of the immune clearance of neurotropic viral pathogens [17]. This process is consistent with the regulated CNS inflammation observed in WNV-infected WT mice (peak inflammatory CNS response at day 8 which was reduced by day 10 p.i. (Figure 4). Thus, in the absence of IL-1b signaling, the enhanced inflammation observed in Il-1r 2/2 mice might actually be detrimental to the host. Indeed, hematoxylin and eosin (H&E) stained histological sections of brain tissue from Il-1r 2/2 animals revealed encephalitis at day 10 p.i. marked by perivascular cuffing and inflamed meninges ( Figure 5A; black arrows, compare top and bottom panels). Tissue damage was associated with multiple regions of the CNS and included edema and hemorrhage ( Figure 5A; Forebrain), mononuclear meningitis ( Figure 5A; meninges) and neuronal dropout (Figure 5A-Midbrain; white arrows) We also observed enhanced macrophage staining (MAC-2) throughout the CNS tissue of WNV-infected, Il-1r 2/2 , mice at day 10 p.i. (Figure 5B). Increased accumulation of MAC-2 + cells was observed in the forebrain and hippocampus-thalamus but was most evident in the midbrain, where we observed a 22-fold increase in MAC-2 expression in Il-1r 2/2 compared to WT controls ( Figure 5B,C). This expression was most prominent in and around blood vessels ( Figure 5B; black arrows), and within the parenchyma of Il-1r 2/2 but not WT brain tissue sections ( Figure 5B, bottom panel). This pattern of staining was comparable to the regions of immune infiltrates identified by H&E staining ( Figure 5A) and suggests that increased infiltration of macrophage subsets is associated with the tissue damage observed in the CNS of Il-1r2/2 animals. Thus we conclude IL-1bdependent control of WNV in the CNS regulates the magnitude of inflammation during acute infection to reduce tissue damage.
IL-1b signaling is required for proper T lymphocyte effector activity in the CNS of West Nile virus infected mice
CD8 + T cell responses are important for clearing WNV infection from the CNS [14][15][16]. We therefore assessed: 1) total Ag-specific cell accumulation, 2) cytokine secretion properties and 3) cytolytic properties of CD8 + T cells within the CNS of WT and Il-1r 2/2 mice. Similar to infiltrating leukocytes, the initial entry of total CD8 + ( Figures animals when compared to WT mice. However, by day 10 p.i. total CD8 + ( Figure 6A), CD4 + (data not shown) and antigenspecific CD8 + T cells ( Figure 6B), were enhanced in Il-1r 2/2 animals, further demonstrating a breakdown in the regulation of inflammation. Interestingly, although the total number of antigenspecific cells was increased in the absence of IL-1b signaling, the frequency of these cells within the CD8 + T cell pool was reduced at each time-point tested compared to wild-type mice ( Figure 6C), thus suggesting that effector activity in the CNS might be altered in these mice.
In order to test this, total cells were recovered from the CNS at day 8 and day 10 p.i. and stimulated ex vivo with the immunodominant T cell epitope peptide-WNV-NS4b (NS4b). We did not observe any differences between Il-1r 2/2 and WT cells in their frequency of cytokine production at day 8 p.i. (data not shown). However, by day 10 p.i., the frequency of TNF-a + /IFNc + double positive and TNF-a + , single cytokine producing cells were significantly lower in Il-1r 2/2 cells when compared to WT infected mice (Figures 6D,E). In contrast, the frequency of IFN-c + single cytokine expressing CD8 + T cells were comparable between the two genotypes, demonstrating an overall loss in multifunctionality but not complete dysfunction in Il-1r 2/2 CD8 + T cells (Figures, D,E). In addition, restimulation of equal numbers of WT or Il-1r 2/2 total brain cells (2.5e 5 ) with the NS4b peptide promoted significantly less IFN-c production at both time-points tested ( Figure 6F) consistent with the reduced frequency of antigen-specific cells at these time-points ( Figure 6C). Cytolytic activity was also found to be compromised in Il-1r 2/2 T cells, as defined by reduced frequency of cells expressing perforin and this was accompanied by a significant increase in granzyme B/perforin Figure 6H,I). Taken together these data serve to link dysregulated inflammation in the CNS of Il-1r 2/2 mice with defective antigen-specific CD8 + T cell effector cytokine and cytolytic activity in the CNS.
Despite a requirement for IL-1b signaling in regulating the T lymphocyte response, humoral immunity remained relatively intact as we observed no differences in the levels of WNV-specific serum IgM and IgG or neutralizing activity of these antibodies between WT and Il-1r 2/2 deficient animals (Table S1). Thus, IL-1b signaling is important for maintaining proper T but not B lymphocyte effector activity during WNV infection.
IL-1b signaling acts in an antiviral manner in vivo to control West Nile virus within the CNS
To determine whether IL-1b signaling acted directly within the CNS to control WNV replication, mice were challenged with a low dose of WNV-TX (5 PFU), via intracranial (i.c.) inoculation and assessed for viral load and cytokine responses within the CNS. Similar to our observations in peripheral challenge, Il-1r 2/2 mice showed a significant enhancement of viral load in the brain ( Figure 7A) as well as spinal cord (data not shown) when compared to WT, i.c. infected, animals, suggesting that IL-1b signaling imparts CNS-intrinsic control of WNV replication. In agreement with this notion, brain tissue sections stained for WNV-antigen showed more widespread and increased intensity staining in Il-1r 2/2 when compared to those from WT animals ( Figure 7D,E). Staining of viral antigen was associated primarily in cortical, hippocampal and mid-brain neurons with sparse staining in the endothelium ( Figure 7D; black arrows indicate staining in endothelial cells) demonstrating a distinct tropism for neuronal cells. Similar to peripherally challenged mice ( Figure 3F) the expression of IFN-b was enhanced at day 4 p.i in Il-1r 2/2 animals ( Figure 7B). Furthermore, in WT mice, i.c. WNV infection triggered IL-1b production in the brain by day 2 p.i. and this was maintained through day 4 p.i. (Figure 7C). Thus, these data indicate that CNS innate immune signaling and IL-1b production are triggered directly by WNV infection and reveal that IL-1b signaling is an essential component of CNS-intrinsic virus restriction.
IL-1b signaling controls West Nile virus infection in cortical neurons
To determine how IL-1b signaling mediates CNS-intrinsic control of WNV, we examined WNV infection ex vivo in cortical neurons, a primary target cell for WNV replication within the CNS. While a direct antiviral role for IL-1b signaling in neurons has not yet been defined, a previous study demonstrated that neurons from Myd88 2/2 animals presented with increased viral load after WNV infection when compared to their WT counterparts [19]. Thus, we reasoned that Myd88-dependent IL-1b signaling might promote an antiviral state in these cells. Cortical neurons were prepared from WT and Il-1r 2/2 mice as previously described [45] and were purified to greater than 95% based on staining for the neuronal marker, NeuN (data not shown). Cultures were infected with WNV (MOI = 1) and singlestep growth curve analysis and their host response to WNV infection was examined.
Il-1r 2/2 cells showed significantly higher viral loads at 24 hr and 48 hr p.i. when compared to their WT controls ( Figure 8A). This increase was also observed in multi-step growth curve analysis after infection with a lower MOI (0.01), although this was not maintained through 48 hrs ( Figure S6A). We found that WT neurons produced IL-1b as early as 12 hr p.i. after virus challenge and this persisted throughout the course of infection ( Figure 8B), thus suggesting that neurons might contribute in part to the total IL-1b response in the CNS. The production of IL-1b was detectable in Il-1r 2/2 cells but at generally lower levels than WT ( Figure 8B). Remarkably, we observed enhanced IFN-b production in Il-1r 2/2 neurons at both MOIs ( Figures 8C, S6B), however this only effective at reducing viral load after low MOI infection ( Figure 8A, S6A). These observations demonstrate that IL-1b signaling contributes to the control of WNV infection in cortical neurons and may operate to enhance the antiviral response mediated by type I IFN signaling.
IL-1b signaling synergizes with type I IFN to control WNV infection in cortical neurons
To determine whether IL-1b could impart suppression of WNV replication in cortical neurons, cells were pre-treated for 24 hrs with either media alone, or IL-1b (10 ng/ml), infected with WNV (MOI = 1) and then assessed for viral load in the supernatant. IL-1b treatment of cells resulted in a reduction of detectable WNV at 24 hr (2.5 fold) and 48 hr (1.7 fold) p.i. compared to non-treated cells ( Figure 8D). This level of reduction was similar to the fold increase in viral titers observed in Il-1r 2/2 neurons ( Figure 8A; 24 hr, 2.3 fold, 48 hr, 2.1 fold) thus demonstrating that IL-1b imparts a response that contributes to the control of WNV replication. In WNV-infected Il-1r 2/2 mice we consistently observed increased levels of type I IFN, either in whole tissue ( Figures 3F, 7B) or from cortical neurons (Figure 8C, S6B) but this level was insufficient to control the viral load. Thus the actions of IL-1b might function to enhance the antiviral effect of type I IFN against WNV infection.
To test this idea, we conducted experiments in which cortical neurons were either left untreated or pretreated for 24 hrs with IL-1b (10 ng/ml), IFN-b (100 IU/ml) or both cytokines together, followed by WNV infection and assessment of viral load. Similar to our previous results, IL-1b treatment led to a 2.1-fold reduction in virus at 24 hr p.i while treatment of neurons with IFN-b reduced viral load by 13.5-fold at 24 hr p.i., consistent with previous studies [11][12] (Figure 8E). Remarkably, when neurons were pretreated with both IL-1b and IFN-b, we observed near complete control of WNV at 24 hr (1500-fold reduction of viral load compared to control; Figure 8E). Further, these results were prolonged as the fold reductions by each cytokine were maintained through 48 hrs p.i. (Figure S6C). We also observed that pretreatment of neurons with both IL-1b and IFN-b in the context of WNV infection led to an increase in mRNA ( Figure S6D) and protein expression ( Figure 8F) of IFN-b and interferon stimulated genes (ISGs), STAT1, IFIT1, IFIT2 and IFIT3, molecules known to participate in control of WNV [46][47]. This response was comparable to induction by type I IFN in the absence of infection ( Figure 8F), suggesting that despite the known ability of WNV to antagonize these responses [44,48], the synergy of IL-1b and type I IFN might act to overcome this antagonism and promote increased viral control. The induction of ISGs was not observed with IL-1b treatment alone ( Figure 8F), demonstrating that combined signals induced by viral infection, type I IFN and IL-1b were required for the synergistic activation of these antiviral genes. Therefore we conclude that IL-1b synergizes with type I IFN to enhance antiviral gene programs that control WNV infection in cortical neurons and thus, IL-1b acts as a key host restriction factor in the control of WNV infection.
Discussion
Our observations support a model in which IL-1b signaling functions as a host restriction factor to control WNV replication within the CNS. WNV restriction by IL-1b occurs in a manner dependent on the ability of IL-1b to synergize with type I IFN to promote a robust antiviral program in neurons. Further, the capability for IL-1b signaling to control viral load is essential for regulating protective CNS inflammation to control WNV disease, as a lack of IL-1b signaling associates with a breakdown of immunity marked by rapid and uncontrolled viral spread through the CNS, hyper-active inflammatory response and defective CD8 + T lymphocyte effector activity. Thus, IL-1b is fundamental for the control of WNV infection and immunity.
To date, three distinct inflammasomes have been described to participate in IL-1b activation in response to viral infection. These include the NLRP3 [32,34,36,49], RIG-I [50] and AIM2 [51][52][53] inflammasomes. While AIM2 is responsible for activation to DNA viruses, both NLRP3 and RIG-I have been associated with IL-1b activation in response to RNA viruses (reviewed in [24]). We found that the defects in immunity observed in Il-1r 2/2 mice infected with WNV were pheno-copied in mice deficient in NLRP3 or Caspase-1 suggesting that the NLRP3-inflammasome acts as the predominant pathway for triggering IL-1b production in vivo during WNV infection. This is not surprising as NLRP3 activation of IL-1b has been shown to occur in response to multiple RNA and DNA viruses including influenza, Sendai and adenoviruses [24] as well as JEV, a WNV related Flavivirus [41]. Further, our observations that NLRP3 signaling is important for limiting lethality and tissue destruction during WNV infection in vivo is consistent with models of influenza infection which have also shown a requirement for NLRP3 and IL-1b signaling in protective immunity against the virus [32,34,36] as well as in limiting collagen deposition and necrosis of lung tissue [34] and thus demonstrate a broad range for NLRP3 signaling in immunity to virus infection. While NLRP3 and Caspase-1 deficient animals succumbed to WNV infection with similar kinetics and frequency we did observe a trend for increased virulence when compared to Il-1r 2/2 animals (Figure 2A,D). While this maps the majority of IL-1 signaling to NLRP3 activation, it is possible that other pathways might contribute to this response. As RIG-I has recently been shown to mediate both the activation of ''signal 1'' as well as ''signal 2'' in response to vesicular stomatitis virus (VSV) in an NLRP3-independent manner [50] and RIG-I [9,44,54] plays important roles in sensing and triggering of innate immune pathways against WNV virus, it is possible that this pathway may contribute in addition to NLRP3 in the full activation of IL-1b activation during WNV infection in vivo.
The contribution of IL-1 signaling to protective immunity against virus infections has largely been attributed to its ability to drive chemokine signaling pathways that recruit immune cells to sites of viral replication [38,39,40]. Although CNS recruitment of CD45 + monocytes and T lymphocytes was largely increased during WNV infection, in the absence of IL-1 signaling, we did observe a reduced frequency and total number of these cells at day 7 post infection ( Figure 4) and this is consistent with previous studies that showed that IL-1b was important for optimal cellular recruitment to the DLN after WNV infection [8]. Therefore, it is possible that this reduction in initial leukocyte and lymphocyte recruitment is sufficient to allow for enough virus to seed infection in neurons and increase the likelihood of CNS spread. Interestingly, despite early differences in cellular recruitment and inflammation in the CNS, by later time-points, we observed a dramatic enhancement in inflammation in Il-1r 2/2 animals (Figure 4). This is in contrast to recent studies examining WNV infection in Myd88 [19] and TLR7 [20] deficient animals which show dramatic reduction in cellular infiltrates and inflammation within the CNS. Thus, it is likely, that while IL-1b signaling contributes to cellular recruitment, TLR signaling via MyD88 might play a dominant role in maintaining these responses over time.
As CD8 + T cells represent a key component in late CNS control of WNV, it is interesting that IL-1 signaling defects were associated with decreased effector activity of these cells. Our data suggest that increased viral load influence this defect directly by driving increased inflammatory cytokine responses and a reduced frequency of antigen-specific cells to total CD8 + cells. However, in addition, it is also possible that IL-1b signaling itself is required to directly promote the optimal T cell effector response. This outcome is similar to observations of influenza infection in which IL-1R and NLRP3-inflammasome deficient animals display defective CD4 + and CD8 + T lymphocyte responses in the absence of increased inflammation [36]. Furthermore, MyD88 signaling was found to act in a T cell intrinsic manner to control both proliferation and acquisition of effector responses during LCMV [55][56] and vaccinia virus infection [57][58], while direct IL-1 signaling in has been shown to influence the development of multiple T lymphocyte effector populations under multiple experimental conditions [23]. Our data are in agreement with a role for IL-1b signaling in driving T cell effector activity during viral infection. Such action might occur against WNV infection by either of two mechanisms: First, IL-1b signaling could indirectly limit inflammation and overstimulation of cells in the CNS that otherwise occurs under conditions of uncontrolled virus replication and dissemination [9,[19][20]. Secondly, IL-1b signaling could directly drive a response, likely in a Myd88-dependent manner, to positively regulate immune cell effector activity against WNV.
We observed a significant reduction in type I IFN in the DLN at day 2 p.i. in Il-1r 2/2 mice that correlated with a modest increase in viral load ( Figures S3D, 3A). This result is notable because the observation occurred at a time-point that is associated with the accumulation of macrophages and DC subsets trafficking into the DLN from the site of primary infection, wherein type I IFN signaling is associated with control of the virus in these cell types [9]. We have also observed reduced type I IFN responses and increased viral replication in primary Il-1r 2/2 macrophages and DCs infected with WNV in vitro (Ramos, HJ; unpublished observations). Furthermore, it has been proposed that viral entry to the CNS could in part be mediated by a ''Trojan horse'' mechanism in which immune cells harboring WNV infiltrate into the CNS and seed viral infection in neurons [5]. Together, these observations raise the possibility that the lack of type I IFN in the DLN in Il-1r 2/2 mice is associated with decreased control of virus within macrophages and DCs, which then infiltrate and seed the CNS with virus in an enhanced manner that promotes the robust inflammatory disease observed in Il-1r 2/2 mice. In this sense IL-1b would serve to control virus levels in infected inflammatory cells and thereby restrict virus seeding into the CNS that might occur through infiltration of infected cells.
Type I IFN and to a lesser extent, IFN-c, and TNF-a, have also been shown to mediate anti-viral responses to WNV [13,18] and our results now implicate IL-1b as a component of innate antiviral signaling and response against WNV infection. In support of this, a protective role for IL-1b in combination with TNF-a was shown to be important to the control of hepatitis B virus [21][22]. In addition, in vitro studies in hepatocyte cell lines revealed that this effect was mediated by the ability of IL-1b to augment type I and type II IFNs antiviral activities in mechanisms dependent on STAT-1 and P38 MAP kinase [21]. This phenomenon of IL-1b synergy has also been observed in epithelial cell infection with VSV [37]. Therefore, we propose that in the context of WNV infection, IL-1 acts synergistically with type I IFN, and possibly type II IFN and/or TNF-a to mediate an antiviral program to control WNV in the CNS.
We found that IL-1b treatment of WT cortical neurons ex vivo resulted in reduced levels of detectable virus in the supernatants of infected neurons however, this response was not sufficient by itself to fully inhibit the virus. Interestingly, the viral suppression we observed occurred between 24 and 48 hrs of treatment suggesting that IL-1b signals a response in neurons that restricts virus amplification rather than initial infection. This kinetics are suggestive of a mechanism by which products of IL-1b -responsive genes serve to impart control of WNV replication, thus suppressing virus spread within the CNS. Indeed, our data demonstrate that IL-1b acts synergistically in the innate antiviral response along with IFNs and possibly other cytokines to ultimately control virus replication. In line with this notion, we observed that IL-1b and type I IFN synergized to enhance the expression of ISGs such as the IFIT family members IFIT1,2,3. This is of interest as recent studies have implicated these antiviral effectors in the control of CNS specific viruses such as VSV [59]. Furthermore, recent data had identified IFITs as important in anti-viral control of WNV [46][47] and this is dependent on their ability to block the viral replication cycle [46]. Therefore, IL-1b might control WNV through its ability to modulate ISGs that impart antiviral actions against WNV replication.
Control of WNV replication in the CNS is paramount for protection against disease. Our study shows that IL-1b production and signaling are important for protective immunity against WNV suggesting that the NLRP3-inflammasome and IL-1 signaling also likely impact immunity to other Flaviviruses. It has been observed that dengue, yellow fever (YFV), St. Louis encephalitis (SLEV) and WNV all trigger IL-1b expression from populations of myeloid derived cells in vitro, while YFV, SLEV and WNV have each been linked to suppression of IL-1b signaling at various levels in vitro [60][61][62]. Taken together, these observations along with those made in this study identify IL-1b as a key host restriction factor in immunity against Flavivirus infection.
Ethics statement
All animal experiments were approved by the University of Washington Institutional Animal Care and Use Committee (IACUC) committee guidelines (protocol number: 4158-01) and follow the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Mouse infections and manipulations were performed under anesthesia of ketamine and xylazine, and every effort was made to limit suffering.
All human subjects provided written-informed consent under a University of California, San Francisco Institutional Review Board approved protocol.
Human subjects
The 43 WNV + subjects included in this study were enrolled in 2009 and 2010 by Blood Systems Research Institute (BSRI) through the United Blood Services network of blood centers. Blood donors who tested positive for WNV RNA by routine donation screening using the WNV Procleix Transcription Mediated Amplification (TMA) assay were asked to return to their local blood donation center for enrollment after informed consent was completed under a University of California, San Francisco Institutional Review Board approved protocol. Upon enrollment, blood donors agreed to return to their blood center for subsequent follow up visits. Infection was confirmed using followup samples showing sero-conversion to anti-WNV IgM. Samples were collected at regional blood centers and were shipped by overnight courier to BSRI. Blood was processed within 24 hours and plasma aliquots were frozen immediately for long term storage. The WNV + subjects were 58% male with an average age of 51 years old. The control subjects (Cntrl.) used in this study were 21 BSRI staff members who consented to donate blood for this study and they were 43% male with an average age of 48 years old.
WNV viral load quantification
Quantification of WNV viral load in follow-up human plasma specimens was assayed by real-time reverse transcription-polymerase chain reaction as previously described [42].
In vivo murine infections C57BL/6 (WT) and IL-1Ra deficient mice were purchased from Jackson Laboratories, Bar Harbor, ME. NLRP3 [63], Caspase-1 [64] and NLRC4 [65] were generously provided by Dr. Vishva Dixit (Genentech, San Francisco, CA), Dr. Chris Wilson (University of Washington, Seattle, WA) and Dr. Alan Aderem (Seattle Biomed, Seattle, WA). All mice were genotyped for positive identification and were bred in specific pathogen-free conditions in the animal facility at the University of Washington. Experiments were performed in accordance with the University of Washington Institutional Animal Care and Use Committee guidelines. Age-matched six to ten week old mice were inoculated subcutaneously (s.c.) in the rear footpad with 100 PFU of WNV-TX 02 in a 10 ml inoculum diluted in phosphate buffered saline (PBS) supplemented with 1% heat-inactivated FBS. Mice were monitored daily for morbidity and mortality. For clinical scoring, infected mice were monitored daily for signs of hind limb dysfunction and paresis. Mice were scored using the following scale from 1-6: 1, ruffled fur/lethargic, no paresis; 2, very mild to mild paresis; 3, frank paresis involving at least one hind limb and/ or conjunctivitis; 4, severe paresis; 5, true paresis; 6, moribund.
Viral tissue quantification
To determine in vivo viral burden, s.c infected mice were euthanized, and perfused with 20-30 ml of PBS to remove blood from tissues. Whole tissue were removed, weighed, and homogenized in 500 ml (spleen, brain) or 200 ml (spinal cord) of PBS containing 1% heat-inactivated FBS using a Precellys 24 at 1500 RPM for 20 seconds (Bertin Technologies, France). Sample homogenates were then titered by plaque assay on BHK cells as previously described [9]. For analysis of viral load within the draining lymph nodes (DLN), the popliteal DLN was harvested and homogenized as described above in 350 ml of RNA extraction buffer (RLT, Qiagen), and total RNA was extracted using an RNeasy kit (Qiagen). DNase treated RNA (Qiagen) was then reversed transcribed to cDNA using a 1:1 mixture of random hexamers and oligodT primers with the iScript select cDNA synthesis kit (Biorad). WNV-specific RNA copy number was measured by single-step Real Time-quantitative PCR (qRT-PCR) using Taqman technology via specific primer sets and probes as previously described [9]. For serum samples, viral RNA was isolated from 50 ml of serum from mock or WNV-TX infected samples using the QIamp Viral RNA isolation kit (Qiagen). Isolated viral RNA was then subjected to cDNA synthesis and qRT-PCR for assessment of WNV copy number as described above.
Primary cell isolation and infection. Cortical neurons were isolated from 15-day-old embryonic mice and cultured as described previously [45] in Neurobasal media (Invitrogen) supplemented with B27 supplement (Invitrogen), HEPES, Lglutamine and antibiotic-antimycotic solution and plated on Laminin coated plates. On day 4 of culture, neurons were infected with WNV-TX02 at an MOI of 0.01 or 1 and supernatants were collected for evaluating viral titers. Cells were collected for RNA analysis by RT-qPCR using specific primer sets. In addition, cell pellets were lysed in 30 ml radioimmunoprecipitation assay (RIPA) buffer for use in western blot analysis. For anti-viral assays, neurons were treated on day 4 post isolation for 24 hrs with either IL-1b alone, 10 ng/ml (Miltenyi biotec), IFN-b, 100 IU/ml (PBL interferon source), both cytokines or left untreated. After 24 hrs, supernatant was removed and cells were infected as described above.
Protein analysis
Whole brain tissue was isolated from mice perfused as described above. Tissue was lysed in 1 ml per brain RIPA buffer containing a cocktail of protease and phosphatase inhibitors (Sigma). Lysis was facilitated by homogenization using the Precellys (Bertin Technologies, France) as described above. Protein extracts (20 mg) were analyzed by immunoblotting. The following primary antibodies were used to probe blots: goat anti-WNV NS3 (R&D systems); rabbit anti-ISG54 (IFIT2) and rabbit anti-ISG49 (IFIT3), kindly provided by Dr. G. Sen; mouse anti-tubulin (Sigma) and rabbit anti-STAT1, (Cell Signaling). Secondary antibodies included peroxidase-conjugated goat anti-rabbit, donkey anti-goat and goat anti-mouse (Jackson Immunoresearch).
IFN-b ELISA. For detection of IFN-b in cell culture supernatants, 100 ml of UV-inactivated supernatant was tested using mouse-specific ELISA kits from PBL Biomedical Laboratories according to the manufacturer's protocol.
WNV-specific antibody ELISA and PRNT analysis WNV-specific IgM and IgG, levels were determined by an ELISA using purified recombinant E protein as previously described [66]. The neutralization titer of serum antibody was determined using the plaque reduction neutralization assay as previously described [67]. The dilution at which 50% of plaques were neutralized was determined by comparing the number of plaques formed from WNV-infected sera samples to mock infected sera samples.
Histological analysis
Mock-infected or WNV-infected mice were sacrificed by exsanguination and perfused with PBS-4% paraformaldehyde, pH 7.3. Brains were embedded in paraffin and 4-6 mm sections were prepared and stained with hematoxylin and eosin (H&E) or immunohistochemistry by the UW Histology and Imaging Core. H&E-stained sections were evaluated for viral-induced neuropathology. Antibodies for immunohistochemical detection included for for macrophages, rat anti mouse MAC-2, Clone M3/38, (Cedarlane, Cat No. CL8942AP); and for West Nile virus, rabbit anti-WNV, clone 7H2 + rabbit anti-WNV polyclonal (BioReliance, Cat No. 81-002, 81-015). Immunohistochemical staining was performed on a Leica Bond Automated Immunostainer with Leica Refine (DAB) detection and hematoxylin counterstain. Images were captured using a Nikon 80i Eclipse microscope with a digital camera with NIS Elements Basic research imaging software. Quantitation for MAC-2 and WNV was performed with the Visiopharm histoinformatics software (Visiopharm) on representative regions of two sets of brain sections and is representative of the ratio of specific staining to the total area of the tissue..
For quantification of WNV and MAC-2, slides were scanned in Brightfield at a 206 objective using the Nanozoomer Digital Pathology slide scanner (Hamamatsu; Bridgewater, New Jersey). The digital images were then imported into Visiopharm software (Hoersholm, Denmark) for analysis. Using the Visiomorph Digital Pathology module, regions of interest (ROIs) were applied around the relevant areas on each slide and consisted of one rectangular ROI within each of three areas-the forebrain, midbrain, and thalamus-on each section. The images were processed in batch using these configurations to generate the desired outputs (areas of staining and normal tissue, in square microns), from which the percent of WNV or MAC2 staining was calculated.
Tissue isolation and flow cytometric analysis
Splenocytes were isolated by grinding on frosted glass slides, washed counted, and re-suspended in RPMI 1640 containing 10% FBS before cell surface staining or in vitro stimulation. For staining, cells were washed once in PBS and once in PBS+0.5% BSA (FACS wash) in a 96 well plate format and then stained in 50 ml FACs wash plus directly-conjugated antibodies specific to CD8, CD4, and CD3 (Biolegend). WNV-specific CD8 + T cells were identified using a Db-restricted NS4b peptide tetramer directly conjugated to either Phycoerythrin (PE) or allophycocyanin (APC). For intracellular cytokine staining, cells were stimulated with 1 mM of the immune-dominant peptide NS4b (SSVWNATTA) for 4 hrs at 37uC in the presence of golgi-stop (BD-PharMingen). After stimulation, cells were washed twice in FACS wash and stained for cell surface markers followed by permeabilization-fixation using the Cytofix-Cytoperm Kit (BD-PharMingen) and stained with a Pacific Blue-conjugated IFN-c and FITC-conjugated TNF-a antibody (eBiosciences) or FITC-conjugated perforin and PE-CY7-conjugated granzyme B at 4uC for 30 min, washed and analyzed by flow cytometry. Flow cytometry was performed on a BD LSRII machine using BD FACSDiva software. Cell analysis was performed on FlowJo (v.8.7.2) software.
For isolation of CNS immune cells, euthanized mice were perfused with 20-30 ml of PBS to remove residual intravascular leukocytes. Brains were isolated and minced in RPMI media, rigorously triturated, and digested with Liberase (Roche) and type I DNase in serum-free RPMI media at 37uC for 1 hr. Immune cells were isolated after gradient centrifugation over 25% Percoll followed by a wash and secondary centrifugation over a 30% Percoll gradient. Cell pellets were isolated and washed twice and counted. Cells were resuspended in 50 ml FACS wash and stained for surface expression of CD4, CD8, CD33, CD11b and CD45 (Biolegend). For intracellular staining, cells were stimulated with 1 mM of NS4b peptide (SSVWNATTA) and stained as described above for IFN-c, TNF-a, and granzyme B and perforin. Cells were analyzed as described above for splenocytes.
ELISA and Luminex
Mice were euthanized and blood was collected for serum preparation via exsanguination. Serum was isolated by separating in microtainer (BD Biosciences) at 10,000 rpm for 8 minutes. Brain, spleen, and lymph nodes were isolated and weighed after perfusion with 20 ml of PBS and homogenized using the Precellys (Bertin Technologies, France) as described above in 1 ml of RIPA buffer. Protein concentration was assessed by Bradford colorimetric assay (Biorad) and 200 mg of total protein were loaded for IL-1b (Biolegend), IL-1a (Biolegend) and IFNb ELISA (PBL Biomedical Laboratories) via the manufactures protocol. Concentration of cytokine was then normalized to the weight of total tissue. For Luminex, 12.5 ml of tissue lysate or serum was run on an 11-plex assay (Miltenyi Biosciences) followed by analysis on a Bioplex 200 (Biorad). Concentration of cytokine was then normalized to total weight of tissue.
For human plasma cytokine analysis, plasma samples were assayed for IL-1b (0.06 pg/ml) and TNF-a (0.05 pg/ml) using the High Sensitivity Human Cytokine kit (Millipore, Billerica, MA, USA), which has a minimum detectable concentration as indicated above and for TNF-a, and for IFN-c (0.1 pg/ml), IFN-a2 (24.5 pg/ml) IL-1ra (2.9 pg/ml), andIL-1a (3.5 pg/ml) using the Human Cytokine/Chemokine kit (Millipore) which has minimum detectable concentrations as indicated above. Plasma samples were assayed following the manufacturer's protocols. Standard curves were run in duplicate, and samples were tested in duplicate. Acquisition was done on a Labscan 100 analyzer (Luminex) using Bio-Plex manager 6.1 software (Bio-Rad).
Statistical analysis
For in vivo viral burden analysis Kaplan-Meier survival curves were analyzed by the log-rank test. For all in vitro studies statistical analysis was performed via unpaired two-tailed student T-test. For in vivo viral burden and immune cell analysis experiments statistical significance was calculated using the Mann-Whitney test. A p-value#0.05 was considered significant.
All data were analyzed using Prism software (GraphPad Prism5).
For human patient studies, cytokine levels were compared between WNV + subjects andnormal controls at each time-point using one-way analysis of variance (ANOVA). Trend analyses were used to evaluate whether a statistically significant increase or decrease of cytokine levels was observed in the time post-index within the WNV + subjects; to detect monotonously increasing or decreasing trend of cytokine quantities over time, Page's trend tests were applied on each cytokine using the R Package 'concord'. A generalized linear model with repeated measures (Proc genmod (GEE) SAS 9.2) was used to analyze in plasma the correlation between WNV viral load and levels of cytokines/chemokines measured at one, three, and six weeks post-index donation. Statistical significance was defined as a p-value#0.05. Figure S3 Expression of IL-1 and IFN-b in tissues associated with WNV replication. Examination of in vivo cytokine expression and viral load in WT and Il-1r 2/2 animals. 6-10 wk old WT mice were infected s.c. with 100 PFU WNV-TX or mock infected and the kinetics of expression of IL-1b (closed circles) or IL-1a (open squares) (A,B) (WT only) or IL-1b (C) or IFN-b (D) in (WT or Il-1r 2/2 ) mice were assessed. Cytokine expression was assessed by quantitative real-time PCR (qRT-PCR) using specific primers for IL-1a and IL-1b (A) or IFN-b (D) and made relative to GAPDH and normalized to mock values in the draining lymph node (DLN) or Luminex array for spleen (B) and Brain (C). Data are shown as the mean +/2 S. E. M. for n = 3-6 mice per time-point. *p,0.05, *** p,0.0005. Dashed lines represent the lower limit of detection for each assay. BLD denotes below limit of detection. (TIF) Figure S4 Deficiency in IL-1b signaling is associated with decreased CNS control of WNV. Examination of viral loads in the CNS of WT or IL-1 signaling deficient animals. Mice were infected with 100 PFU WNV-TX and viral loads were assessed in the brain by plaque assay for WT (closed circles) and The antibody response to West Nile virus is not altered in IL-1R or inflammasome deficient animals. Mice were infected with WNV and serum was isolated at days 6 or 8 p.i. from WT and inflammasome deficient animals. Serum IgG and IgM were detected by ELISA for antibody specific to WNV-E protein. For PRNT assay, serum was used to neutralize purified WNV-TX02 in BHK infections. Data is presented as the dilution at which antibody was detected at three standard deviations above mock or for PRNT, the dilution required to neutralize virus by 50%. (DOCX)
==
Domain: Biology Medicine
|
[{"added": "2016-05-04T20:20:58.661Z", "created": "2012-11-01T00:00:00.000", "id": "15301167", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plospathogens/article/file?id=10.1371/journal.ppat.1003039&type=printable", "pdf_hash": "c88dcc62f9fb2d95fa8697a5ca42d594faeb5ba0", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:33766", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "c88dcc62f9fb2d95fa8697a5ca42d594faeb5ba0", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Chenodeoxycholic acid stimulates glucagon‐like peptide‐1 secretion in patients after Roux‐en‐Y gastric bypass
Abstract Postprandial secretion of glucagon‐like peptide‐1 (GLP‐1) is enhanced after Roux‐en‐Y gastric bypass (RYGB), but the precise molecular mechanisms explaining this remain poorly understood. Plasma concentrations of bile acids (BAs) increase after RYGB, and BAs may act as molecular enhancers of GLP‐1 secretion through activation of TGR5‐receptors. We aimed to evaluate GLP‐1 secretion after oral administration of the primary bile acid chenodeoxycholic acid (CDCA) and the secondary bile acid ursodeoxycholic acid (UDCA) (which are available for oral use) in RYGB‐operated participants. Eleven participants (BMI 29.1 ± 1.2, age 37.0 ± 3.2 years, time from RYGB 32.3 ± 1.1 months, weight loss after RYGB 37.0 ± 3.1 kg) were studied in a placebo‐controlled, crossover‐study. On three different days, participants ingested (1) placebo (water), (2) UDCA 750 mg, (3) CDCA 1250 mg (highest recommended doses). Oral intake of CDCA increased plasma concentrations of GLP‐1, C‐peptide, glucagon, peptide YY, neurotensin, total bile acids, and fibroblast growth factor 19 significantly compared with placebo (all P < 0.05 for peak and positive incremental area‐under‐the‐curve (piAUC)). All plasma hormone concentrations were unaffected by UDCA. Neither UDCA nor CDCA changed glucose, cholecystokinin or glucose‐dependent insulinotropic polypeptide (GIP) concentrations. In conclusion, our findings demonstrate that the primary bile acid chenodeoxycholic acid is able to enhance secretion of gut hormones when administered orally in RYGB‐operated patients—even in the absence of nutrients.
Introduction
Roux-en-Y gastric bypass (RYGB) is an effective treatment of severe obesity (Sj€ ostr€ om 2013; Madsbad et al. 2014) and type 2 diabetes (Madsbad et al. 2014;Schauer et al. 2014). Interestingly, improved glycemia occurs within days after surgery indicating that weight-loss cannot fully explain improvement in glucose tolerance (Pories et al. 1995). Early improvement in hepatic insulin sensitivity (Bojsen-Moller et al. 2014), accelerated absorption of nutrients (Dirksen et al. 2013a), altered gut hormone secretion , and enhanced insulin secretion ) all appear to contribute to the weight loss and improved glucose metabolism seen postoperatively. After RYGB, studies have consistently reported a 7-10 fold enhanced postprandial secretion of glucagon-like peptide-1 (GLP-1), which is of major importance for the improved beta-cell function in patients with preoperative type 2 diabetes (Jørgensen et al. , 2013 and linked to postoperative weight loss . However, the mechanisms behind the enhanced GLP-1 secretion after RYGB remain poorly understood. Several studies have demonstrated increased plasma bile acids (BAs) after RYGB (Nakatani et al. 2009;Jansen et al. 2011;Pournaras et al. 2012;Ahmad et al. 2013;Kohli et al. 2013a;Steinert et al. 2013;Werling et al. 2013;Dutia et al. 2015;Jørgensen et al. 2015), and endogenous BAs have therefore been proposed as keymediators of the improvements in glucose metabolism (Pournaras et al. 2012;Kohli and Seeley 2013;Kohli et al. 2013b;Ryan et al. 2014;Baud et al. 2016). Moreover, studies have demonstrated an association between plasma BAs, GLP-1 and improved glucose tolerance months to years after RYGB (Nakatani et al. 2009;Pournaras et al. 2012;Kohli et al. 2013a;Steinert et al. 2013;Werling et al. 2013;Dutia et al. 2015;Jørgensen et al. 2015). Within the first week after RYGB, GLP-1 secretion is enhanced and glucose tolerance markedly improved while plasma BAs still seem unchanged Jørgensen et al. 2015), although one study reports increased BAs already 4 days after RYGB (Pournaras et al. 2012). Hence, it remains unclear to what extent endogenous BAs are involved in the exaggerated GLP-1 and insulin responses after RYGB.
Administration of BAs has been demonstrated to stimulate GLP-1 secretion in vitro via activation of the TGR5 receptor, which is activated by most BAs (Kawamata et al. 2003;Brighton et al. 2015). A stimulated secretion was also observed in humans after rectal (Adrian et al. 2012;Wu et al. 2013a) and intracolonic infusions (Adrian et al. 1993) in non-operated healthy volunteers (Adrian et al. 1993;Wu et al. 2013a) and patients with type 2 diabetes (Adrian et al. 2012). In contrast, neither intragastric nor upper intestinal administration of BAs were associated with clinically relevant GLP-1 responses Wu et al. 2013b;Hansen et al. 2016), which might suggest that an exposure of the distal small intestine is required for BAs to stimulate robust GLP-1 secretion. As a consequence of the bypass of stomach and upper small intestine after RYGB, nutrients and other ingested molecules will rapidly reach the distal part of the small intestine where the density of GLP-1 producing L-cells is high (Eissele et al. 1992).
The aim of our study was to evaluate whether oral administration of BAs is able to increase GLP-1 secretion in RYGB-operated participants. We studied intake of two different BAs that are available as pharmacological agents used for treatment of gallstones: the primary BA chenodeoxycholic acid (CDCA) and the secondary BA ursodeoxycholic acid (UDCA). We chose to use the highest recommended daily dose for each bile acid according to the label, which was 1250 mg for CDCA and 750 mg for UDCA. We hypothesized that oral intake would stimulate GLP-1 secretion in RYGB-operated participants.
Participants
Eleven (six males and five females) participants were recruited from Hvidovre Hospital's bariatric surgery outpatient clinic (Table 1). Inclusion criteria were uncomplicated primary RYGB >3 months prior to inclusion and normal glucose tolerance (fasting plasma glucose <7.0 mmol/L and HbA1c <48 mmol/mol), and exclusion criteria were anemia (plasma hemoglobin <6.5 mmol/L), dysregulated hypothyroidism, use of anti-thyroid medication, renal insufficiency, prior pancreatitis, certain RYGB complications (severe reactive hypoglycemia, dysphagia), pregnancy or lactation. All participants maintained stable weight during the study period. Medications include vitamins (multivitamins, zink, D-vitamins and B12-injections), proton-pump inhibitors and antihypertensive drugs. None of the participants had cholecystectomy prior to inclusion; however, one participant underwent acute cholecystectomy before the last study day due to an acute attack of simple gallstones. One participant only completed the UDCA day and another participant did not complete the CDCA day; moreover, one participant was excluded from data analysis from the CDCA day due to excessively high morning C-peptide concentrations indicating a non-fasting state. Written informed consent was obtained from all participants before entering the study. The protocol was approved by the Municipal Ethical Committee of Copenhagen in accordance with the Helsinki II declaration, by the Danish Data Protection Agency and was registered at www. ClinicalTrials.gov (NCT 02340247).
Study design
Participants were examined on three separate days separated by at least 2 days with the following interventions: Placebo: Oral intake of 150 mL water; UDCA: Oral intake of UDCA 750 mg (Ursofalk â capsules, Dr. Falk Pharma GmbH, Freiburg, Germany) suspended in 50 mL water followed by oral intake of 100 mL water; CDCA: Oral intake of CDCA 1250 mg (Xenbilox â capsules, Juers Pharma ImportExport GmbH, Hamburg, Germany) suspended in 50 mL water followed by oral intake of 100 mL water. Dosages corresponded to the highest recommended daily dose of CDCA and UDCA respectively.
Sample collection and laboratory analyses
Blood was collected into clot activator tubes for C-peptide analysis and into EDTA-coated tubes for analyses of glucose, GLP-1, glucagon, PYY, GIP, CCK, neurotensin, total bile acids (TBA) and fibroblast growth factor 19 (FGF19). Clot activator tubes were left at room temperature to coagulate for 30 min, whereas EDTA-tubes were immediately placed on ice until centrifuged at 4°C (2000g for 10 min). Plasma glucose was measured using the glucose oxidase technique (YSI model 2300 STAT Plus; YSI, Yellow Springs, OH). Serum for measurements of C-peptide was frozen and stored at À80°C and plasma at À20°C until batch analysis.
Serum C-peptide concentrations were determined using Immulite 2000 analyzer (Siemens Healthcare Diagnostics Inc., Tarrytown, NY, USA). Plasma samples were assayed for total GLP-1 immunoreactivity using antiserum 89390 as previously described (Ørskov et al. 1994). Glucagon was measured by in-house radioimmunoassay (RIA) using antibody 4305 . Total PYY was measured by an in-house RIA using a monoclonal antibody MAB8500 as described elsewhere .
Statistical analyses and calculations
Basal concentrations were calculated as the mean of the three time points (t = À10, À5 and 0 min). Positive incremental area-under-the-curve (piAUC) was calculated using the trapezoid rule as the AUC above basal concentrations to assess postprandial secretions. Data are expressed as mean AE SEM.
Differences between the three study days were analyzed by ANOVA in a linear mixed effects model with study day (placebo, UDCA, CDCA) as fixed effect and individual participants as random effect. Post hoc testing was performed comparing CDCA and UDCA with placebo if ANOVA revealed significant differences between study days. Statistical analysis was performed in R version 3.1.2 (www. R-projec.org) with the package "nlme" for the linear mixed effects models. P < 0.05 were considered statistically significant.
Basal hormone concentrations
Basal hormone concentrations did not differ between study days (Table 2).
Primary outcome: GLP-1 secretion GLP-1 secretion differed significantly between study days (ANOVA: piAUC P < 0.01 and peak P = 0.01). Oral intake of CDCA increased GLP-1 secretion from a basal of 14.3 AE 1.5 pmol/L to a mean peak concentration of Fig. 1A). One participant underwent cholecystectomy before the CDCA day, but GLP-1 responses (piAUC: 1032 pmol/L 9 min, peak: 38 pmol/L) on this day did not differ markedly from the overall mean response to CDCA.
Plasma glucose, serum C-peptide, and plasma glucagon Plasma glucose concentrations did not differ between study days (Table 2, Fig. 1B). C-peptide secretion differed between study days (ANOVA: piAUC P = 0.03) and was slightly increased after oral intake of CDCA (P = 0.01) in contrast to an absent response after UDCA compared with placebo. However, peak C-peptide concentrations were not significantly affected (ANOVA: peak P = 0.54; Table 2, Fig. 1D). Glucagon secretion also differed between study days (ANOVA: P < 0.01 for both piAUC and peak) with increased secretion following CDCA but no effect of UDCA (Table 2, Fig. 1F).
Other gut hormones PYY secretion differed between study days (ANOVA: P ≤ 0.01 for both piAUC and peak) with increased secretion after CDCA compared with placebo, whereas UDCA did not stimulate PYY secretion (Table 2, Fig. 1C). Intake of CDCA resulted in a prolonged PYY secretion, i.e. concentrations did not return to basal level within 180 minutes after intake (P = 0.02). Secretion of neurotensin also differed between study days (ANOVA: P < 0.01 for both piAUC and peak) with clearly enhanced secretion after CDCA in contrast with no stimulation during UDCA compared with placebo (Table 2, Fig. 1E). GIP was not affected by any of the two BAs (Table 2, Fig. 1H). Peak of CCK concentrations (ANOVA: P = 0.05) tended to increase after CDCA (P = 0.06) but not after UDCA (P = 0.49) compared with placebo, while piAUC was unaffected by both BAs (ANOVA: P = 0.57; Table 2, Fig. 1G).
Plasma TBA and FGF19
Both plasma TBA and FGF19 differed between study days (ANOVA: P < 0.01 for both piAUC and peak). CDCA increased plasma TBA significantly, while only a slight numerical increase in piAUC (P = 0.07) and no change in peak concentrations (P = 0.41) were observed following UDCA (Table 3, Fig. 2A) compared with placebo. FGF19 concentrations increased after CDCA compared with placebo (P ≤ 0.01 for both piAUC and peak) and were unchanged following UDCA (P = 0.91 for piAUC, P = 0.16 for peak; Table 3, Fig. 2B).
VAS ratings, blood pressure, and heart rate
Oral intake of BAs neither changed VAS ratings for appetite perception (hunger or satiety), nausea, or stomach pain nor affected blood pressure or heart rate (data not shown). However, one patient did report increased VAS (64 mm from baseline) in stomach pain 30 min after intake of CDCA. This was followed by a single defecation after which VAS returned to baseline.
Discussion
We investigated the effects of oral administration of two different bile acids (BAs), ursodeoxycholic acid (UDCA) and chenodeoxycholic acid (CDCA) in RYGB-operated participants.
We found that administration of the primary BA, CDCA, stimulated secretion of GLP-1, PYY, neurotensin, C-peptide and glucagon, whereas plasma glucose, CCK and GIP concentrations were unaffected. In contrast, ingestion of the secondary BA, UDCA, did not affect any hormone concentrations significantly. Interestingly, CDCA induced peak GLP-1 concentrations of 29 pmol/L and notably, this was obtained without concomitant intake of nutrients or calories. After RYGB, passage of orally ingested substances is accelerated through the gastric pouch that drains directly into the lower part of the jejunum (Dirksen et al. 2013a); accordingly, oral intake of BAs corresponds to administration directly into distal parts of the small intestine. Thus, our findings add to the existing literature and support that GLP-1 secretion after BA administration depends on the site of delivery. This may be particularly relevant for unconjugated BAs, which normally will be passively absorbed in the upper intestine before reaching the distal small intestine after oral administration in non-RYGB individuals (Ma and Patti 2014).
In RYGB-operated patients, however, GLP-1 secretion may reach peak concentrations up to~150 pmol/L after a liquid mixed meal , hence CDCA does not seem to activate the secretory machinery of the L-cell to its full capacity-at least not when BAs are ingested alone without nutrients. Nevertheless, the GLP-1 secretion after CDCA was associated with a small increase in C-peptide concentrations even without co-ingestion of nutrients and without changes in glucose. Whether this insulin secretion was actually mediated by GLP-1 cannot be determined from this study, but if so, we would not expect a major response given that the insulinotropic effect of GLP-1 is glucose-dependent (Hvidberg et al. 1994).
Administration of BAs has been demonstrated to stimulate PYY in vitro via activation of TGR5 (Bala et al. 2014). Stimulation also appears to occur in vivo (Adrian et al. 1993(Adrian et al. , 2012Meyer-Gerspach et al. 2013;Wu et al. 2013a), as demonstrated after intraduodenal , intracolonic (Adrian et al. 1993), and rectal (Adrian et al. 2012) infusions of BAs (CDCA, deoxycholate (DCA) and TCA, respectively), and further supported by our findings of an increase in PYY concentrations after oral intake of CDCA. Notably, markedly higher peak concentrations of PYY have been reported after ingestion of mixed meals in both non-operated subjects and RYGB-operated patients as well as after rectal administration of BAs (Adrian et al. 2012;Wu et al. 2013a). Neurotensin appears to be a co-player with GLP-1 and PYY in regulating metabolic functions, including appetite, and glucose homeostasis regulation (Svendsen et al. 2015;Grunddal et al. 2016) and as for GLP-1 and PYY, RYGB is associated with greatly exaggerated postprandial secretion of this peptide (Dirksen et al. 2013b). We found that administration of CDCA stimulated secretion of neurotensin and the secretory profile was rather similar to that of GLP-1. Interestingly, the secretory profiles of GLP-1 and neurotensin appeared to differ from the profile of PYY on the CDCA day, with GLP-1 and neurotensin returning to baseline level after 120-150 min, whereas PYY levels remained high throughout the 180 min follow-up. The present observation is in contrast to the parallel dynamic changes of the hormones often observed during a mixed meal test . A similar prolonged secretion of PYY has not been reported previously and caused us to speculate that CDCA may, in succession, first stimulate the L-cells in the proximal ileum followed by stimulation of the L-cells in the distal part of the intestine including the colon. Indeed, there is data to suggest that the L-cells in the distal part of the gut primarily secrete PYY, as demonstrated in several animal models (Svendsen et al. 2015;Grunddal et al. 2016;Wewer Albrechtsen et al. 2016). Our results suggest that CDCA is able to simultaneously enhance the secretion of several enteroendocrine cell products (GLP-1, neurotensin, PYY), which could have synergistic beneficial actions with respect to appetite inhibition (Schmidt et al. 2016).
We observed increased concentrations of glucagon only after CDCA intake, which is in line with previous observations Hansen et al. 2016). It is well established that RYGB-operated patients have increased postprandial glucagon concentrations concomitantly with high glucose, GLP-1 and insulin concentrations Jørgensen et al. 2013;Svane et al. 2016). The cause of this paradoxical hyperglucagonemia is still unknown, but extrapancreatic secretion may occur, as observed after pancreatectomy using Whipple procedure (Holst et al. 1983;Lund et al. 2016), which implies similar gastrointestinal reconstructions as after RYGB. Another explanation of our findings could be a direct stimulatory effect of CDCA on the alpha-cells in pancreas. However, in contrast with the case for CDCA-stimulated GLP-1 secretion (Brighton et al. 2015), activation of TGR5 is unlikely be involved as the receptor has been found not to be expressed in primary murine alpha-cells . BA administration did not change GIP or CCK concentrations in our study, although there was a tendency for slightly increased peak concentrations of CCK after CDCA. In line with these results, other studies have demonstrated increased CCK after intraduodenal ) and intragastric (Hansen et al. 2016) CDCA infusions and no effect on GIP concentrations (Hansen et al. 2016).
In this study, we found a marked difference in gut hormone responses between administration of CDCA (1250 mg) and UDCA (750 mg). As this study was an exploratory study of CDCA and UDCA in RYGB-operated participants, we chose to use the highest recommended daily dose for each bile acid according to the label, which was 1250 mg for CDCA and 750 mg for UDCA. The study was therefore not designed to compare the relative effects of CDCA and UDCA on gut hormone secretion, but rather to investigate whether these BAs stimulated gut hormone secretions at all. Dose-response studies of different BAs with respect to GLP-1 secretion have not been performed in humans, but an in vitro study has demonstrated that CDCA activates TGR5 while UDCA was without effect (at equal concentrations) (Kawamata et al. 2003). Differences in the absorption of the two BAs could also be of importance for the ability to stimulate release of gut hormones. In a recent study, BAs activated basolateral but not apically located TGR5 receptors, indicating that BAs must be absorbed from the intestinal lumen before they can stimulate GLP-1 secretion (Brighton et al. 2015). In addition to the acute TGR5 mediated effects, BAs may have additional and potentially longer lasting effects via activation of the nuclear receptor FXR (Ma and Patti 2014). Activation of FXR regulates insulin secretion (Renga et al. 2010) and induces expression and secretion of FGF19, which stimulates glycogen synthesis and reduce hepatic gluconeogenesis (Ma and Patti 2014). Knockout of FXR in mice eliminated the beneficial effects of sleeve gastrectomy on glucose metabolism (Ryan et al. 2014). Activation of FXR does not seem to have stimulatory effects on GLP-1 secretion (Brighton et al. 2015) and may even decrease secretion (Trabelsi et al. 2015). CDCA is the most potent ligand for stimulating and activating FXR, while UDCA has negligible FXR-activities (Ma and Patti 2014). Whether a single administration of BAs can activate the FXR-pathway is unknown, but in this study we demonstrated that CDCA increased FGF19 secretion after 90 min, although we do not know whether this was indeed a result of FXR pathway activation.
After RYGB, plasma BA concentrations increase both in the fasting state and postprandially (Pournaras et al. 2012;Kohli et al. 2013a;Steinert et al. 2013;Werling et al. 2013;Jørgensen et al. 2015). Endogenously secreted BAs in humans consist mostly of the primary BAs, cholic acid (CA) and CDCA and the secondary BA, DCA; most BAs are conjugated with glycine and the remaining with taurine (Werling et al. 2013 support BAs as key mediators of the improvements in glucose metabolism after RYGB and sleeve gastrectomy (Pournaras et al. 2012;Kohli and Seeley 2013;Kohli et al. 2013b;Ryan et al. 2014;Baud et al. 2016), but inconsistencies with respect to timing of the postoperative changes in BAs and the effects of RYGB on GLP-1 secretion and glucose metabolism are observed in human studies (Ahmad et al. 2013;Steinert et al. 2013;Dutia et al. 2015;Jørgensen et al. 2015). Notably, administration of exogenous BAs causes much higher plasma TBA concentrations (obviously depending of the BA dosage) than seen in RYGB-operated patients. In our study, CDCA administration increased TBA concentrations up to 5-fold with peak concentrations of 55 lmol/L. In comparison, postprandial TBA concentrations after RYGB reach peak concentrations of only 8-9 lmol/L 12-15 months postoperatively (Werling et al. 2013;Jørgensen et al. 2015). Thus, despite our findings of increased GLP-1 secretion in RYGB-operated participants in response to pharmacological doses of CDCA, we cannot make any conclusions regarding the possible physiological effects of endogenous BAs for improved glucose metabolism after RYGB, which obviously is of great interest. The use of only one dose of each BA, and two different doses of BAs, further limit the conclusions with respect to relative effects of CDCA versus UDCA on gut hormone secretion. Another interesting research question is whether exogenous administered BAs can be used to enhance endogenous gut hormone secretion in combination with meal intake in obese and RYGB-operated patients. If so, BA administration could be a potential treatment for patients with obesity in general, and in particular RYGB-operated patients with inadequate glycemic control or poor weight loss responses after surgery.
In conclusion, oral administration of the bile acid, chenodeoxycholic acid, to RYGB-operated participants stimulated GLP-1, PYY, neurotensin, C-peptide, and glucagon secretion in the absence of nutrients and without changed glucose concentration. No effect was seen following ursodeoxycholic acid. Chenodeoxycholic acid may therefore be viewed as a molecular enhancer of several gut hormones of potential importance for the future treatment of diabetes and obesity.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T03:05:22.446Z", "created": "2017-02-01T00:00:00.000", "id": "530586", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.14814/phy2.13140", "pdf_hash": "e21701bef3d98e9b1c2fe92fc6440b8a21e2e635", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39695", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "e9cbc048c5abc1dca2fadf1ff3ae156696cbd133", "year": 2017}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
CYP2B6rs2279343 Is Associated with Improved Survival of Pediatric Rhabdomyosarcoma Treated with Cyclophosphamide
Background Rhabdomyosarcoma (RMS) is a small round blue cell malignant tumor, representing 7% of childhood malignancies, and over 50% of all soft tissue sarcomas. Cyclophosphamide (CPA) is a prodrug and is the mainstay of RMS treatment. CYP2B6 is a highly polymorphic drug metabolizing enzyme involved in CPA bioactivation. The influence of CYP2B6 single nucleotide polymorphisms (SNPs) on the survival of RMS is still unknown. Methods We genotyped CYP2B6SNPs rs2279343, rs3745274, and rs3211371 by restriction fragment polymorphism (RFLP) after PCR amplification in a cohort of 73 pediatric RMS patients treated with CPA-based first line treatment. We then analyzed the association between those genotypes and survival outcome of RMS. Results The frequencies of CYP2B6 rs2279343, rs3745274, and rs3211371 were 63%, 45.2%, and 5.5%, respectively. There was no association between rs3745274, rs3211371 genotypes and survival outcomes of RMS. However, the carriers of at least one mutant allele CYP2B6rs2279343 had significantly longer event-free survival (p-value = 0.03). Conclusion Our results demonstrated that CYP2B6 rs2279343 may predict EFS in RMS patients and warrants future studies to clarify the pharmacogenetics of CPA in pediatrics. If validated, integration of genetic factors with clinical and molecular characteristics could be used for a composite algorithm to better stratify risk prior to treatment.
Introduction
Rhabdomyosarcoma (RMS) is a small round blue cell tumor, representing 7% of all childhood malignancies [1]. Approximately, 350 new cases of RMS aged less than 19 years old are diagnosed annually in the United States with an incidence of 0.4414 cases per 100,000 children per year [2]. RMS can arise from a variety of anatomic sites and is pathologically subdivided into two distinct subgroups; the alveolar and the embryonal. The embryonal subtype is more frequent and shows better prognosis [3]. Currently, treatment of RMS is based on risk stratification into low, intermediate, and high risk which is determined by both the pre-treatment clinical TNM staging and the surgical grouping [4]. Risk stratification and multimodality care contributed to treatment improvement over the years, where cure rates went from 25% in the early 1970s to 70% over the last 40 years [5]. The combination of vincristine, actinomycin, and cyclophosphamide (VAC) has been used for decades as the standard chemotherapy regimen for treating RMS by the Intergroup Rhabdomyosarcoma Study Group (IRSG). About 35% of RMS can achieve complete response (CR) with chemotherapy alone after initial biopsy or incomplete excision [6].
Cyclophosphamide (CPA) is an oxazaphosphorine alkylating chemotherapeutic agent and is considered the mainstay of RMS treatment. It is a prodrug, which 70-80% of the dose is biotransformed to its active metabolite (4-hydroxycyclophosphamide). Different hepatic cyto-chromeP450 (CYP450) enzymes have been implicated in CPA activation, including CYP2B6, CYP2C9, CYP2C19, and CYP3A4/5. Studies proved that CYP2B6 is the primary CYP450 responsible for CPA activation [7,8]. CYP2B6 is highly polymorphic, with several known variant alleles [9,10]. The CYPalleles website (www.cypalleles.ki.se/cyp2b6.htm) currently lists > 100 known single nucleotide polymorphisms (SNPs) in CYP2B6 in humans with distinct ethnic frequencies [11]. The CYP2B6 polymorphic variants have been shown to exhibit different metabolic properties than the wild type enzymes [12,13]. With the emerging concepts of personalized medicine, investigators started looking at the role of using common functional CYP2B6 SNPs to predict response to different CYP2B6 substrates and to tailor doses for each patient [14]. Nevertheless, their role remains controversial and definitive evidence is lacking, due to sparse, and even contradicting information [15][16][17].
Understanding the factors affecting CPA metabolism may aid to clarify causes of inter-individual variability in CPA response. The ability to predict treatment failure would help improve CPA dosing regimens. Our hypothesis is that SNPs in CYP2B6 might affect the functionality of this enzyme in activating CPA, which could be reflected on the response to treatment. Based on such considerations, we conducted this study to investigate the prevalence of 3 CYP2B6 common functional SNPs rs2279343 (A785G, CYP2B6 Ã 4, exon 5, K262R), rs3745274 (G516T, CYP2B6 Ã 9, exon 4, Q172H), and rs3211371 (C1459T, CYP2B6 Ã 5, exon 9, R487C) in our cohort of Egyptian pediatric RMS patients. Our aim was to examine the possible influence of CYP2B6 polymorphism on the chemotherapy treatment response, and survival outcomes in RMS.
Patient eligibility and treatment
This retrospective study included 73 histopathologically confirmed RMS patients under the age of 18 years. They were recruited at the Children's Cancer Hospital Egypt (CCHE) from March 2008 until December 2012 and were treated with a combination of vincristine, actinomycin and cyclophosphamide (VAC) as the first line standard clinical treatment for RMS according to IRS-IV protocol. This regimen comprises Vincristine (1.5 mg/m 2 ) administered as an IV push (max. 2 mg), actinomycin (0.045mg/kg) as a 5 min intravenous infusion and cyclophosphamide (1.2-2.2 gm/m 2 ) infused over 30-60 min, with hydration continued at 3000 ml/m 2 / day followed by a weekly dose of vincristine for 2 weeks. Local control (radiotherapy or surgery) was given in accordance with the treatment protocol. Patients were followed-up till March 2015.
Ethics statement. This study was approved by CCHE Institutional Review Board. A written informed consent was obtained from parents/guardians before a patient was enrolled in the study.
Blood sampling. Peripheral blood (4 ml) was withdrawn from participants and collected in EDTA vacutainers. Genomic DNA was extracted by the salting out procedure [18]. DNA purity and concentration were measured using Nanoquant ™ spectrophotometer (Infinite M200, TECAN, Switzerland). DNA was kept at -20°C until pharmacogenetic analysis.
CYP2B6 genotyping
Genotyping for CYP2B6 SNPs rs3745274 in exon 4 (G516T), rs2279343 in exon 5 (A785G), and rs3211371 in exon 9 (C1459T) was performed by restriction fragment length polymorphism (RFLP) after polymerase chain reaction (PCR) amplification. All PCR amplifications were carried out in similar standard protocol conditions [10]. Slight modifications were done, for rs3745274 the annealing temperature was 56°C and the extension step was reduced to 40 s while, for rs3211371, a prolonged extension step of 1 min 30 s and the annealing temperature was 65°C. All primers' sequence for both forward and reverse pairs, restriction enzymes and interpretation of post-RFLP products for the studied SNPs are specified in (Table 1).
DNA sequencing. For laboratory validation and substantiation of the protocol and the results capillary Sanger sequencing was used. PCR amplification was performed according to the PCR protocol for the PCR-RFLP assay. Subsequently, the PCR products were purified using the DNA Clean & Concentrator™-25 (DCC™-25) kit, followed by sagner sequencing of both strands using the BigDye1 Terminator Cycle Sequencing Kit v.3.1 (Applied Biosystems). The reaction mixture for the sequencing reaction contained 4 μl ABI PRISM Big Dye Terminator (Applied Biosystem), 10 μl water, 2 μl BigDye1 Terminator Buffer (Applied Biosystems), 1 μl primer (3.2 pmol) and 3 μl of purified template (40 ng); a total volume of 20 μl. The sequencing conditions were: one cycle at 96°C for 1 min and 25 cycles (96°C for 30 s, 56°C for 5 s, 60°C for 4 min). Forward and reverse sequencing primers were used to sequence both strands of the gene and were the same used for the PCR reactions (see Table 1). The sequence products were purified by gel filtration chromatography using Centri-Sep™ (Thermo Fisher, United States), in order to eliminate excess primers and/or unincorporated dideoxynucleotides (dNTPs). All amplified products were resuspended in 10 μl formamide, and before sequencing analysis, submitted to denaturation at 96°C for 3 min. After being separated on an automated sequencer (ABI PRISM-3130 Genetic Analyzer, Applied Biosystems), DNA sequencing reactions were carried out using a 3730 48-Capillary DNA Analyzer (Applied Biosystems) and MicroAmp Optical 96 Well Plates (Applied Biosystems).
Clinical evaluation criteria
According to the response evaluation criteria in solid tumors (RECIST) guidelines version 1.0 [19], all patients' tumor response was evaluated as complete response (CR), partial response (PR), stable disease (StD), or progressive disease (PD).
Statistical Analysis
The analysis was done using SPSS, version 17 for Windows (IBM Corp., USA), and a p-value of 0.05 was considered a statistically significant result. Descriptive statistic was presented as mean and standard deviation for continuous measures and frequencies and percentages for categorical measures. Tumor objective response outcomes were defined as responders (CR, PR, StD) and non-responders (NR, PD). Correlation between various genotypic variants with treatment outcomes was examined using univariate regression. Association was expressed as odds ratios (OR) with 95% confidence interval (CI). The primary endpoints of the current study were event-free survival (EFS) and overall survival (OS), which were estimated according to Kaplan-Meier method and log-rank test. EFS was defined as the time from the initial diagnosis until the date of disease progression, recurrence, or death due to any cause. The OS was defined as the interval from the initial diagnosis to death or time of the last contact.
Association of CYP2B6 genotypes with objective clinical response, OS, and EFS. The survival data showed that 3-years EFS after VAC chemotherapy was 53.1%. Table 4 summarizes the association of CYP2B6 SNPs with the objective response and the survival outcome. Carriers of at least one mutant allele rs2279343 (n = 46/73) when compared to wild-type homozygotes, showed a significant association with the clinical objective response (pvalue = 0.01), and a significantly longer EFS (p-value = 0.03) (Fig 3), but there was no discernable effect on the OS (p-value = 0.48). On the other hand, there was no observed association between neither rs3745274 nor rs3211371 with any of the study end points. Discussion CPA is an anticancer prodrug, which in order to exert its antitumor activity requires bioactivation to the active alkylating metabolite, 4-OH-CPA [20]. There is a significant heterogeneity in CPA treatment response. CYP2B6 was identified as the major CPA 4-hydroxylase catalyzing the metabolism of CPA [7,8] [URL]-3. Strategies to identify patients at risk of treatment failure and those who will require further dose adjustment might help in decisions regarding treatment doses, intensifying chemotherapy doses or length of treatment [21]. Many studies investigated the impact of CYP2B6 on the clinical outcome of different substrates (e.g bupropion, efavirenz, bevacizumab) and concluded that CYP2B6 effect is substrate specific [14][15][16][17]. A previous study has found a significantly higher incidence of relapse and graft failure in 66 pediatric patients receiving a busulfan-based conditioning regimen carrying a homozygous reduced functional CYP2B6 allele compared with carriers of at least one wild allele [22]. On the other hand, some studies did not show any effect of CYP2B6 polymorphisms on CPA treatment outcome [23,24].
Altogether in vivo and in vitro studies have not reached a conclusive evidence to support the role of different CYP2B6 polymorphisms in predicting CPA effect. Helsby and Tingle attributed this inconsistency of data to different study size design as well as lack of consistency in allele definition and genotype information among studies [25]. Yet, our study is one of few pharmacogenetic studies carried out on pediatric population aiming to investigate the influence of CYP2B6 variants on CPA treatment outcome.
Based on this, rather than choosing a "genome association study approach" with complex data analysis, our study design was based on a "candidate SNP approach" to examine particular SNPs in CPA activation pathway that are thought to influence response to treatment. We have selected 3 common functional SNPs (rs3745274, rs2279343, and rs3211371) found in CYP2B6 exons 4, 5, and 9, respectively. They were carefully selected based on evidence that they are common in Caucasians [10], they cause changes in CYP2B6 coding that would affect CYP2B6 enzyme expression or activity and are associated with a change in CPA activation and metabolism [24,26]. Our hypothesis is that the pharmacological activity of CPA could be altered by the functionality of CYP2B6 which might be reflected on the treatment response. Thus, interpatient variability in CPA response may be due in part to altered CPA metabolism by the CYP2B6 polymorphic variants. In light of these considerations, the aim of this study was to analyze the frequencies of allelic variants of CYP2B6 in pediatric Egyptian RMS patients. Furthermore, to determine if they influence treatment efficacy and survival outcome and hence, can be used as a prognostic biomarker for response. A total of 73 children with RMS were recruited and genotyping of the selected SNPs were analyzed with all the clinical data and outcome.
CYP2B6 is known to have inter-ethnic variability, resulting in a high genetic diversity among different populations as shown in Table 5. In our cohort, the percentage of patients carrying at least one copy of CY2B6rs2279343, rs3745274, and rs3211371 were 63%, 45.2%, and 5.5%, respectively. The main finding in our study was that in RMS patients treated with VAC as the first line of treatment, there was an association between CYP2B6rs2279343 (carrying G mutant allele) and response to CPA-based treatment. Patients who carried at least one mutant allele CYP2B6rs2279343 (n = 46/73), had better objective clinical response compared to the homozygous wild allele carriers (p-value = 0.01). Also, our results demonstrate that this group of patients who carried at least one mutant allele in CYP2B6rs2279343 (G allele), had a significant longer EFS (p-value = 0.034) but not OS (p-value = 0.48) compared to those carrying the homozygous wild allele (A allele). Our results have provided evidence for the involvement of CYP2B6rs2279343 in the response of RMS patients to CPA-based treatment. This is in accordance with a previous study that identified a strong functional impact of rs2279343 on enhancing the catalytic activation of CPA into 4-OH-CPA and a tendency to reduce protein expression [8]. In contrast to several studies which associated variants including CYP2B6rs2279343 with lower CPA hydroxylation or with poor response [17,29]. Raccor et al. found that CYP2B6rs2279343 genotype caused 50% reduction in 4-OH-CPA formation yet the sample size was too small to draw a reliable conclusion. In addition, they noted that carrying rs2279343 alone or rs3745274 alone displayed a reduction in intrinsic CPA clearance compared with the wild type [30]. In this context, our hypothesis generated in this study is that patients carrying CYP2B6rs2279343 (heterozygous or mutant homonozygous genotypes) have a predicted rapid metabolic phenotype; consequently, maintaining an increased CPA activation to 4-OH-CPA, while patients with the wild-genotype will have a higher ratio of CPA to 4-OH-CPA in plasma. Thus, carriers of at least one mutant allele CYP2B6rs2279343 have a higher success rate for RMS therapy with CPA.
In summary, our results provide a picture of the role of CYP2B6 polymorphisms in RMS. Our data suggests that pretreatment evaluation of CYP2B6rs2279343 may explain some interindividual differences in treatment response. However, this should be replicated in a larger cohort and more studies are warranted to clarify the pharmacogenetics of CPA in pediatrics. If validated, integration of genetic factors with clinical and molecular characteristics could be used for a composite algorithm to better stratify risk prior to treatment. Moreover, this may facilitate clinical decision, improve CPA treatment, and hence improve the clinical outcome. Nevertheless, one limitation is that our study did not correlate the genetic variation with serum drug metabolite levels. The pharmacokinetic modeling of CPA activation will give a more clear understanding of the inter-individual variation in drug response and how to stratify RMS patients based on their genetic makeup.
==
Domain: Biology Medicine
|
[{"added": "2018-04-03T04:21:37.229Z", "created": "2016-07-07T00:00:00.000", "id": "15451083", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0158890&type=printable", "pdf_hash": "891e31bc60af92fe8d3959e739c2b78f0196285d", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:35919", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "891e31bc60af92fe8d3959e739c2b78f0196285d", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Minor Role of Mitochondrial Respiration for Fatty-Acid Induced Insulin Secretion
An appropriate insulin secretion by pancreatic beta-cells is necessary to maintain glucose homeostasis. A rise in plasma glucose leads to increased metabolism and an elevated cytoplasmic ATP/ADP ratio that finally triggers insulin granule exocytosis. In addition to this triggering pathway, one or more amplifying pathways—activated by amino acids or fatty acid—enhance secretion by promoting insulin granule recruitment to, and priming at, the plasma membrane. The aim of this study was to clarify the impact of the mitochondrial respiratory activity on fatty acid-induced insulin secretion that was assessed by an extracellular flux analyzer. Treatment of isolated mouse islets with glucose (20 mM) increased insulin secretion 18-fold and correlated with ATP-synthesizing respiration. Furthermore, oxygen consumption rate (OCR) significantly increased by 62% in response to glucose, whereas the addition of palmitate resulted only in a minor increase of OCR at both 2.8 mM (11%) and 20 mM glucose (21%). The addition of palmitate showed a pronounced increase of coupling efficiency (CE) at 2.8 mM glucose but no further insulin secretion. However, treatment with palmitate at 20 mM glucose increased insulin secretion about 32-fold accompanied by a small increase in CE. Thus, fatty acid induced respiration has a minor impact on insulin secretion. Our data clearly demonstrate that fatty acids in contrast to glucose play a minor role for respiration-mediated insulin secretion. In the presence of high glucose, fatty acids contribute partially to amplifying pathways of insulin secretion by further increasing mitochondrial activity in the islets of Langerhans.
Introduction
Insulin secretion from islets initiated in response to nutrients is the main physiological reaction to maintain normal blood glucose levels and markedly depends on ATP. Glucose, fatty acids and amino acids are the three substrates an organism can use to maintain metabolic homoeostasis and are required for the generation of energy (e.g., as ATP). The prime pathway for the degradation of fatty acids is mitochondrial fatty acid β-oxidation [1,2], a key metabolic pathway for energy homoeostasis in organs such as the liver, heart and skeletal muscle. In general, organs and cells are flexible in the choice of the substrate used for energy production. Under normal conditions, glucose is the preferred substrate for oxidation, whereas under specific conditions, e.g., during fasting, fatty acids and ketone bodies become more important as an alternative energy source. This reciprocal relationship or competition between the oxidation of fatty acids and glucose is also known as the glucose-fatty acid or Randle cycle [3].
Pathophysiological conditions like type 2 diabetes are associated with changes in circulating glucose and lipid levels which influence the function of pancreatic islets and insulin secretion [4,5]. Insulin secretion is regulated via two different pathways, the triggering and the amplifying pathway [6]. The triggering pathway includes conversion of glucose to ATP via glycolysis and the respiratory chain, closure of ATP-dependent K + channels with depolarization of the plasma membrane resulting in an influx of calcium and the release of insulin. Superimposed on this pathway, one or more amplifying pathways are described to enhance secretion by promoting insulin granule recruitment and fusion with the plasma membrane. These include the receptor-mediated generation of cAMP and other signals and metabolic mediators such as NADPH [7]. However, all substrates and mechanisms of the amplifying pathways are incompletely understood. Fatty acids have enormous capacity to amplify glucose-stimulated insulin secretion [8,9] which is particularly operative in situations of beta-cell compensation for insulin resistance. They for instance bind to the free fatty acid receptor 1 resulting in enhancement of glucose-stimulated accumulation of cytosolic Ca 2+ and consequently insulin secretion [10,11]. However, chronically elevated fatty acids, particularly in the presence of elevated glucose levels, reduce insulin biosynthesis [12], secretion [13] and participate in beta-cell loss [4,14,15].
Since it is known that both the triggering and the amplifying pathways are absolutely dependent on the mitochondrial metabolism of glucose, the aim of this study was to investigate the contribution of palmitate-induced insulin secretion by activation of the respiratory chain. Recently, Wikstrom et al. [16] described the development of a novel islet respirometry assay to determine the bioenergetic efficiency of islets as a tool to study islet mitochondrial function. Thus, in parallel to measurements of insulin secretion in isolated islets treated with glucose and palmitate we used the Seahorse XF24 extracellular flux analyzer to evaluate oxygen consumption rate and to calculate ATP synthesis, proton leak and coupling efficiency.
Results and Discussion
Before comparing the capacities of nutrient-stimulated insulin secretion and the mitochondrial metabolism of intact cells the detection of oxygen consumption with the Seahorse XF24 extracellular flux analyzer was optimized. In previous studies islets of both DBA/2J and B6. V-Lep ob mice were used without showing differences in their glucose-dependent increase in oxygen consumption rate (OCR). Since B6. V-Lep ob mice exhibit a higher number of islets the following experiments were exclusively performed with islets of this mouse strain. We tested the oxygen consumption of islets of different size and observed that islets with a maximal size of 150 µm but not islets larger than 150 µm gave a significant increase of OCR in response to elevated glucose concentrations (Figure 1a). In contrast to the in vivo situation isolated islets are not connected to vessels and consequently a limited diffusion of oxygen from the inner beta cells of larger islets and/or an impaired fuel supply may occur. Therefore all following studies were performed with islets <150 µm. Figure 1b illustrates the insulin secretion of isolated islets that were exposed to both low and high glucose concentrations in either the absence or presence of palmitate. The basal release of insulin at a glucose concentration of 2.8 mM increased 1.2-fold in response to 0.5 mM palmitate. Under high glucose condition (20 mM) insulin secretion increased 18-fold and the addition of 0.5 mM palmitate resulted in a further augmentation of insulin release through fatty-acid stimulation (32-fold in comparison to 2.8 mM glucose without palmitate). These data confirmed that in the acute state fatty acids enhance the glucose-stimulated insulin secretion [14,17,18], thereby strengthening the hypothesis that an activated triggering pathway is essential for a sufficient increase in insulin release via the amplifying pathway [19].
In a parallel set of experiments we used isolated islets to evaluate the impact of palmitate at low (2.8 mM) and high (20 mM) glucose concentrations for the oxygen consumption via mitochondrial metabolism. With the extracellular flux analyzer we observed that 0.5 mM palmitate increased basal OCR by 11% ( Figure 1c). In contrast, basal OCR was increased by 62% in response to 20 mM glucose. The subsequent addition of 0.5 mM palmitate in the presence of 20 mM glucose increased the OCR by 21% (Figure 1d), indicating that palmitate is generally metabolized for ATP production via the respiratory chain however, to a low extent in comparison to its strong effect on insulin secretion. Thereby, we have demonstrated for the first time that fatty acids induce a modest activation of mitochondrial metabolism and ATP synthesis in isolated islets and conclude that their ability to augment insulin secretion at high glucose conditions is mainly mediated via the amplifying pathway. The detection of the ATP/ADP ratio of isolated islets treated with glucose alone or glucose plus palmitate would give further information about the mitochondrial capacity. Due to limited islet material this approach could not be included in the actual study, however it will be a focus in future experiments using beta-cell lines such as Min6 cells.
We next measured OCR after the application of different stimuli to calculate the coupling efficiency (CE) which reflects the efficiency of mitochondrial energy conversion after blocking ATP production and mitochondrial respiration. The basal OCR of 0.69 ± 0.08 pmol·min −1 ·ng DNA −1 was elevated to 1.03 ± 0.11 pmol·min −1 ·ng DNA −1 by 20 mM glucose (Figure 2a). In order to study the relationship between insulin secretion and ATP synthase-dependent OCR we calculated linear regression during different stimulatory conditions. Thereby, we were able to determine insulin secretion and OCR of isolated islets from the same animal. We calculated a strong correlation between ATP synthase-dependent OCR and insulin secretion of isolated islets treated with low glucose (low Glc; R 2 = 0.6137) or high glucose concentrations (high Glc; R 2 = 0.7275). In contrast, no correlation between ATP synthase-dependent OCR and insulin secretion exists in response to palmitate either at low glucose (low Glc + PA; R 2 = 0.0328) or at high glucose exposure (high Glc + PA; R 2 = 0.0985) (Figure 2b). Calculations of CE which is about 0.43 in the basal state showed a significant increase in response to glucose (to 0.60), whereas palmitate induced only a slight increase of CE at low (0.54) and high glucose concentrations (0.66) (Figure 2c). The weak increase of oxygen consumption and coupling efficiency in response to palmitate indicates that the amplification itself is nearly independent of mitochondrial activity. However, since we cannot exclude non-linear kinetics of mitochondrial respiration and insulin secretion a parallel detection of the membrane potential of isolated beta-cells will be required in future studies. The amplifying pathways are very poorly understood, in terms of how fatty acids and/or amino acids couple to the mechanics of exocytosis of insulin carrying granules [20,21]. Alquier et al. [10] suggested that binding of fatty acids to a G-protein coupled receptor, the free fatty acid receptor 1, results in elevation of intracellular calcium as a part of the amplifying pathway [10,11]. Moreover, fatty acids are known to be degraded in mitochondria via beta-oxidation. However, high concentrations of glucose and its metabolite malonyl-CoA inhibit carnitine palmitoyltransferase 1 (CPT1) and the subsequent import of fatty acids into the mitochondria [20,21] leading to an accumulation of fatty acids in the cytosol. This overwhelming increase of fatty acids within the cell has two consequences, on the one hand the production of toxic ceramides [21] and on the other hand the ATP-consuming storage of lipids as triglycerides [5,20]. This cycling of lipids and its intermediates after fatty-acid stimulation is suggested to produce signals which participate in the amplifying pathway and augments insulin secretion nearly independent of fatty-acid induced mitochondrial activation.
Furthermore, calculations of proton leak confirmed that islets are highly uncoupled [16]. However, no alterations in uncoupling resulted after stimulation with glucose or palmitate (Figure 2d). Fatty acids are described to produce superoxides [22,23] and to activate uncoupling via UCP2, a highly expressed proton transporter in islets of Langerhans [23,24]. During the process of uncoupling protons of the respiratory chain reenter the mitochondrial matrix through a so called proton leak mediated by e.g., UCP2 without producing ATP. In contrast to former studies performed with isolated mitochondria of INS1 cells [25,26] we showed that fatty acids did not have an impact on proton leak when applied to isolated islets. This might be due to the fact that the characterization of mitochondrial action of the entire islet has the disadvantage that not all cells are beta-cells and that the other cell types (e.g., alpha-cells) also participate in oxygen consumption. Therefore, further studies with isolated and FACS sorted beta-cells are needed to specify their exact contribution. These data indicate that a sufficient defense of isolated islets exists against fatty-acid mediated reactive oxygen species production under short-term stimulation. Additionally, palmitate is probably not involved in the regulation of uncoupling in isolated mammalian islets of Langerhans. However, these data should be confirmed by studies of beta-cells lines such as INS cells with regard to short-and long-term incubation of fatty acids. It would be possible that fatty acids increase proton leak due to prolonged exposure when defense against reactive oxygen species becomes insufficient and thus insulin secretion would be influenced by superoxide production and uncoupling.
In summary, we showed that fatty acids exhibit a low capacity to increase respiratory activity in isolated islets especially at low glucose concentrations when palmitate has no apparent influence on insulin secretion. Moreover, palmitate appears to play only a minor role in fatty-acid activated mitochondrial respiration under high glucose conditions and cannot explain the 32-fold enhancement of insulin secretion that was induced by the addition of palmitate at 20 mM glucose. The minor role of fatty acids for the induction of insulin secretion via an increased mitochondrial response and ATP production is plausible. If a general increase of fatty acids-for instance in response to fasting when blood glucose levels are low and free fatty acids levels increase-would result in an elevated beta-oxidation and thereby in an increased ATP production, its induction of insulin release should decrease the blood glucose levels to a dramatically low level. Therefore, the inferred role of malonyl-CoA is that it is essential for maintaining glucose homeostasis because it derives from glucose metabolism [27] and inhibits fatty acid oxidation by allosteric inhibition of CPT1. This increases the availability of LC-CoA for lipid signaling to cellular processes that are needed to increase exocytosis of insulin granuels. Figure 2. Impact of palmitate on mitochondrial respiration. (a) Changes in oxygen consumption rate (OCR) of isolated islets in response to indicated stimuli; (b) Correlation of insulin secretion and ATP synthase-dependent OCR in response to 2.8 mM glucose (low Glc) or 20 mM glucose (high Glc) without or with 0.5 mM palmitate (+PA). In parallel measurements of insulin secretion and oxygen consumption were performed at different conditions with isolated islets. Islets were treated with Krebs-Ringer buffer for 1 h. Insulin was measured by an ELISA and normalized to islet DNA content. ATP synthase-dependent OCR was calculated by subtracting mitochondrial proton leak from the respiration under the indicated conditions; (c) Calculation of islet coupling efficiency in response to 2.8 mM or 20 mM glucose without and with 0.5 mM palmitate (caluculated as: CE = 1 − (OCR proton leak /OCR mito.respiration )); (d) Proton leak of isolated islets after treatment with indicated substrates (calculated as: OCR proton leak = OCR mito.respiration − OCR ATPsynthase ). Values in (a), (c) and (d) represent means of 5-10 independent experiments with islets obtained from different mice ± SEM; plotted points in (b) represent secreted insulin of one experimental animal and the corresponding ATP synthase-dependent OCR value at a particular condition. * p < 0.05; # p < 0.01.
Mice
Studies of islet-size dependent function were performed with islets from 24 week old male DBA/2J. All following analyses were carried out with isolated islets form 22-26 week old male B6. V-Lep ob mice. Animals were housed in air conditioned rooms (temperature 20 ± 2 °C) with a 12 h light-dark cycle in accordance with the National Institutes of Health guidelines for the care and use of laboratory animals. All experiments were approved by the ethics committee of the State Agency of Environment, Health and Consumer Protection (State of Brandenburg, Germany).
Preparation of BSA-Conjugated Palmitate
Palmitate has a low solubility in aqueous solutions. To produce an aqueous-soluble and absorbable fatty acid complex we conjugated palmitate (Sigma, St. Louis, MO, USA) with bovine serum albumin (BSA, essentially fatty acid free; Sigma). Preparation of BSA-conjugated palmitate was performed as described previously [28]. Briefly, sodium palmitate solution preheated to 75 °C was slowly dissolved in a 48 °C preheated BSA solution. Palmitate was complexed in a 6 to 1 molar ratio with BSA. To exclude possible side effects of BSA we compared islets of Langerhans treated with BSA-conjugated palmitate and BSA alone ( Figure S1).
Detection of Insulin Secretion
Isolated islets were kept in Krebs-Ringer buffer containing 2.8 mM glucose for 1 h and then incubated at 2.8 mM or 20 mM glucose without and with 0.5 mM palmitate (PA) for 1 h. Insulin content of supernatant was measured by Mouse Insulin Ultrasensitive ELISA (DRG Instruments, Germany) and normalized to the DNA content of islets.
Measurement of Oxygen Consumption Rate
Oxygen consumption rates (OCR) of isolated islets were determined with an XF24-3 extracellular flux analyzer (Seahorse Bioscience, Billerica, MA, USA). The analyzer uses fluorescence sensors for detection of rate changes of dissolved O 2 in the surrounding media of isolated islets. To determine the flux of oxygen a chamber of ~7 µL was created mechanically within the multi-well plate during the measurement [29]. After each measurement a period of mixing and waiting followed to re-equilibrate the islets with the whole buffer volume. Four different ports (A, B, C and D) nearby the fluorescence sensor enabled a controlled injection of substances to the media. To measure cellular and mitochondrial oxygen consumption rate 70 islets were placed in 500 µL Krebs-Ringer buffer (111 mM NaCl, 4.7 mM KCl, 2 mM MgSO 4 , 1.2 mM Na 2 HPO 4 , 0.5 mM carnitine, pH 7.4) containing 2.8 mM glucose in an islet capture microplate. Islets were incubated at 37 °C without CO 2 for 1 h before starting the experiment. Measurement of OCR was performed according to manufacture's instructions, with duration for mixing, waiting and measurement of 1, 2 and 3 min, respectively. The treatment time with each of the compounds was adjusted to reach steady state conditions (plateau phase of OCR). After measurement of OCR at basal conditions (2.8 mM glucose) different substrates were injected in separated experiments: (a) no further substrate application, to measure at low glucose conditions (low Glc); (b) 0.5 mM palmitate (low Glc + PA); (c) 20 mM glucose (high Glc); (d) 20 mM glucose followed by 0.5 mM palmitate (high Glc + PA). Each condition was followed by an application of oligomycin (6 µM) as penultimate step before a cocktail of rotenone (5 µM) and antimycin A (4 µM) was injected. Using different inhibitors enabled analysis of islet respiration. Oligomycin inhibits F O subunit of ATP synthase and therewith it is suitable to calculate mitochondrial ATP synthase-dependent OCR. Additionally, rotenone and antimycin A block mitochondrial electron flux of complex I and III of the respiratory chain whereby determination of non-mitochondrial respiration is possible. Mitochondrial respiration (OCR mito.respiration ) represents the sum of oxygen consumption due to the proton leak (green colored segment Figure 2a) and ATP synthesis-linked respiration (pink colored segment Figure 2a), whereas non-mitochondrial respiration (OCR Rotenone/Antimycin A ) was subtracted. ATP synthase-dependent OCR was calculated as followed: OCR ATPsynthase = (OCR condition − OCR Roteneone/Antimydin A ) − (OCR Oligomycin − OCR Rotenone/Antimycin A ). To evaluate proton leak, which describes flux of protons through the mitochondrial membrane thereby producing heat but no ATP, we used the equation: OCR proton leak = OCR mito.respiration − OCR ATPsynthase . Moreover, we calculated the efficiency of mitochondrial energy conversion, termed coupling efficiency (CE) as the proportion of respiration converted to ATP, considering the proportion that is lost as heat by proton leak (CE = 1 − (OCR proton leak /OCR mito.respiration )) [26]. For normalization DNA content of islets was measured with Quant-iT PicoGreen (Invitrogen, Darmstadt, Germany).
Statistical Analysis
Data are presented as mean ± SEM. Statistically significant differences between conditions were defined as p < 0.05 by using two-tailed Student's t test and one way ANOVA.
Conclusions
In conclusion, glucose-induced insulin secretion of isolated islets is mediated by an increase in mitochondrial respiration and coupling efficiency, known as the triggering pathway. In contrast, a further augmentation of insulin secretion through palmitate at high glucose concentrations is only mediated to a minor extent via mitochondrial oxygen consumption indicating that palmitate induces insulin secretion nearly exclusively through the amplifying pathway.
==
Domain: Biology Medicine
|
[{"added": "2016-03-22T00:56:01.885Z", "created": "2013-09-01T00:00:00.000", "id": "5069071", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/1422-0067/14/9/18989/pdf", "pdf_hash": "f9b9fa1007b1e21b33fbe9f4f745eccfe1a1015e", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39262", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "f9b9fa1007b1e21b33fbe9f4f745eccfe1a1015e", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Contribution of the PALB2 c.2323C>T [p. Q775X] Founder mutation in well-defined breast and/or ovarian cancer families and unselected ovarian cancer cases of French Canadian descent
Background The PALB2 c.2323C>T [p. Q775X] mutation has been reported in at least three breast cancer families and breast cancer cases of French Canadian descent and this has been attributed to common ancestors. The number of mutation-positive cases reported varied based on criteria of ascertainment of index cases tested. Although inherited PALB2 mutations are associated with increased risks of developing breast cancer, risk to ovarian cancer has not been fully explored in this demographically unique population. Methods We screened the PALB2 p. Q775X variant in 71 families with at least three cases of breast cancer (n=48) or breast and ovarian cancers (n=23) that have previously been found negative for at least the most common BRCA1 and BRCA2 mutations reported in the French Canadian population and in 491 women of French Canadian descent who had invasive ovarian cancer and/or low malignant potential tumors of the major histopathological subtypes. Results We identified a PALB2 p. Q775X carrier in a breast cancer family, who had invasive ductal breast carcinomas at 39 and 42 years of age. We also identified a PALB2 p. Q775X carrier who had papillary serous ovarian cystadenocarcinoma at age 58 among the 238 serous subtype ovarian cancer cases investigated, who also had breast cancer at age 52. Conclusion Our findings, taken together with previous reports, support adding PALB2 c.2323C>T p. Q775X to the list of cancer susceptibility genes for which founder mutations have been identified in the French Canadian population.
Background
Carriers of PALB2 mutations in a heterozygous state have been associated with increasing the risk of developing breast cancer [1][2][3][4][5][6][7][8][9][10]. PALB2 is a partner and localizer of the BRCA2 breast-ovarian cancer susceptibility protein to DNA damage sites [9,11]. Penetrance estimation for conferring breast cancer risk has been hampered by the paucity of cases, although estimates of 2-to 6-fold increased risk to breast cancer have been suggested [12,13], thus classifying PALB2 as a moderate breast cancer risk allele [9,[12][13][14]. Germline mutations in PALB2 have also been identified in familial pancreatic cancer [15,16]. PALB2 is comprised of 13 exons spanning a 38 kb region on chromosome 16p12.1 and mutation screening is complicated by the diversity of variants (including missense mutations) identified in cancer cases. The PALB2 c.2323C>T mutation, which results in the introduction of a stop codon at amino acid position 775 (p. Q775X), has been reported in at least three French Canadian breast cancer families [5], and along with other protein truncating PALB2 mutations found in breast cancer cases, is strongly suspected to be deleterious [17]. The French Canadian population of Quebec exhibits an unique genetic demography [18][19][20]. About 40% of French Canadian cancer families with at least three cases of breast and/ ovarian cancer carry a pathogenic BRCA1 or BRCA2 mutation [20][21][22][23][24][25]. Although 15 different mutations in these genes have been reported in French Canadian cancer families, six specific mutations in BRCA1 and BRCA2 have been shown to account for a significant majority of mutation-positive families [20][21][22][23][24][25][26]. This has been attributed to a shared ancestry of mutation carriers due to common founders of the French Canadian population of Quebec [25][26][27][28].
The number of PALB2 p. Q775X mutation-positive cases that have been reported thus far in studies involving the French Canadian population vary according to criteria and catchment area of ascertainment of index breast cases tested [5,29]. To further assess the contribution of PALB2 p. Q775X mutation in the French Canadian population, we report the results of screening this variant in 71 well defined cancer families with at least three confirmed cases of breast and/ovarian cancer found negative for the most common BRCA1 and BRCA2 mutation reported in this population. We report the cancer phenotype of a new p. Q775X mutation-positive family. We also report the screening 385 invasive ovarian cancer cases and 106 low malignancy potential ovarian tumors not selected for family history of cancer that were ascertained from the French Canadian population, and describe the cancer history of the p. Q775X cases identified in this screen. We describe our findings in the context of previous studies describing mutation screens of PALB2 in individuals of French Canadian descent.
Subjects and cancer families
The study subjects fall within two defined groups. The first group contains index cases from 71 independently ascertained families ( Table 1). The index cases tested for mutations were recruited to the study through the hereditary cancer clinics in Montreal as part of research studies assessing the contribution of BRCA1 and BRCA2 in breast and/or ovarian cancer families as described previously [21,25]. They have a family history of breast cancer (n= 48) or breast and ovarian cancer (n=23) according to the following criteria: in addition to the index case affected with breast cancer at less than 66 years of age, the families contained at least two other confirmed cases of invasive breast and/or epithelial ovarian cancer in the same familial branch. The affected index cases from 26 breast cancer families (HBC) and 14 breastovarian cancer (HBOC) families were previously screened and found negative for BRCA1 and BRCA2 sequence variants by commercial DNA sequencing (Myriad Genetics, Myriad Genetics Laboratories, Salt Lake City, UT, USA). The index affected cases from the remaining 22 HBC and 9 HBOC families were found negative for 20 BRCA1 and BRCA2 mutations reported in French Canadian cancer families of Quebec which include the following most common BRCA1 (c.4327C>T (R1443X), c.2834_2836delG-TAinsC) and BRCA2 (c.8537_8538delAG, c.5857G>T (E1953X), c.3167_3171delAAAAG) mutations reported in this population, as described previously [22,23,25]. All index cases in this study self-reported grandparental French Canadian ancestry. The second group contained 385 females with epithelial ovarian carcinomas and 106 low malignant potential tumors (Table 2), who were recruited to Banque de tissus et de données of the RRCancer of the Fonds recherché Québec-santé tumor bank between April 1991 and October 2007. At least 88% of all women with malignant serous, endometrioid or undifferentiated malignant ovarian cancer cases from RRCancer Tumor self reported French Canadian ancestry (unpublished data). All women with serous LMP (Low Malignant Potential) tumors self-reported FC ancestry [30]. None of these subjects were selected for family history of cancer. Histopathology according to criteria established by the International Federation of Gynecology and Obstetrics (FIGO), age at diagnosis and personal history of cancer were provided for each case. Written consent to participate was obtained and the study protocols approved by the ethics review boards of the University of Montreal Hospital Center, McGill University Health Centre and Jewish General Hospital.
PALB2 mutation analysis
Mutation analysis was performed on DNA extracted from peripheral blood leukocytes or from fresh frozen tumor The numbers in the brackets refer to number of index cases where the complete coding region of PALB2 genes was sequenced. Abbreviations: breast cancer (BC), hereditary breast cancer (HBC) hereditary breast and ovarian cancer (HBOC), and ovarian cancer (OC). tissue. A sequence analysis of protein coding exons of PALB2 for the index cases of 14 families was performed as described previously [5,10,17] (Table 1). The targeted analysis for PALB2 c.2323C>T (p. Q775X) variant was performed using an allelic specific assay as described [5]. The variant-positive cases were confirmed by DNA sequencing using 3730XL DNA analyzer system platform from Applied Biosystems (Carlbad, CA, USA) at the McGill University and Genome Quebec Innovation Centre (Montreal, PQ, CDN). Sequences were compared with PALB2 NCBI Reference Sequence NM_024675 as described in GenBank (www.ncbi.nlm.nih.gov). For the Loss of heterozygosity (LOH) analysis, fresh frozen ovarian tumor tissue from the PALB2 p. Q775X mutation-positive case was macrodissected and DNA was extracted from the collected cells using the QIAamp DNA Mini Kit (Qiagen). The PCR was carried out in a volume of 50 μL, as previously described [5,10]. Sequence data were analyzed using the Lasergene SeqMan Pro sequence analysis software by DNASTAR, Inc. (Madison, WI, USA) and Chromas 2.31 from Technelysium Pty Ltd. (Helensvale, Australia) and compared to the sequences from lymphocyte DNA from PALB2 p. Q775X mutation-positive and mutation-negative cases.
The PALB2 p. Q775X variant is also annotated as p. Q775* according to a recently proposed nomenclature alteration for nonsense changes by the Human Genome Variation Society (www.hgvs.org). However for historical purposes the p. Q775X designation is maintained in this report.
Results
PALB2 c.2323C>T [p. Q775X] carriers in breast and/or ovarian cancer families One PALB2 p. Q775X positive case was identified among the cancer families not previously investigated for PALB2 mutations. The index carrier case was identified among the total of 48 (2.1%) HBC families or 71 (1.4%) HBC and HBOC families that share a phenotype defined by at least three or more confirmed cases of breast and/or ovarian cancer in the same familial branch (Table 1). These families were previously found negative for BRCA1 and BRCA2 mutations or the most common pathogenic mutations in these genes found in the French Canadian population.
The PALB2 carrier had bilateral invasive ductal carcinomas of the breast at ages 39 and 42 and is part of the breast cancer family F1469 (Figure 1). Although breast cancer was reported in both paternal and maternal branches of her family (Figure 1), only the aunt and cousin from the paternal branch of the family were confirmed to have had breast cancer. Her paternal aunt also had bilateral invasive ductal carcinoma at ages 41 and 42, as well as atypical stomach carcinoma that was identified at age 42 but not further explored due to death soon thereafter. The carrier's paternal cousin had an invasive breast cancer of mixed ductal and lobular histopathology at 52 and was still living at the time of pedigree analysis. Notable in this pedigree is that lack of ovarian cancer cases typified by BRCA1 or BRCA2 mutation carrier families. Her father had esophageal cancer and cancers of other sites reported (some confirmed) for both branches of her family. To our knowledge no other cases are available for genetic testing and thus transmission of the mutation is uncertain in this family. The family structure and associated phenotypes does not appear to overlap previously described PALB2 p. Q775X positive families [5].
PALB2 c.2323C>T [p. Q775X] carriers in ovarian cancer cases One PALB2 p. Q775X positive case was identified among the 491 women with ovarian cancer or low malignant potential tumors. The carrier was diagnosed with a papillary serous cystadenocarcinoma at age 58. The carrier was identified among the 385 (0.3%) invasive ovarian carcinomas of all histopathological subtypes and among 238 (0.4%) invasive serous ovarian carcinomas ( Table 2).
There were 21 cases that also had a prior personal history of breast cancer and the PALB2 p. Q775X carrier was in the group of 10 invasive serous ovarian carcinoma cases with this history ( Table 2). The carrier had a breast cancer at age 52 years of undisclosed histological type. Genetic analysis of genomic DNA from ovarian cancer specimens did not indicate LOH of the PALB2 locus (Figure 2) or identify a second mutation in the other coding exons of PALB2.
Discussion
Our results are consistent with previously established frequency of PALB2 c.2323C>T [p. Q775X] carriers in breast cancer families of French Canadian descent (Table 3). Initial reports of comprehensive screening of all of the protein encoding exons of PALB2, identified no variants in 38 breast cancer families, where 22 families had a prior probability of greater than 10% of harboring a BRCA1 or BRCA2 mutation [10]. The same group reported one of 50 (2%) breast cancer families, and two of 356 (0.6%) cases of early age (< 50 years) breast cancer with this mutation [5]. Pedigree analysis of the index PALB2 p. Q775X positive cases from these three families indicated that they are not immediately related to each other and haplotype analysis was consistent with this being a founder mutation [5]. Four PALB2 p. Q775X positive cases were also identified in a subsequent study involving 564 (0.7%) breast cancer cases not selected for family history of cancer, which also showed that 6% of cases harbored common BRCA1, BRCA2 or CHEK2 mutations [29]. This latter study also included the 356 early age breast cancer cases reported in a previous study, and one of these cases was found related to a member of previously reported mutation-positive family (P28031) and two other cases were the same carriers identified in a same previous study (P31030 and P26007) [5] (Table 3). In an independent study involving the investigation of a new BRCA2 variant, c.9004 G>A (E3002K), a family (F1573) harboring both a BRCA2 variant and the PALB2 p. Q775X mutation was reported [22]. However, pedigree inspection revealed that family F1573 was related to one of three PALB2 p. Q775X (P28031) families described in the initial report of this variant in the French Canadian breast cancer families [5]. The family history of PALB2 p. Q775X carrier P36470 also identified in a screen of 564 breast cancer cases not selected for family history [29] appears not to be related to the PALB2 p. Q775X carrier families described based on pedigree inspection, including family F1469 reported in this study. In summary, six PALB2 p. Q775X breast cancer carriers which include five occurring in apparently unrelated cancer families have thus far been identified in screening breast cancer cases or breast cancer families. In contrast, no PALB2 variants were reported in two other independent studies involving 99 [31] and 21 [2] independently ascertained breast cancer families of French Canadian descent. Genealogy and genetic studies have reported variability of founder effects in various regions of Quebec [32], suggesting that demography may also be a factor in the paucity of PALB2 p. Q775X carriers in some studies of French Canadian cancer families. The majority of our cases were ascertained in Montreal [20], whereas independent groups have ascertained their families from the Quebec City region [31]. This possibility could also account for the lack of PALB2 p. Q775X carriers found in a screen of 6,440 newborns of French Canadian descent as the majority of these newborns were from the Quebec City region [5]. The young ages of breast cancer diagnoses and number of breast cancer cases per family in PALB2 p. Q775X carrier families suggest that carriers of this mutation are at high risk for breast cancer (Table 3), as has been posited with some PALB2 mutation carrier families [5,10,17]. Our findings here support this notion, as the PALB2 p. Q775X carrier identified had a bilateral case of breast cancer diagnosed before age 45 years and a strong family history of breast cancer (Figure 1).
It is interesting that the PALB2 p. Q775X carrier found among the ovarian cancer cases examined in this study had a prior history of breast cancer (Table 3). Notable is that her personal history of cancer does not match any of the cases that appear in the pedigrees of PALB2 p. Q775X positive French Canadian cancer families described thus far (including the new carrier family identified in this study). The role of PALB2 in ovarian cancer is uncertain, as there have been few documented ovarian carcinoma cases harboring germline mutations in this gene. Two PALB2 mutation carriers were identified in a study of 339 unrelated ovarian cancer cases of Polish descent [3]. The carriers had high grade carcinomas of different histopathological types: serous (case diagnosed at 61 years) and endometrioid (case diagnosed at 54 years) subtypes, where the latter carrier also harbored a BRCA2 mutation [3]. Two (0.6%) PALB2 mutation carriers were reported in a study of 360 ovarian cancer cases that were also screened for BRCA1, BRCA2 and other recently described cancer susceptibility genes [33]. Neither of these two high-grade serous carcinoma PALB2 mutation-carriers (diagnosed at ages 51 and 58) had a personal history of breast cancer, although the ovarian cancer case diagnosed at age 58 years had a family history of breast and/or ovarian cancer [33]. A low frequency of PALB2 carriers (0.4%) was also recently reported in an investigation of 253 ovarian cancer cases from the Volga-Ural region of Russia [34], with the Table 2) Not known This study only carrier identified in this study having a bilateral (moderate grade) serous ovarian carcinoma at age 46 and a prior history of melanoma. The low frequency of PALB2 mutation carriers identified thus far may argue a minor role for this gene in conferring ovarian cancer risk compared with higher frequency of mutation carriers observed in breast cancer cases and breast cancer families. This is consistent with recent findings estimating that PALB2 heterozygotes were 1.3-fold more likely to have a relative with ovarian cancer in the context of HBOC family history [2]. Our genetic analyses of the carrier ovarian cancer specimen harboring the PALB2 p. Q775X mutation did not exhibit evidence of LOH of the PALB2 locus. This could also be consistent with sufficient contamination of normal stromal DNA such that it would obscure an imbalance of alleles. It has been suggested that PALB2 contributes to carcinogenesis through haploinsufficiency and/or a dominant negative effect given the paucity of LOH observed in the majority of breast cancer cases from PALB2 carriers [3,4,6,10], with the exception of one study where a high frequency of LOH was seen [2]. LOH was observed for both ovarian cancer cases identified in one study [33]. Promoter methylation silencing has also been reported in four of 53 sporadic ovarian cancer cases [35]. The significance of these findings is unknown and warrants further investigation to elucidate the role of PALB2 in both breast and ovarian carcinogenesis.
Conclusion
The PALB2 c.2323C>T [p. Q775X] mutation confers increased risk for breast cancer in the French Canadian population of Quebec. The contribution of PALB2 c.2323C>T [p. Q775X] to the causation of breast cancer in French-Canadians appears to be lesser than that attributable to the most common founder alleles in BRCA1 and BRCA2, but the young age at diagnoses and associated familial history of breast cancer suggest that this variant should be added to the panel of deleterious mutations screened for assessing breast cancer risk in this unique population. Indeed during the preparation of this manuscript another PALB2 carrier harboring the p. Q775X variant was identified in the Hereditary Cancer Clinics affiliated with McGill University Health Centre. The carrier had bilateral breast cancer at ages 34 and 42 years and a strong family history of breast cancer further supporting the notion that PALB2 p. Q775X carriers are at increased risk for breast cancer.
==
Domain: Biology Medicine
|
[{"added": "2016-05-12T22:15:10.714Z", "created": "2013-01-09T00:00:00.000", "id": "10010461", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://bmcmedgenet.biomedcentral.com/track/pdf/10.1186/1471-2350-14-5", "pdf_hash": "cf984fbb3db949f4b1998db1d1325e62edf98aa7", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:30430", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "cf984fbb3db949f4b1998db1d1325e62edf98aa7", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Adaptation of a RAS pathway activation signature from FF to FFPE tissues in colorectal cancer
Background The KRAS gene is mutated in about 40 % of colorectal cancer (CRC) cases, which has been clinically validated as a predictive mutational marker of intrinsic resistance to anti-EGFR inhibitor (EGFRi) therapy. Since nearly 60 % of patients with a wild type KRAS fail to respond to EGFRi combination therapies, there is a need to develop more reliable molecular signatures to better predict response. Here we address the challenge of adapting a gene expression signature predictive of RAS pathway activation, created using fresh frozen (FF) tissues, for use with more widely available formalin fixed paraffin-embedded (FFPE) tissues. Methods In this study, we evaluated the translation of an 18-gene RAS pathway signature score from FF to FFPE in 54 CRC cases, using a head-to-head comparison of five technology platforms. FFPE-based technologies included the Affymetrix GeneChip (Affy), NanoString nCounter™ (NanoS), Illumina whole genome RNASeq (RNA-Acc), Illumina targeted RNASeq (t-RNA), and Illumina stranded Total RNA-rRNA-depletion (rRNA). Results Using Affy_FF as the “gold” standard, initial analysis of the 18-gene RAS scores on all 54 samples shows varying pairwise Spearman correlations, with (1) Affy_FFPE (r = 0.233, p = 0.090); (2) NanoS_FFPE (r = 0.608, p < 0.0001); (3) RNA-Acc_FFPE (r = 0.175, p = 0.21); (4) t-RNA_FFPE (r = −0.237, p = 0.085); (5) and t-RNA (r = −0.012, p = 0.93). These results suggest that only NanoString has successful FF to FFPE translation. The subsequent removal of identified “problematic” samples (n = 15) and genes (n = 2) further improves the correlations of Affy_FF with three of the five technologies: Affy_FFPE (r = 0.672, p < 0.0001); NanoS_FFPE (r = 0.738, p < 0.0001); and RNA-Acc_FFPE (r = 0.483, p = 0.002). Conclusions Of the five technology platforms tested, NanoString technology provides a more faithful translation of the RAS pathway gene expression signature from FF to FFPE than the Affymetrix GeneChip and multiple RNASeq technologies. Moreover, NanoString was the most forgiving technology in the analysis of samples with presumably poor RNA quality. Using this approach, the RAS signature score may now be reasonably applied to FFPE clinical samples. Electronic supplementary material The online version of this article (doi:10.1186/s12920-016-0225-2) contains supplementary material, which is available to authorized users.
Background
Colorectal cancer (CRC) is the third most common cancer in men and women [1]. Nearly one-third of the patients will eventually die of the disease. Hyperactivation of the RAS signaling pathway is a driver of many cancers, including CRC [2,3]. Activating mutations in the K-ras protooncogene (KRAS) are found in approximately 40 % of colorectal tumors [4]. Thus, the RAS pathway activation has become a major focus of drug targeting efforts, including prediction of response to targeted therapies [5][6][7][8]. For example, the epidermal growth factor receptor (EGFR) is a major therapeutic target in metastatic colorectal cancer [7][8][9][10][11][12][13][14]. The fact that nearly 60 % of patients with a wild type KRAS fail to respond to combination therapies involving EGFRi [5,15], however, strongly suggests that there are additional genes, beyond KRAS, that contribute to RAS pathway activation. It has been recently reported that mutations in BRAF and NRAS that also activate the RAS pathway may account for EGFRi therapy resistance in some of the wild-type KRAS CRCs [7,10,12,14].
A number of gene expression signatures have been developed using multiple types of cancer cell lines and human fresh frozen (FF) samples to predict RAS pathway dependence in association with drug response [2,3]. For example, a 147-gene RAS pathway signature has been reported to be superior to KRAS mutation status alone for the prediction of dependence on RAS signaling, and it could predict response to PI3K and RAS pathway inhibitors in lung and breast tumors [3]. Low RAS pathway signature score was associated with a higher cetuximab response rates in a retrospective analysis of metastatic CRC [3]. Another RAS pathway signature (18 genes) was developed from multiple types of cancer cell lines and human tumors, including CRC, to specifically assess MEK functional output and activation of the RAS/RAF/MEK/ERK pathway [2]. While measuring mutations in individual genes such as KRAS and NRAS can predict EGFRi response, their level of accuracy is low with up to 60 % of patients still notresponding [15]. For this reason, multi-gene expression signatures hold promise in being able to more robustly assess pathway activation than single gene mutations, and thus there is an interest in translating them for use with FFPE clinical samples.
One of the challenges in using these gene expression signature scores is that many have been developed using fresh-frozen (FF) tissues on the Affymetrix GeneChip (microarray) platform. In order for these signature scores to be clinically useful, they need to be adapted to the more commonly available archival formalin-fixed paraffinembedded (FFPE) tissues [16,17]. However, microarrays that can assess thousands of transcripts are not only expensive but also lack reproducibility, especially when evaluating FFPE samples having low RNA quality [18,19].
Determinants of RNA quality from FFPE samples have been reported to include storage time and conditions, fixation time and specimen size [20]. RT-qPCR and NanoString technologies have been reported to be useful for gene expression quantification in FFPE tissues [17,[21][22][23]. However, the recently developed, probebased NanoString method was shown to be superior to the RT-qPCR approach in archived FFPE samples [22].
To date, the RAS pathway signatures developed in FF samples for prediction of drug response have not been validated in CRC using FFPE samples. Thus, in this study, we elected to evaluate the translation of an 18-gene RAS signature score [2] from FF to FFPE in 54 selected CRC cases in a head-to-head comparison of five technology platforms: Affymetrix GeneChip (Affy), NanoString nCounter™ (NanoS), whole genome RNASeq (Illumina RNA-Access (RNA-Acc), targeted RNASeq (t-RNA), and Illumina Total stranded RNA-rRNA-depletion (rRNA).
Tissue sample selection
Fifty-four (54) FFPE evaluable tumor specimens were selected from a larger multi-center cohort of 468 wellcharacterized colorectal adenocarcinoma patients whose tissues were obtained between October 2006 and September 2010 [24]. In all cases, tissue and clinical data were collected on patients with the University of South Florida institutional review board approval [25]. All tumors were collected from curative survival resections and snap frozen in liquid nitrogen within 15-20 min of extirpation. The sample cohort was composed of tumor samples that were available as matched fresh-frozen (FF) and formalin-fixed paraffin-embedded (FFPE) pairs. As shown in Additional file 1, the 54 samples had mutant KRAS (25/54 or 46 %) and BRAF (2/54 or 4 %), but no NRAS mutations.
The Affymetrix GeneChip, NanoString, whole genome RNASeq, and targeted RNASeq assays on the 54 FFPE samples were performed at LabCorp, Inc., Seattle, USA. Whole genome RNASeq was further comprised of two library preparation methods: Illumina RNA-Access (RNA-Acc) and Illumina Total stranded RNA-rRNAdepletion (rRNA), which were analyzed as separate datasets. Targeted RNA sequencing data (t-RNA) was based on the RAS 18-gene signature [2].
The flowchart (see Fig. 1) below shows the steps followed in the pre-processing and analysis of the data. The statistical methods used include [1] the Robust Microarray Average (RMA) method [26] for the normalization of Affy_FF and Affy_FFPE samples; [2] principal component analysis (PCA) [19,27] to identify "bad" samples from the Affy_FFPE data; [3] correlation analyses among the datasets; and [4] the nearest shrunken centroids algorithm for predicting the mutation type of a sample.
Data pre-processing
Both the 54 Affy_FF and matching Affy_FFPE samples were normalized using the RMA method [26]. For the NanoString data, we used the reference (housekeeping) gene normalization method, as described in the nCounter® Expression Data Analysis Guide (available at [URL]). The 11 housekeeping genes were BIRC6, EMC8, HADHA, MAEA, MRPL18, ORMDL1, PSMD11, RBM4, STX6, TRIM39, and UBE2K. The geometric mean of these reference genes was obtained for each sample (lane) and the average of these means across all samples calculated. The normalization factor for each sample was the overall mean divided by the geometric mean. We multiplied this factor by the mRNA transcript count for each of the 18 RAS genes in the sample. For the targeted RNA data, we used median normalization. For that platform, we obtained the median for each of the samples and subtracted this number from each of the gene counts for the sample. Notably, global normalization using median-centering is commonly used to correct for sample-specific bias (due to experimental artefacts) and render the gene expression levels comparable in differential gene expression analysis in microarrays [28]. With the advent of RNASeq technology, the method has been adopted to render counts from different samples, which may have been sequenced to different depths, comparable [29]. Thus, gene expression values could be positive or negative numbers relative to a reference (e.g. median). For the whole transcriptome RNASeq (RNA-Acc and rRNA) platforms, the data was first processed by STAR aligner [30] and cufflinks [31], then the resulting FPKM was log2-transformed and z-score-normalized.
Probe filtration
After normalization, filtration of probes was performed for both the Affy_FF and Affy_FFPE data. Probes were retained if they had at least 1.5-fold change in either direction of the median expression level in at least 20 % of the samples and if they had at most 50 % missing values across the samples. The entire probe filtration process was implemented by the BRB-ArrayTools software [32]. The NanoS_FFPE, RNA-Acc_FFPE, t-RNA and rRNA_FFPE Fig. 1 Flow-chart of the procedure followed in the pre-processing and analysis of the data. Six datasets (1 FF and 5 FFPE, each with 54 samples and 18 genes) underwent quality control procedures before analysis. Thirty-nine [39] "good" samples and 16 "good" genes were retained. Correlation analyses were performed using mean scores from the sample pairs. The predictive ability of the 16-gene set was validated using the Affymetrix FF, Affymetrix FFPE and NanoString gene expression data, by the PAM method datasets did not have probe-level data and so were not subjected to the probe filtration process.
Calculation of RAS pathway activation scores before probe filtration
The next step was to calculate the RAS pathway activation scores from the normalized 54 Affy_FF and Affy_FFPE samples. For genes with multiple probes/probesets in the dataset, probes/probesets from the same gene were averaged to yield one value of the expression level for each gene. The mean of these expression levels across the 18 RAS genes was calculated for each sample to yield the sample mean score. The sample mean scores for the NanoS_FFPE, RNA-Acc_FFPE, t-RNA_FFPE and rRNA_FFPE were obtained by averaging across the 18 RAS genes.
Calculation of RAS pathway activation scores after probe filtration
The probe filtration resulted in a reduction of probes from 60,607 to 23,765. Some genes that were represented by only one probe in the dataset were filtered out in this filtration process. For the remaining genes with multiple probes/probesets, the probes/probesets were reduced to one per gene by selecting the probe with the highest mean signal strength across the samples. The mean expression levels across the remaining 16 RAS genes were calculated for each sample to yield the sample mean score. The sample mean scores for the NanoS_FFPE, RNA-Acc_FFPE, t-RNA_FFPE and rRNA_FFPE were obtained by averaging across the 16 RAS genes.
Statistical analysis
The FF -FFPE sample pairs of mean scores were used in the correlation analyses, using SAS software version 9.4 (SAS Institute, Cary, NC, USA). There were 15 possible combinations of the sample pairs, among the six datasets (Affy_FF, Affy_FFPE, NanoS_FFPE, RNA-Acc_FFPE, t-RNA-FFPE, and rRNA_FFPE), yielding 15 pairwise Spearman correlations. We also assessed the effect of removing "bad" samples and probes on the Spearman correlations across the five platforms. To identify the "bad" samples, we performed a principal component analysis (PCA) of the 54 Affy_FFPE samples, with the entire set of 60,607 probes, to generate the first two principal components (PC1 and PC2), using the SAS software version 9.4. The PC1 and PC2 scores were identified as the eigenvectors of the covariance matrix of the 54 Affy_FFPE samples that accounted for the highest and the second-highest variation in the data, respectively. A scatterplot of PC2 vs PC1 was used to show the location of the possibly "bad" samples.
Samples were classified as either KRAS/BRAF mutant or KRAS/BRAF wild-type (WT). The nearest shrunken centroids algorithm [33] was employed in predicting the mutation type of a sample, based on the gene expression profiles of the 16 genes from the 18-gene RAS signature. This algorithm was implemented by the Prediction Analysis of Microarrays (PAM) tool in BRB-ArrayTools software [32]. The algorithm builds several linear models (classifiers) containing up to 16 genes and selects the model with the minimal prediction error. The prediction error of the models are estimated using leave-one-out cross-validation (LOOCV) as described in [34]. For each leave-one-out training set, the entire model building process was repeated, including the gene selection process. The proportion of times when classifiers incorrectly predicted the class (miss-classification rate) of the excluded samples was recorded for the entire training set of samples.
Identification and removal of 15 "poor" quality samples improves the FF -FFPE correlations Sfakianos et al. used the PCA procedure to detect outliers and showed that the outliers were associated with poor quality samples [19]. More recently, Guinney et al. performed quality control analysis for outlier detection using PCA [11]. We adopted this procedure to identify samples that could possibly have "poor" RNA quality. A scatterplot (see Fig. 2) of the first and second principal component (PC1 and PC2) scores identified fifteen samples with low PC1 scores (hereby less than -0.10) that were considered to be "outliers", or samples likely to have "poor" RNA quality. Notably, as compared to other 39 "good" samples (see Table 1B), most of 15 samples identified also had low standard deviations (signalto-noise ratios) across probes (data not shown). Furthermore, these 15 samples all had below average Affy_FFPE mean scores in contrast to their wide-spread pattern for Affy-FF scores (see Fig. 3a). Thus, these data support the notion that low PC1 scores were likely associated with "poor" RNA quality, However, these 15 "bad" samples did not stand out among the NanoS_FFPE mean scores (see Fig. 3b), suggesting that NanoString technology may be more forgiving of poor RNA quality inherent to these samples. The 15 samples identified with potential "poor" quality were removed, leaving 39 samples available for subsequent analyses.
As shown in between Affy_FF and Affy_FFPE changed from being insignificant (r = 0.233, p = 0.090) to significant (r = 0.556, p = 0.0002). Notably, while the 39 "good" samples had mutant KRAS (19/39) and BRAF (1/39), the 15 "bad" samples had mutant KRAS (6/15) and BRAF (1/15). No signification association was seen between sample RNA quality and KRAS/BRAF genotypes. For example, the Fisher's exact test of association of BRAF mutation status and sample RNA quality was insignificant (p = 0.5711). This suggests that the "badness" of the 15 samples is likely not due to a biological reason (e.g. BRAF V600E enrichment), but rather to a "technical aspect of the sample preparation".
Reduction of probes and associated genes in attempt to improve further the FF-FFPE correlations
The 18 RAS signature genes were represented by 51 probes and 50 probesets, in the Affy_FF and Affy_FFPE datasets, respectively, with 48 probes in common to both. Probe selection was performed to exclude probes that were not sufficiently differentially-expressed across the 39 samples. The selection was performed from the entire set of 60,607 probes on 39 Affy_FFPE samples. We first filtered out those probes with less than 1.5-fold change in either direction of the probe's median value and then filtered out those with at least 1.5 fold change but in less than 20 % of the samples, resulting in 23,765 probes in 10,031 genes. We then reduced the number of probes to one per gene, by selecting the probes with the highest mean expression values. Due to the probe selection, all probes from the genes LZTS1 and ZFP106 were dropped, yielding a 16-gene signature that was then applied to the 54 and 39 samples, respectively.
Removal of 2 "problematic" probes/genes, while retaining all 54 tumors, resulted in a modest increase in two pairwise correlations among Affy_FF vs. [1] Affy_FFPE and [2] NanoS_FFPE (comparing Table 1A vs. Table 2A). Of interest, the performance of t-RNA_FFPE and rRNA_FFPE were also improved with probe reduction.
Building to improve the model, the removal of both 15 samples and 2 probes/genes further enhanced the correlations of Affy_FF vs. Affy_FFPE, NanoS_FFPE, and RNA-ACC_FFPE (comparing Table 1A vs. Table 2B).
Using a PAM classifier to predict KRAS/BRAF mutation status using the 16-gene expression data The Affy_FF gene expression score might be considered a new "gold" standard because of its potential capacity to more inclusively identify tumors with RAS pathway activation not necessarily linked to RAS mutation. RAS mutation status, however, taken by itself, could also be considered a "gold" standard, and in fact is the current clinical standard used to qualify the administration of EGRFi therapies. We therefore sought to validate the known mutational status of previously sequenced CRC samples (n = 54) using our Affy_FF, Affy_FFPE and NanoS_FFPE datasets in conjunction with the modified 16-gene RAS signature score. In this regard, the samples were classified as either KRAS/BRAF mutant or KRAS/BRAF wild-type, resulting in two classes.
Notably, no NRAS mutation was detected in the 54 samples (Additional File 1). For each dataset, we developed linear models utilizing gene expression profiles of the 16 genes to predict the class (mutation type) of future samples. Table 3 shows the sensitivity and specificity values for the classifier, together with the LOOCV missclassification rates. Class prediction was performed using the gene expression data (n = 54) from the Affy_FF, Affy_FFPE and NanoS_FFPE samples. The 16-gene Affy_FF classifiers performed best in predicting KRAS/ BRAF mutation status (error rate = 19 %), with an optimal sensitivity of 0.852 and specificity of 0.778. Reduction in sample size was ineffective in improving KRAS/BRAF mutation status predictions (results not shown). Table 4 shows the reduced gene sets in the selected predictive model (one with the minimal error of prediction) for each of the validation datasets (Affy_FF, NanoS_FFPE and Affy_FFPE) out of the 16 genes.
Discussion
Gene expression signatures have been identified for prediction of RAS pathway dependence and drug response [2,3]. One obstacle to clinical translation is that these signatures were developed using cell lines and fresh frozen (FF) tissues, whereas usually only formalin-fixed, paraffin embedded (FFPE) tissue of lower quality is readily available for clinical use [19][20][21][22]. A number of studies have been reported on gene expression quantitation in FFPE samples using FF as a standard, usually employing one or two technologies, including RT-qPCR, Nano-String, and/or Affymetrix GeneChip [19,21,22,[35][36][37]. In this study, we simultaneously compared five technology platforms: [1] Affymetrix GeneChip; [2] NanoString; [20,21,[35][36][37][38]. While Lebbe and co-workers used the expression levels of a set of reference genes to construct a statistic for differentiating "bad" melanoma samples from "good" ones [36], Sfakianos et al. used PCA analysis to identify "bad" samples in ovarian cancer FFPE samples [19]. We adopted the PCA method to identify and filter out 15 "outlier" samples with "poor" RNA quality. The removal of the "outlier" samples improved the correlations of Affy_FF (as a "gold" standard) significantly with Affy_FFPE, but only slightly with NanoS_FFPE; the 15 "outlier" samples identified for the Affy_FFPE platform did not appear to be outliers for NanoS_FFPE. A plausible explanation is that in contrast to Affymetrix and RNASeq technologies, NanoString is a more "direct" technology (hybridization-based) to detect the number of RNA transcripts, so it does not need steps of mRNA reversetranscription into cDNA and subsequent cDNA amplification. Reverse-transcription and cDNA amplification are known to be sensitive to the RNA quality issue caused by RNA cross-linking in FFPE samples.
Since multiple different gene-specific probes (used in Affymetrix technologies) may have different sensitivities to the RNA quality of FFPE samples [21], we used the mean signal scores for the probes coupled with their fold-change information, to filter out 2 genes that were insufficiently expressed across the 39 samples. Notably, our probe filtration approach here differs from the filtration methods used previously in the literature [39]. The removal of these two genes improved the FF to FFPE translation by both Affymetrix and NanoString methods. This indicates that the RNA quality and probe problems are two different confounding factors for the translation of the RAS signature scores. Notably, we observed poor correlations and no significant improvement upon removal of the "outlier" samples and/or "bad" probes for Illumina Total RNA-stranded rRNA-depletion, and targeted RNASeq. However, the cause was not clear.
Moreover, NanoString mean scores were most significantly and consistently correlated with Affymetrix FF, Affymetrix FFPE and RNA-Access mean scores, in the presence or absence of bad samples and probes. Furthermore, while our data suggest that removing "bad" samples can improve the translation of a test from FF to FFPE tissues in Affymetrix FFPE and RNA-Acces platforms, identifying samples with poor RNA quality is not always an easy and practical task. Within a potential future diagnostic setting, it is impractical to perform a PCA across multiple samples to identify "bad" samples. Even if this were practical, it is far from ideal to exclude patients from diagnostic assessment because their FFPE samples happened to have lower quality RNA than usual. In addition, attempting to identify and remove poor quality samples adds an additional step to any analysis. Thus, due to its lower apparent sensitivity to In the assessment of the predictive ability for KRAS/ BRAF mutation status, the Affymetrix _FF 16-gene classifier produced the lowest misclassification rate (19 %) on the 54 samples. Our PAM analysis could further reduce the modified RAS pathway signature gene set from 16 to 7 genes in the Affy_FF classifier. Whereas all 18 genes were selected for capacity to identify MEK pathway activity independent of tumor genotype, the majority of the selected genes have particularly strong and direct relationships to the RAS/MEK/ERK pathway activation. DUSP4/6 [40] and PHLDA1 (TGAD51) [41] are known transcriptional targets of MEK/ERK. ETV4/5 [42] can replace RAS/MAPK pathway activation and TRIB2 can enhance ERK phosphorylation [43]. These relationships point to the strength of the signature genes identified by the algorithms applied to our sample sets. Of interest, SERPINB1 was retained in Affy_FF and NanoS_FFPE sample sets but appeared to have no direct relationship to RAS pathway activation.
Conclusions
Of the five technology platforms tested, NanoString technology was more adaptive to the translation of the RAS pathway signature from FF tissues to commonly available FFPE tissues than were the Affymetrix GeneChip and RNASeq technologies. NanoString was the most forgiving FFPE technology in reproducing the "gold" standard analysis on matched FF tissues. NanoString technology appears to rescue samples with poor RNA quality, permitting more samples to be scored. These critical analyses pave the way for a RAS pathway signature score to be used to assess FFPE CRC samples for applications such as prediction of EGFRi response to therapy.
==
Domain: Biology Medicine
|
[{"added": "2017-08-03T01:51:22.047Z", "created": "2016-10-19T00:00:00.000", "id": "15371753", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://bmcmedgenomics.biomedcentral.com/track/pdf/10.1186/s12920-016-0225-2", "pdf_hash": "2dc24de67d4386ccacc4b922dcb542f7bb74254c", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:31902", "s2fieldsofstudy": ["Medicine", "Biology"], "sha1": "2dc24de67d4386ccacc4b922dcb542f7bb74254c", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Metabolic effects of orally administered small-molecule agonists of GPR55 and GPR119 in multiple low-dose streptozotocin-induced diabetic and incretin-receptor-knockout mice.
AIMS/HYPOTHESIS
Abnormal cannabidiol (Abn-CBD) and AS-1269574 are potent selective agonists for GPR55 and GPR119, respectively. The present study evaluated the actions and ability of these small-molecule agonists to counteract experimental diabetes in mice.
METHODS
Diabetes was induced in NIH Swiss mice by five consecutive daily intraperitoneal injections of 40 mg/(kg body weight) streptozotocin. Diabetic mice received daily oral administration of Abn-CBD or AS-1269574 (0.1 μmol/kg) or saline vehicle (0.9% wt/vol. NaCl) over 28 days. Body weight, food intake, fluid intake, plasma glucose, insulin, glucose tolerance, insulin release, lipid profile and pancreatic morphology were examined. Mechanism of action of agonists was assessed in acute studies using incretin-receptor-knockout mice.
RESULTS
Abn-CBD and AS-1269574 decreased plasma glucose (20-26%, p < 0.05) and increased circulating insulin (47-48%, p < 0.05) by 10-28 days, compared with saline-treated diabetic controls. Food intake and polydipsia were reduced by both agonists (21-23%, p < 0.05 and 33-35%, p < 0.01, respectively). After 28 days of treatment, plasma glucagon concentrations were reduced (p < 0.01) and glucose tolerance was enhanced by 19-44% by Abn-CBD (p < 0.05 or p < 0.001) and AS-1269574 (p < 0.05 to p < 0.001). Plasma insulin responses were improved (p < 0.01) and insulin resistance was decreased (p < 0.05 or p < 0.01) in both Abn-CBD- and AS-1269574-treated groups. Triacylglycerols were decreased by 19% with Abn-CBD (p < 0.05) and 32% with AS-1269574 (p < 0.01) while total cholesterol was reduced by 17% (p < 0.01) and 15% (p < 0.05), respectively. Both agonists enhanced beta cell proliferation (p < 0.001) although islet area was unchanged. Acute studies in Gipr- and Glp1r-knockout mice revealed an important role for the glucagon-like peptide 1 (GLP-1) receptor in the actions of both agonists, with the glucose-lowering effects of Abn-CBD also partly mediated through the glucose-dependent insulinotropic peptide (GIP) receptor.
CONCLUSIONS/INTERPRETATION
These data highlight the potential for fatty acid G-protein-coupled receptor-based therapies as novel insulinotropic and glucose-lowering agents acting partly through the activation of incretin receptors.
Introduction
NEFAs play a complex role in glucose homeostasis and the pathogenesis of type 2 diabetes [1] and recent studies have shown that G-protein-coupled receptors (GPCRs) are involved in sensing NEFAs [2,3]. NEFAs have an emerging role in improving glucose-stimulated insulin release from pancreatic beta cells and improving insulin sensitivity in liver and skeletal muscle [4]. GPCRs have different affinities for fatty acids of varying chain lengths. Thus whereas mediumand long-chain fatty acids activate GPR40 and GPR120, short-chain fatty acids are known to serve as ligands for GPR41 and GPR43 [4]. Furthermore, novel receptors GPR55 and GPR119 have recently been shown to affect blood glucose control, and activation of these receptors with novel synthetic fatty acids may have therapeutic potential for type 2 diabetes [5].
GPR119 has been identified on pancreatic beta cells and intestinal L cells and K cells and has the ability to enhance glucose-stimulated insulin release and the secretion of both glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP) [6]. These incretin hormones have a variety of actions in pancreatic islets including enhancement of insulin biosynthesis and glucose-stimulated insulin release, stimulation of beta cell proliferation and inhibition of beta cell apoptosis [7]. GLP-1 additionally inhibits glucagon secretion [8]. GPR119 has been shown to bind a variety of lipid-derived ligands, as well as a range of small synthetic molecules. Several endogenous and synthetic GPR119 agonists have been shown to exhibit insulinsecretory properties in clonal beta cells, isolated islets and in vivo in mice [9,10]. AS-1269574 is a specific and potent fatty acid GPR119 agonist that enhances glucose-dependent insulin secretion [9,10]. AS-1269574 has also been implicated in GLP-1 secretion and increased proglucagon gene promoter activity via GPR119 in a mouse L cell line (GLUTag) [11]. A recent study found that a synthetic GPR119 agonist (PSN-632408) can enhance beta cell regeneration, improve islet cell graft survival and augment plasma active GLP-1 [12]. Previous reports have found that GPR119 agonists exhibit enhanced potency in vivo in combination with inhibition of enzymes such as fatty acid a m i d e h y d r o l a s e i n h i b i t o r ( U R B -5 9 7 ) [ 1 3 ] or dipeptidyl-peptidase-IV (DPP-IV; sitagliptin) [14,15].
GPR55 is expressed in the central nervous system, ileum, adipose tissue and endocrine pancreas, being predominantly on insulin-secreting beta cells [16,17]. GPR55 was initially de-orphanised as a cannabinoid receptor and this receptor binds many cannabinoid compounds. However, some GPR55 ligands are not cannabinoids and do not bind to either cannabinoid receptor 1 (CB1) or 2 (CB2) [16]. Cannabidiol (CBD) is a known GPR55 antagonist and has structural similarity to cannabinol and Δ 9 -tetrahydrocannabinol [18]. A range of synthetic CBD analogues have been synthesised, including abnormal cannabidiol (Abn-CBD), O-1918 and O-1602, which all act via GPR55 [5]. GPR55 has been associated with several physiological roles including anti-inflammatory activity [19], osteoclast function [20], insulin secretion and glucose homeostasis [17,21,22]. GPR55 agonists exhibited insulinotropic properties in clonal beta cells, isolated islets and in vivo in mice [17,21,22], with Abn-CBD noted as having the maximum potency and selectivity for GPR55 [21]. A recent study also using mice found that ablation of the GPR55 receptor increases adiposity and insulin resistance, selectively decreasing physical activity [23]. Further studies are required to assess the efficacy of Abn-CBD administration in the treatment of diabetes and other obesity-related diseases.
Recently we identified Abn-CBD and AS-1269574 as potent selective agonists for GPR55 and GPR119, respectively; both agonists exhibited acute glucose-lowering and insulinotropic properties in mice [9,21]. In the present study we have evaluated the long-term glucose-lowering effects of small-molecule agonists Abn-CBD and AS-1269574 in mice with diabetes induced by multiple-low-dose streptozotocin (STZ). To gain further information on the potential mechanism of action of these GPCR agonists, glucose-lowering and insulin-releasing properties were assessed in vivo using Giprand Glp1r-knockout mice.
See electronic supplementary material (ESM) Methods for details of materials.
Animals Male NIH Swiss mice (10-16 weeks old) were purchased from Harlan, UK. Gipr-and Glp1r-knockout mice crossed with the C57BL/6 strain (courtesy of B. Thorens, Lausanne, Switzerland and D. J. Drucker, Toronto, ON, Canada) and age-matched control wild-type C57BL/6 mice were obtained from an in-house breeding colony. The study was conducted in accordance with the Guide for the Care and Use of Laboratory Animals (2011) and the UK Animal (Scientific Procedures) Act 1986 and ARRIVE guidelines for reporting experiments involving animals [24].
Chronic administration of Abn-CBD and AS-1269574 in STZ-induced diabetic mice In a long-term study (28 days), the effects of daily oral administration of Abn-CBD and AS-1265974 (both at 0.1 μmol/[kg body weight]) [9,21] or saline vehicle (0.9% wt/vol. NaCl) were examined in multiple-low-dose streptozotocin-induced diabetic NIH Swiss mice. To induce diabetes, NIH Swiss mice fasted for 4 h received five consecutive daily intraperitoneal injections of STZ (40 mg/[kg body weight]). See ESM Methods for further details.
Histology Pancreatic tissues were removed at 28 days and processed as previously reported [21]. See ESM Methods for further details.
Biochemical analysis Analysis of blood samples was undertaken as previously reported [21,25]. See ESM Methods for further details.
Statistics Data are expressed as the means ± SEM. Results were compared using the Student's t test or one-way Fig. 1 Effect of daily administration of Abn-CBD and AS-1269574 on body weight (a), food intake (b), fluid intake (c) and plasma glucose (d) in STZ-induced diabetic mice. Variables were measured before and during 28 day treatment with Abn-CBD, AS-1269574 or saline vehicle (treatment period indicated by the horizontal black bar). Black squares, non-diabetic mice treated with saline (normal); white circles, STZ-induced diabetic mice treated with saline vehicle; black triangles, diabetic mice treated with Abn-CBD; white triangles, diabetic mice treated with AS-1269574. Values are means ± SEM for six mice. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with normal mice; + p < 0.05, ++ p < 0.01 and +++ p < 0.001 compared with diabetic mice In (a) variables were measured before and during 28 days of treatment with Abn-CBD, AS-1269574 or saline vehicle (treatment period indicated by the horizontal black bar) and in (b-d) they were measured after 28 days of treatment. Black squares, non-diabetic mice treated with saline (normal); white circles, STZ-induced diabetic mice treated with saline vehicle; black triangles, diabetic mice treated with Abn-CBD; white triangles, diabetic mice treated with AS-1269574. Values are means ± SEM for six mice. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with normal mice; + p < 0.05, ++ p < 0.01 and +++ p < 0.001 compared with diabetic mice circles, non-diabetic mice treated with saline (normal); white squares, STZ-induced diabetic mice treated with saline vehicle; black triangles, diabetic mice treated with Abn-CBD; black squares, diabetic mice treated with AS-1269574. Values are means ± SEM for six mice. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with normal mice; + p < 0.05, ++ p < 0.01 and +++ p < 0.001 compared with diabetic mice ANOVA on Prism graph pad version 5.0. Differences in data were considered to be statistically significant for p < 0.05.
Discussion
Much interest has been focused recently on fatty acid GPCRs in relation to their potential beneficial effects on glucose homeostasis in type 2 diabetes and other obesity-related diseases [1][2][3]26]. Several novel fatty acid receptors, including GPR55 and GPR119, have shown promise as emerging targets of diabetes therapies. Previous research has highlighted the potent acute glucose-lowering and insulinotropic ability of the GPCR agonists Abn-CBD and AS-1269574 [9,21] and our acute in vivo studies, using GPR55 and GPR119 antagonists, have confirmed the specificity of Abn-CBD and AS-1269574 in islets.
The present study evaluated the glucose-lowering and insulinotropic properties of small-molecule agonists Abn-CBD and AS-1269574 in multiple-low-dose STZ-induced diabetes in mice. This is a commonly employed means of inducing diabetes [27,28], which in the present study was associated with decreased body weight, hyperphagia, polydipsia, moderate hyperglycaemia, hyperglucagonaemia and hypoinsulinaemia but with good numbers of surviving functional beta cells, representing a model of mild type 1 diabetes. Daily oral administration of Abn-CBD and AS-1269574 lowered plasma glucose, plasma glucagon and food and fluid intake. In harmony with this, plasma insulin and pancreatic insulin content were increased in both treatment groups. Following long-term administration Values are means ± SEM for six mice, 25 islets per group. ***p < 0.001 compared with nondiabetic mice treated with saline; +++ p < 0.001 compared with diabetic mice. Arrows indicate insulin/Ki-67 co-positive beta cells of the agonists, glucose tolerance and insulin sensitivity were markedly improved. Additionally, total cholesterol and triacylglycerols were decreased by Abn-CBD and AS-1269574 and HDL-cholesterol was augmented by AS-1269574. Interestingly, our data would suggest a link between GPCR activation and suppression of hyperphagia. We found that GPR55 and GPR119 agonists decreased food consumption and appetite, though we observed no effect on overall body weight. This is in contrast to some published literature suggesting a link between GPR55 antagonism and reduced food consumption and body weight gain [22]. The accompanying decrease in fluid intake in the mice in our study most likely reflects the improved hyperglycaemic status following treatment with the two GPCR agonists. Moreover, GPR55 expression has been found to be increased in adipose tissue of obese individuals and further so in obese patients with type 2 diabetes, with GPR55 expression correlating with body weight, BMI and percentage fat mass [29]. Ex vivo studies using both adipose tissue explants and differentiated primary adipocytes show that L-α-lysophosphatidylinositol, an endogenous GPR55 agonist, increased the expression of genes stimulating fat deposition in adipose tissue and in differentiated adipocytes from visceral fat of obese patients it raised intracellular Ca 2+ concentration. The GPR119 data is generally consistent with a recent study which found that AS-1907417, a modified form of AS-1269574, improved lipid profile, plasma insulin, plasma glucose and pancreatic insulin content over a 4 week treatment period in db/db mice [30]. These recent studies suggest that activation of GPR55 and GPR119 could play a role in weight management, energy load and lipid metabolism.
Very few previous studies have assessed long-term administration of AS-1269574 and Abn-CBD in animal models of diabetes. Interestingly, bone mineral content, assessed by DEXA, was increased by agonising GPR119 with AS-1269574. GPR55 is expressed on osteoclasts [20] and, in contrast to our observations, GPR55-receptor-knockout mice have been reported to exhibit enhanced quantity and Values are means ± SEM for six mice. *p < 0.05, **p < 0.01 and ***p < 0.001 compared with 18 mmol/l glucose thickness of trabecular bone [20]. This difference may reflect an adaptive response in mice deficient in GPR55. No data are currently available on GPR119 and bone metabolism. Of interest, recent studies have highlighted the importance of GIP [31] and GLP-1 receptors [32] in bone physiology, including bone strength and quality. This may explain the enhancement in bone mineral density and content by the GPCR agonists in this study as the receptors, in particular GPR119, has been implicated in GIP and GLP-1 secretion [6,33,34].
In the present study, multiple-low-dose STZ induced a relatively moderate form of diabetes associated with hypoinsulinaemia and mild effects on islet architecture. The islets contained good numbers of surviving positively stained insulin beta cells, increased numbers of glucagon-secreting alpha cells and decreased expression of the GPR119 receptor. Oral administration of Abn-CBD or AS-1269574 resulted in increased numbers of insulin-secreting beta cells and reduced hyperglucagonemia confirmed by decreased circulating Fig. 10 Acute effect of GPR119 agonist in combination with GPR119 antagonist exendin (9-39) on plasma glucose (a, c) and plasma insulin (b, d). Fasted male lean NIH Swiss mice (n = 6) received glucose (18 mmol/[kg body weight]) alone or in combination with GPR119 agonist AS1269754 and/or GPR119 antagonist exendin (9-39) (0.1 μmol/[kg body weight]). Values are means ± SEM. Black circles, glucose alone; white squares, glucose + AS-1269574; black triangles, glucose + exendin (9-39); black squares, glucose + AS-1269574 + exendin (9-39). *p < 0.05, **p < 0.01 and ***p < 0.001 compared with glucose; + p < 0.05, ++ p < 0.01 and +++ p < 0.001 compared with glucose + AS-1269574 Fig. 11 Acute effect of GPR55 agonist in combination with GPR55 antagonist CBD on plasma glucose (a, c) and plasma insulin (b, d). Fasted male lean NIH Swiss mice (n = 6) received glucose (18 mmol/[kg body weight]) alone or in combination with GPR55 agonist Abn-CBD and/or GPR55 antagonist CBD. Black circles, glucose alone; white squares, glucose + Abn-CBD; black triangles, glucose + CBD; black squares, glucose + Abn-CBD + CBD. Values are means ± SEM. **p < 0.01 and ***p < 0.001 compared with glucose; ++ p < 0.01 and +++ p < 0.001 compared with glucose + Abn-CBD plasma glucagon levels. The expression of GPR55 and GPR119 was enhanced by Abn-CBD and AS-1269574, respectively, with GPR119 reverting to normal expression levels. In addition, treatment with GPR55 and GPR119 agonists augmented beta cell proliferation and regeneration. Consistent with this, a recent study reported that the GPR119 agonist PSN-632408 increased GLP-1 secretion and beta cell proliferation as assessed by Ki-67 and BrdU staining; these effects were further augmented by DPP-IV inhibitor sitagliptin [14]. Many factors are known to play a role in the regulation of beta cell proliferation, including incretins, and further studies are required to investigate the possibility of parallel changes in apoptosis and to discover whether the increased cellular proliferation and replication observed in this study with long-term administration of GPCR agonists is a result of the action of incretins on beta cell proliferation and beta cell function. Both GIP and GLP-1 have been implicated in a number of protective functions in pancreatic beta cells including proliferation, neogenesis and anti-apoptosis [35].
This study has confirmed [6,33,34] that GPR119 agonists stimulate the release of GLP-1 from L cells. The GPR55 agonist Abn-CBD exhibited potent glucose-lowering and insulin-releasing properties in C57BL/6 wild-type mice, effects that were abolished in both Giprand Glp1r-knockout mice. The GPR119 agonist AS-1269574 displayed strong glucose-lowering and insulinotropic ability in wild-type C57BL/6 mice as well as Gipr-knockout mice. In contrast, GPR119 activation with AS-1269574 removed the glucose-lowering effect of the ligand in Glp1r-knockout mice. These data indicate that, in addition to direct stimulatory effects of the agonists on beta cells, the beneficial effects of Abn-CBD also involve the GIP and GLP-1 receptors and AS-1269574 activation of the GLP-1 receptor.
Overall these results indicate that GPCR small-molecule agonists Abn-CBD and AS-1269574 exert a broad spectrum of actions on glucose homeostasis, islet and enteroendocrine cells, lipid profiles and pancreatic and bone composition. Interestingly, few studies have assessed the role of GPR55 in diabetes [16,18], with no literature probing the involvement of incretin hormones in the mechanism of action of GPR55. In contrast, several studies have identified the importance of GLP-1 [11,12,14,15] and GIP [6] in GPR119 activation in glucose homeostasis. Similar to the findings of this study, the GPR119 agonist AS-1269574 has also been found to enhance plasma GLP-1 concentrations [11,34]. Few studies have assessed the role of GPR119 in GIP signalling and the exact molecular mechanisms remain unclear. The synthetic agonist AR-231453 enhanced both GIP and GLP-1 levels in wild-type mice but not in Gpr119-knockout mice [34]. Additionally, AR-231453 stimulated GIP and GLP-1 release in wild-type mice and Glp1rand Gipr-knockout mice [6]. Further studies are clearly warranted to assess the role of GPR119 and GPR55 agonists on the incretin pathway.
In conclusion, long-term administration of GPR55 agonist Abn-CBD and GPR119 agonist AS-1269574 improved glycaemic control in a multiple-low-dose STZ-induced mouse model of diabetes, resulting in decreased plasma glucose and glucagon and improved plasma insulin, GLP-1 and lipid profiles. Additionally, glucose tolerance and insulin sensitivity were enhanced, with positive actions on the pancreas and bone. The results also indicate an important action through the incretin pathway, suggesting that agonists capable of agonising both fatty acid GPCRs and incretin secretion may have therapeutic potential in the future.
Funding These studies were supported by the Department of Education and Learning, Northern Ireland, and Ulster University Strategic Funding.
Contribution statement All authors made substantial contributions to conception and design, acquisition of data, reviewed the literature, drafted and revised the manuscript and approved the final submitted manuscript. AMK is the guarantor of this work.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
Open Access This article is distributed under the terms of the Creative Comm ons Attribution 4.0 International License ( [URL]:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
==
Domain: Biology Chemistry Medicine
|
[{"added": "2017-08-02T18:10:37.783Z", "created": "2016-09-27T00:00:00.000", "id": "10701214", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://link.springer.com/content/pdf/10.1007/s00125-016-4108-z.pdf", "pdf_hash": "13783265ed6b5cb945e3e3da84fd3688bda5c857", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39382", "s2fieldsofstudy": ["Medicine", "Biology", "Chemistry"], "sha1": "13783265ed6b5cb945e3e3da84fd3688bda5c857", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of the Natural Peptide Crotamine from a South American Rattlesnake on Candida auris, an Emergent Multidrug Antifungal Resistant Human Pathogen
Invasive Candida infections are an important growing medical concern and treatment options are limited to a few antifungal drug classes, with limited efficacies depending on the infecting organism. In this scenario, invasive infections caused by multiresistant Candida auris are emerging in several places around the world as important healthcare-associated infections. As antimicrobial peptides (AMPs) exert their activities primarily through mechanisms involving membrane disruption, they have a lower chance of inducing drug resistance than general chemical antimicrobials. Interestingly, we previously described the potent candicidal effect of a rattlesnake AMP, crotamine, against standard and treatment-resistant clinical isolates, with no hemolytic activity. We evaluated the antifungal susceptibility of several Candida spp. strains cultured from different patients by using the Clinical and Laboratory Standards Institute (CLSI) microdilution assay, and the antifungal activity of native crotamine was evaluated by a microbial growth inhibition microdilution assay. Although all Candida isolates evaluated here showed resistance to amphotericin B and fluconazole, crotamine (40–80 µM) exhibited in vitro activity against most isolates tested. We suggest that this native polypeptide from the South American rattlesnake Crotalus durissus terrificus has potential as a structural model for the generation of a new class of antimicrobial compounds with the power to fight against multiresistant Candida spp.
Introduction
Infections caused by Candida spp. have progressively increased over the last decades, and this phenomenon is mainly associated with the increasing number of critically ill patients exposed to invasive
Identification of Candida spp. by Sequencing of the ITS Region of rDNA
All isolates were identified at the species level by sequencing of rDNA ITS as previously described [29]. Total genomic DNA was extracted from the Candida isolates using PrepMan ® Ultra Sample Preparation Reagent (Applied Biosystems Inc., Foster City, CA, USA) according to the manufacturer's instructions. PCR for the amplification of the ITS region was performed using the forward primer V9G (5 -TTACGTCCCTGCCCTTTGTA-3 ) and the reverse primer LS266 (5 -GCATTCCCAAACAACTCGACTC-3 ) [30].
Assays for Checking In Vitro Susceptibility of Candida spp. against Antifungal Drugs
The antifungal susceptibility tests were performed using the Clinical and Laboratory Standards Institute (CLSI) microdilution assay [31]. In brief, susceptibility tests were performed in a 96-well plastic microplate containing RPMI 1640 (Sigma-Aldrich Corp., St. Louis, MO, USA) buffered at pH 7.0 with 0.165 M morpholinopropanesulfonic acid (Sigma-Aldrich Corp.), as outlined in the CLSI-M60 document [31]. Plates were incubated at 35 • C for 24 and 48 h. The following antifungal drugs were tested: Fluconazole (FLC) and amphotericin B (AMB) (from Sigma-Aldrich Corp.) and micafungin (MICA, provided by Astellas Pharma Inc., Tokyo, Japan). The final concentrations tested ranged from 0.03 to 16 µg/mL for AMB and MICA and from 0.5 to 64 µg/mL for FLC. The final inoculum density ranged from 0.5 to 2.5 × 10 3 cells mL −1 . The MIC results for each agent were determined visually. In the absence of international clinical breakpoint values for AMB and FLC against both species tested, we adapted the reference values for antifungal resistance recently suggested by the Centers for Disease Control and Prevention (CDC) [32] for C. auris: FLC ≥ 32 µg/mL and AMB ≥ 2 µg/mL, tested by the Clinical and Laboratory Standards Institute (CLSI) microdilution assay.
Assays for Checking In Vitro Susceptibility of Candida spp. for Natural Peptides Including Crotamine
For the antifungal susceptibility assay, the yeast Candida spp. was sown with disposable handles in Petri dishes (90 × 15 mm) containing 65 g/L of Sabouraud dextrose agar (Becton, Dickinson and Company, Franklin Lakes, NJ, USA), and the plates were incubated at 37 • C for 48 h. Candida spp. isolates were cultured in 5 mL of 25 g/L of PDB medium (Potato Dextrose Broth, Becton, Dickinson and Company), at 37 • C for 18 h in a shaker model New Brunswick™ Innova ® 43 (VWR International LLC, Radnor, PA, USA). The in vitro antifungal activity of crotamine and other native peptides was evaluated essentially as described in Yamane et al. [15]. Crotamine was tested in the concentrations of 10, 20, 40, 80, and 160 µM, while other natural peptides were tested in a single concentration (1 mM each), as also described in the Figure 1 legend. These native peptides were added in a 96-well plate containing the isolates of Candida spp. (about 2.5-5.0 × 10 3 cells/100 µL) in PDB medium, before incubation at 37 • C for 24 h. The Candida growth rate was determined by measurements at 595 nm using a plate reader SpectraMax (Molecular Devices LLC, San Jose, CA, USA). Briefly, the inhibition of growth was calculated by subtracting the density of the strain in the presence of the peptides studied from the maximum density value of this strain growth in the absence of peptides.
Statistical Analysis
Statistical analysis was performed using the two-way ANOVA test. GraphPad Prism Software (La Jolla, CA, USA) was employed for data analyses. The significance threshold was considered at p ≤ 0.05. Table 1 summarizes the in vitro susceptibilities of all isolates selected for the present study, after evaluation by the CLSI microbroth assay using FLC, AMB, and MICA. The South American and Middle East clinical isolates were resistant to at least two different classes of antifungal drugs. The Asian C. auris CBS 10913 reference strain was susceptible to all three antifungal drugs tested here. The antifungal activity of crotamine and other natural peptides was monitored by a microbial growth inhibition assay in a liquid medium essentially as described earlier [15], as it is the most commonly employed assay to evaluate and express the antifungal activity of native compounds [33], and also to allow direct comparison to previous data [15,25].
Susceptibility Tests
The activity of several natural peptides from invertebrate or vertebrate animals and plants were evaluated using fixed concentrations of about 1 mM of each peptide, and they did not show any important or significant inhibition of the growth of resistant C. auris clinical isolates from the South America or Middle East outbreaks [26,27], although the Asian C. auris CBS 10913 reference was more susceptible to these peptides, with the highest activity observed for cheliferin from the pseudoscorpion C. cancroides ( Figure 1). auris CBS 14916 (hatched dark gray columns). Peptides 1 and 2 are from the centipede Scolopendra subspinipes, cheliferin is from pseudoscorpion Chelifer cancroides, oligoventin is from the eggs of the Brazilian armed spider Phoneutria nigriventer, and comosusin is from pineapple peels. The percentage of inhibition determined by each peptide (about 1 mM each) was calculated by the ratio between the growth of the microorganisms in the presence/absence (which corresponds to 100% growth) of the peptides. The dashed line indicates approximately 50% inhibition of growth. Differences were considered statistically significant for values of p ≤ 0.05 for two-way ANOVA, for the multiple comparisons post-hoc Bonferroni, N = 4, and they are indicated by the bars.
Although crotamine was also more effective in inhibiting the Asian C. auris CBS 10913 reference compared to the resistant C. auris clinical isolates from South America or the Middle East outbreaks at low concentration, the activity of native crotamine against these resistant C. auris strains was demonstrated with an inhibition of about 50% of the yeast growth for most strains at concentrations of about 80-160 µM (which corresponds to approximately 0.4-0.8 mg/mL of crotamine). At these higher concentrations of crotamine (namely above 80-160 µM), a more pronounced effect of crotamine against the multiresistant C. auris 467/2015 and CBS 14916 strains could be observed ( Figure 2). Moreover, a trend for a more effective inhibition of resistant C. auris 470/2015 and 484/2015 strains compared to the CBS 10913 reference strain could be noticed at 80 µM of crotamine, while at the highest concentration used here (namely 160 µM) the differences were statistically significant for the C. auris 484/2015 multiresistant strain also (Figure 2). Although not different among the reference and clinical multiresistant strains, at 40 µM of crotamine close to 40% of inhibition of growth was observed for all evaluated strains, and therefore, we considered as the minimum inhibitory concentration (MIC) values ranging from 40 to 80 µM, which correspond to about 0.2-0.4 mg/mL of crotamine. The in vitro fungicidal activity of crotamine against all these clinical multiresistant C. auris strains was demonstrated here. Although crotamine was also more effective in inhibiting the Asian C. auris CBS 10913 reference compared to the resistant C. auris clinical isolates from South America or the Middle East outbreaks at low concentration, the activity of native crotamine against these resistant C. auris strains was demonstrated with an inhibition of about 50% of the yeast growth for most strains at concentrations of about 80-160 µM (which corresponds to approximately 0.4-0.8 mg/mL of crotamine). At these higher concentrations of crotamine (namely above 80-160 µM), a more pronounced effect of crotamine against the multiresistant C. auris 467/2015 and CBS 14916 strains could be observed ( Figure 2). Moreover, a trend for a more effective inhibition of resistant C. auris 470/2015 and 484/2015 strains compared to the CBS 10913 reference strain could be noticed at 80 µM of crotamine, while at the highest concentration used here (namely 160 µM) the differences were statistically significant for the C. auris 484/2015 multiresistant strain also (Figure 2). Although not different among the reference and clinical multiresistant strains, at 40 µM of crotamine close to 40% of inhibition of growth was observed for all evaluated strains, and therefore, we considered as the minimum inhibitory concentration (MIC) values ranging from 40 to 80 µM, which correspond to about 0.2-0.4 mg/mL of crotamine. The in vitro fungicidal activity of crotamine against all these clinical multiresistant C. auris strains was demonstrated here.
In addition, the closely related resistant C. haemulonii clinical isolates (9873/2014 and 1112/2016) were also evaluated with crotamine, but inhibition by less than 40% for concentrations up to 80 µM of crotamine was observed. In addition, no significant differences in inhibition efficiency with increasing concentrations of crotamine could be observed (Figure 3). In addition, the closely related resistant C. haemulonii clinical isolates (9873/2014 and 1112/2016) were also evaluated with crotamine, but inhibition by less than 40% for concentrations up to 80 µM of crotamine was observed. In addition, no significant differences in inhibition efficiency with increasing concentrations of crotamine could be observed (Figure 3).
Discussion
C. auris is an emergent multiresistant Candida species able to disseminate in the hospital environment [3]. Antifungal susceptibility data published so far point out that some C. auris strains exhibit elevated minimum inhibitory concentration (MIC) for the three major classes of antifungal
Discussion
C. auris is an emergent multiresistant Candida species able to disseminate in the hospital environment [3]. Antifungal susceptibility data published so far point out that some C. auris strains exhibit elevated minimum inhibitory concentration (MIC) for the three major classes of antifungal drugs, i.e., azoles, polyenes, and echinocandins [7]. C. haemulonii complex isolates are considered to be emergent species related to C. auris that may also exhibit a multiresistant phenotype to antifungal drugs [2]. So far, several authors have tried to explore different strategies of drug combinations to find new tools to combat this new emerging multiresistant pathogen [34][35][36].
The value of screening natural compounds for searching for new compounds with antimicrobial activity against multiresistant strains is well recognized. However, the several different peptides from invertebrate and plants, with already demonstrated antimicrobial activity, showed important activity only against the Asian C. auris CBS 10913 reference strain, with no significant effect against the multiresistant C. auris clinical isolates from South America or the Middle East. On the other hand, the effective antifungal activity of crotamine against these multiresistant clinical C. auris strains from different patients with fungemia points out the potential of crotamine as a structural model for the development of a new generation of antimicrobial drugs against multidrug-resistant clinical strains. At this point, it is also worth considering that although few clinical strains were evaluated, all those strains were from independent emergences of different clonal populations on different continents [26][27][28].
In the present study, we also decided to evaluate the antifungal activity of crotamine against clinical multiresistant C. haemulonii, which is often misidentified by commercial identification methods and presents a multidrug-resistant profile as confirmed here (Table 1) and as described by others [2], but with no important effect in concentrations up to 80 µM of crotamine.
Crotamine is a well-characterized polypeptide with multiple biological activities [37], including the in vitro activity against a selected panel of Candida species [23]. Crotamine is also a recognized member of the antimicrobial peptide (AMPs) class whose members are efficient in killing most microbes, and for which the development of resistance is rare, mainly due to the characteristic mechanism(s) of action based on the rapid interaction with and disruption of lipid cell membranes [15,19,38]. AMPs represent therefore a powerful drug candidate with reduced risk of resistance development and potential reduction in the duration of treatment [38].
The action of crotamine on mitochondria was also previously demonstrated by us [13], and phenolic compounds such as flavonoids, when combined with FLC, were demonstrated to show activity against Candida tropicalis strains resistant to FLC, by promoting mitochondrial depolarization, apoptosis, and exposure of phosphatidylserine in the plasma membrane [39]. Interestingly, crotamine is also able to promote mitochondrial depolarization and apoptosis [12,13], besides having an affinity for negatively charged lipids such as phosphatidylserine [19,38]. Furthermore, although the lipid-membrane-disrupting activity of crotamine and its shorter derived peptides was previously demonstrated by us [19,38], at this point we cannot simply come to the conclusion that the negative charges in the cell membrane of FLC-resistant strains could be playing a role in the candicidal effect against the C. auris resistant strains. In addition, the low ability of native crotamine in inhibiting the closely related C. haemulonii resistant clinical isolates also deserves special attention, and further studies are planned to clarify this selective activity of native crotamine only against the C. auris clinical strains.
Conclusions
Based on our present experiments, we suggest that this native polypeptide from the South American rattlesnake has potential as a structural model compound for the generation of a new class of antimicrobial compound with power against multiresistant nosocomial Candida strains, representing a possible new road to overcome the microbial resistance challenge against emerging opportunistic human fungal pathogens. At this point, it remains unclear if the concentrations of crotamine required to inhibit C. auris and C. haemulonii strains could be safely administrated for treating human infections. However, we may certainly suggest that these data encourage further studies to explore the possible use of crotamine as topical antifungal agents or to study the structural aspects of crotamine as a molecular template for modeling and designing of new molecules with higher efficacy against multiresistant clinical strains.
However, further studies are still necessary to determine the molecular mechanism of action underlying the crotamine activity against the multidrug-resistant strains of C. auris and its relative inefficiency against the C. haemulonii clinical strains, and this may certainly be the next target of our future studies.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2019-05-30T23:43:21.908Z", "created": "2019-05-28T00:00:00.000", "id": "169039090", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/2218-273X/9/6/205/pdf", "pdf_hash": "54a525fb7a023cea189058d684faae1227f2ad90", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:40062", "s2fieldsofstudy": ["Medicine", "Biology", "Environmental Science"], "sha1": "54a525fb7a023cea189058d684faae1227f2ad90", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Immunomodulatory Effects of Edible and Medicinal Mushrooms and Their Bioactive Immunoregulatory Products
Mushrooms have been valued as food and health supplements by humans for centuries. They are rich in dietary fiber, essential amino acids, minerals, and many bioactive compounds, especially those related to human immune system functions. Mushrooms contain diverse immunoregulatory compounds such as terpenes and terpenoids, lectins, fungal immunomodulatory proteins (FIPs) and polysaccharides. The distributions of these compounds differ among mushroom species and their potent immune modulation activities vary depending on their core structures and fraction composition chemical modifications. Here we review the current status of clinical studies on immunomodulatory activities of mushrooms and mushroom products. The potential mechanisms for their activities both in vitro and in vivo were summarized. We describe the approaches that have been used in the development and application of bioactive compounds extracted from mushrooms. These developments have led to the commercialization of a large number of mushroom products. Finally, we discuss the problems in pharmacological applications of mushrooms and mushroom products and highlight a few areas that should be improved before immunomodulatory compounds from mushrooms can be widely used as therapeutic agents.
Introduction
In clinical practice, immunomodulators are usually classified into three categories: immunosuppressants, immunostimulants, and immunoadjuvants [1]. Their market share has increased rapidly over the past few years due to wide-ranging medical applications for patients that require human immune system modulations. Immune system modulations are also commonly used as prophylactic medicine for an increasing number of healthy people [2,3]. While most immunomodulators are synthetic or semi-synthetic compounds, there has been a growing interest in natural immunomodulators. Many natural compounds have shown significant immunomodulatory and overall health-benefiting effects to humans, with no or minimal toxicity. These natural-based products with potential pharmacological and beneficial effects are increasingly perceived as safer than synthetic compounds by the general public [4,5]. Indeed, many of the currently used chemical drugs have negative side effects
Medicinal Mushrooms
As mentioned above, medicinal mushrooms refer to all macroscopic fungi whose extracts or powder form from any stages of the mushroom development have shown documented beneficial effects on health [13]. These beneficial effects may have been shown in the forms of in vitro, ex vivo, or in vivo activities. Their effects may cover different groups of organisms such as antagonistic effects against human pathogens and parasites, and/or beneficial effects for human and animal cell lines, or animal and human individuals [11]. Since many edible mushrooms and their products have shown to be a beneficial component of the human diet, some of these edible mushrooms are also commonly included as medicinal mushrooms [7]. In our literature search, a large number of MMs have been documented. For example, terpenes and terpenoids from Ganoderma lucidum could stimulate the expressions of genes coding for proteins in the nuclear factor (NF)-kB pathway and modulate immune system functions [14]. Heteroglycan and heteroglycan-peptide from the mushroom of Hericium erinaceus can modulate the immuno-effects by inducing nitric oxide production and increasing expression of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-12 [15]. These mushrooms belong to two fungal phyla, Ascomycota and Basidiomycota. Most of the MMs are in phylum Basidiomycota. Table 1 shows the major medicinal mushrooms, including their taxonomy and geographic/ecological distributions. As can be seen, some of these mushrooms are broadly distributed (e.g., the button mushroom Agaricus bisporus) while others are geographically more restricted (e.g., the Himalayan caterpillar fungus Ophiocordyceps sinensis). Some of the mushrooms included in the table e.g., Amanita phalloides are highly poisonous when Some of these MM species have been used as herbal medicine for centuries, including Ganoderma lucidum, Ganoderma lingzhi, Lentinula edodes, Inonotus obliquus, Fomitopsis officinalis, Piptoporus betulinus, and Fomes fomentarius [7,17]. While these mushrooms have attracted most of the medical attention among the MMs, other species in multiple genera have also shown immunomodulatory and anti-tumor effects, such as those in genera Agaricus, Albatrellus, Antrodia, Calvatia, Clitocybe, Cordyceps, Flammulina, Fomes, Funlia, Ganoderma, Inocybe, Inonotus, Lactarius, Phellinus, Pleurotus, Russula, Schizophyllum, Suillus, Trametes, and Xerocomus [12]. Figure 1 shows a few representative medicinal mushroom species in their natural habitats. Some medicinal mushrooms are only found in the wild, e.g., the ectomycorrhizal mushrooms Boletus edulis and Russula lepida. However, a large number of medicinal mushrooms are also commercially cultivated, including Shiitake, Ling-zhi, and Lion's Mane. Figure 2 shows a few representative medicinal mushrooms under cultivation.
Immunomodulatory Compounds and Commercial Products of Medicinal Mushrooms
As shown above, there is a large number of medicinal mushrooms in diverse taxonomic groups. Some of these medicinal mushrooms are commercially cultivated for food but there is an increasing
Immunomodulatory Compounds and Commercial Products of Medicinal Mushrooms
As shown above, there is a large number of medicinal mushrooms in diverse taxonomic groups. Some of these medicinal mushrooms are commercially cultivated for food but there is an increasing trend for developing the immune-active compounds from these cultivated mushrooms into nutraceuticals. Tables 2-5 summarize the major groups of bioactive compounds in medicinal mushrooms and their demonstrated immunomodulatory effects to specific pathologies, including the relevant references.
The main classes of compounds from medicinal mushrooms with immunomodulatory properties are terpenes and terpenoids, lectins, fungal immunomodulatory proteins (FIPs), and polysaccharides (particularly β-d-glucans, but also include polysaccharopeptides and polysaccharide proteins) [1]. Below we describe specific examples in each of these groups.
Polysaccharides
Among the bioactive compounds derived from mushrooms with immunomodulatory activity, those based on polysaccharides, with or without side chain modifications (including polysaccharopeptides and polysaccharide proteins) are the most reported during the last several decades [18]. Table 2 presents a list of polysaccharides from medicinal mushrooms that have shown immunomodulatory activities. Among the reported polysaccharides with immunomodulatory and antitumor activities, the best-known is lentinan, isolated from shiitake (L. edodes), as well as schizophyllan from Schizophyllum commune. Both lectinan and schizophyllan contain β-1,3-d-glucans with β-1,6 branches. Specifically, lentinan showed immunomodulatory properties against gastric cancer while schizophyllan was effective against head and neck cancer. Both products have been licensed and approved in Japan since 1986 for clinical use, in combination with chemotherapy against the two respective cancers [19,20].
Other polysaccharide-based compounds showing immunomodulatory properties have a similar core polysaccharide chemical structure but contain different branching linkages and/or branches with different conjugates. These polysaccharide-conjugate complexes are called heteroglucans, with α(1-4)and β(1-3) glycosidic linkages to protein components. For example, in the presence of fucose (a hexose deoxy sugar with the chemical formula C6H12O5), the turkey tail mushroom Trametes versicolor produces a Krestin bound β-glucan polysaccharide K (PSK). PSK is commercially produced from this mushroom in Japan and has been approved for clinical use since 1977 [21]. Several subsequent reports confirmed the effectiveness of PSK as an adjuvant to conventional cancer therapies through inhibition of cancer metastasis [22], induction of cancer cell apoptosis [23], improvement of inflammatory cytokines gene expression [24,25].
Another compound isolated from T. versicolor is a polysaccharide peptide (PSP). PSP contains rhamnose and arabinose, two monosaccharides not found in PSK. In addition, the conjugated protein was also different. PSP has been commercially available in the Chinese market since 1987 [21]. It has been documented to improve the quality of life in cancer patients by providing substantial pain relief and enhancing immune status in 70-97% of patients with stomach, esophagus, lung, ovary and cervical cancers. Specifically, PSP has been shown to be capable of boosting immune cell production, ameliorating chemotherapy symptoms, and enhancing tumor infiltration by dendritic and cytotoxic T-cells [26].
Similarly, a large number FIPs have been identified. The FIP names, the medicinal mushrooms that produce them, and evidence for their specific immunomodulatory effects are presented in Table 4. Among these, the best known is probably Ling-Zhi-8 from G. lucidum which acts as an immunosuppressive agent [1]. In addition, aside from immunomodulation, many FIPs have also shown antitumor activities in pharmacological tests, including the inhibition of cell growth and proliferation, the induction of apoptosis and autophagy, and the reduction of invasion and migration of tumor cells. At present, most of these tests are conducted using tissue cultures. Further tests using animal models and clinical trials are needed in order to confirm their safety and efficacy in humans. If confirmed, these FIPs could be more efficiently produced and commercialized through genetic engineering for clinical use. Table 3. Major immunomodulatory lectins from medicinal mushrooms and their immunomodulatory effects.
Source
Lectin name
Immunomodulation and Other Human Health Effects of Medicinal Mushrooms
The human immune system is tightly linked to tumor development. With the increasing impact of tumor on human health, a large number of studies have been undertaken to identify mushroom extracts/fractions/compounds with antitumor activities. Indeed, some of the observed antitumor activities by medicinal mushroom extracts were based on the activation of the immune system (Tables 2-5). Davis et al. (2020) recently suggested that 17 medicinal mushroom species (A. brazei, Cordyceps militaris, Flammulina velutipe, F. fomentarius, F. officinalis, Ganoderma applanatum, G. lucidum, Ganoderma oregonense, G. frondosa, Hericium erinaceus (syn. Hericium erinaceum), I. obliquus, L. edodes, Tropicoporus linteus (syn. Phellinus linteus), P. betulinus, Pleurotus ostreatus, S. commune) could support both immune-activation for cancer treatments and help resolve host defense-induced inflammatory reactions and facilitate a post-response return to homeostasis for cancer patients. Furthermore, a medicinal mushroom formulation consisting of G. lucidum, L. edodes and G. frondosa showed synergistic antitumor and immuno-modulatory activity in human macrophages [137].
As shown in Tables 2-5, many medicinal mushrooms each can produce different categories of compounds with immunomodulatory effects. Furthermore, different extractions of the same mushroom may show non-overlapping but complementary activities. For example, in L. edodes, its heterogalactan (fucomannogalactan) has anti-inflammatory properties [138], lentinan has anti-tumor effect [139], crude water-soluble polysaccharides can activate macrophages and increase the productions of nitric oxide (NO), cytokines, and proteins related to phagocytosis [140], and polysaccharides with both antioxidant effects [141] and antiviral activities [142].
The significance of functional components from medicinal mushrooms has been shown not only from clinical perspectives but also from foods. Because many medicinal mushrooms are commercially cultivated for food, there has been an increasing trend of including mushrooms and their components into other foods to develop functional foods, including adding new flavors or promoting certain types of functions. For example, Ulziijargal et al. used mushroom mycelia of T. camphoratus, A. blazei, H. erinaceus, and P. linteus to substitute 5% of wheat flour to make bread. The final product contains substantial amounts of the amino acids gaminobutyric acid (GABA) and ergothioneine and showed beneficial health effects [143]. Kim et al. developed noodles that contained L. edodes paste, resulting in a higher quality, fibre-rich functional food with antioxidant and hypocholesterolemic properties [144]. Components of other mushrooms, e.g., Pleurotus sajor-caju dry powder, A. bisporus extracts, the freeze-dried powder from A. aegerita, Suillus luteus, and Coprinopsis atramentaria have been used to develop snacks and cheese-related products with much commercial success [145][146][147][148]. The benefits include increased proteins, minerals, crude fiber or ingredients with antioxidant potentials and free radical scavenging capacities. The diversified flavors and tastes as well as the enhanced nutritional values of food due to the addition of mushroom components represent an exciting direction for the edible and medicinal mushroom industries.
Mechanisms for the Immunomodulation Effects of Medicinal Mushroom Compounds
The immune system consists of a network of cells, tissues and organs that work together to defend the body against attacks by "foreign" invaders [2]. The network is connected by lymphatic vessels from organ to organ. The network includes protective barriers that constantly communicate with lymphatic fluid rich in white blood cells and leukocytes. When pathogens break our physical barriers (i.e., skin and mucosal membranes of the mouth, nose, the gastrointestinal tract, and the urogenital tract), the next line of the body's defense response is activated. This line of defense includes granulocytes and monocytes that also function as antigen-presenting cells (APCs) for helper T lymphocytes. These cells synthetize and secrete lipid mediators such as prostaglandins as well as cytokines which act as messengers in regulating immune response and stimulating adaptive immunity. For example, natural killer (NK) cells can recognize infected and abnormal cells, such as cancer cells and kill these cells by inducing them to undergo apoptosis or by producing cytokines, such as interferon-gamma (IFN-γ). They also activate macrophages and kill phagocytosed microbes. Figure 3 presents the overview of the key immune response against microbial pathogens. urogenital tract), the next line of the body's defense response is activated. This line of defense includes granulocytes and monocytes that also function as antigen-presenting cells (APCs) for helper T lymphocytes. These cells synthetize and secrete lipid mediators such as prostaglandins as well as cytokines which act as messengers in regulating immune response and stimulating adaptive immunity. For example, natural killer (NK) cells can recognize infected and abnormal cells, such as cancer cells and kill these cells by inducing them to undergo apoptosis or by producing cytokines, such as interferon-gamma (IFN-γ). They also activate macrophages and kill phagocytosed microbes. Figure 3 presents the overview of the key immune response against microbial pathogens. It is well known that the human immune system can be modulated by foods, supplements or endogenous bioactive agents [2]. Different types of immunoregulatory compounds have been isolated from medicinal mushrooms, including mushroom fruiting bodies and fermented mycelia. It is well known that the human immune system can be modulated by foods, supplements or endogenous bioactive agents [2]. Different types of immunoregulatory compounds have been isolated from medicinal mushrooms, including mushroom fruiting bodies and fermented mycelia. Tables 2-5 present the specific components isolated from different mushroom species that have shown significant immunomodulatory activities, including their (potential) mechanisms of action. For example, two polysaccharides from two different mushroom species have shown significant immunoenhancing activities. In the first, a glucuronoxylomannan TAP-3 obtained from Naematelia aurantialba (syn. Tremella aurantialba) showed marked immune enhancement activity and promoted NO, IL-1β and TNF-α secretions from macrophages [149]. Similarly, another study showed that at a concentration of 40 µg/mL, polysaccharide CCP from Craterellus cornucopioides strengthened the phagocytic function of macrophages, increased the expression of cytokines, upregulated the expression of cell membrane receptor TLR4 and downstream protein kinase products through activation of the TLR4-NFκB pathway [150].
Some of these bioactive compounds can also directly attack cancer cells while showing immunoregulatory effects. For example, Li et al. reported that polysaccharide LRP-1 purified from Leccinum rugosiceps inhibited the growth of human hepatoma cells HepG2 and human breast carcinoma MCF-7 cells, and induced the secretion of NO, IL-6 and TNF-α in vitro [151]. Similarly, a recent report showed that an aqueous extract of Sarcodon imbricatus (SIE) effectively inhibited the growth, migration, and invasion properties of breast cancer cells in vitro and reduced tumor growth in vivo, while showing increased expression of PD-L1 and increased NK cell viability [152]. Furthermore, Xue et al. reported that a triterpenoid EAe from Pleurotus eryngii inhibited MCF-7 cell lines proliferation with an EC50 of 298.26 µg/mL, and significantly inhibited the growth of CD-1 tumors (inhibition rate of 65%) in mice in a dose-dependent manner without toxicity [153].
Relationship between Structure and Activity of Immunomodulatory Compounds from Medicinal Mushrooms
Immunomodulators from medicinal mushrooms have been shown to be capable of stimulating both innate and adaptive immune responses. They activate innate immune system components such as natural killer (NK) cells, neutrophils, and macrophages, and stimulate the expression and secretion of cytokines. These cytokines in turn activate adaptive immunity by promoting B cell proliferation and differentiation for antibody production and by stimulating T cell differentiation to T helper (Th) 1 and Th2 cells, which mediate cellular and humoral immunities, respectively [154].
As shown in Tables 2-5, a large number of immunoregulatory compounds from medicinal mushrooms have been reported. These compounds differ greatly in their molecular weight and structure. Below we describe the relationships between their molecular structures and immunoregulatory activities.
Polysaccharides
Polysaccharides are the most commonly reported natural immunomodulators from mushrooms. The immunomodulating polysaccharides are highly diversified in their sugar compositions, main chain polymer structures, degrees of branching, conformations, molecular weights, and other physical properties, which together have significant effects on the bioactivity and mode of action of each polysaccharide [18]. These polysaccharides are either homoglycans (polysaccharides that contain residues of only one type of monosaccharide molecule) or heteroglycans (polysaccharides that contain residues of two or more types of monosaccharide molecules), and are able to combine with other molecules such as oligo-or poly-peptides to make peptidoglycan or polysaccharide-protein complexes. In general, higher molecular weight polysaccharides exhibit greater bioactivity [155]. These large polysaccharides are not able to penetrate the immune cells, but instead act to bind cell receptors. For example, the highest immunomodulatory activity of PSK was associated with the highest molecular weight fraction of this compound, at >200 kDa [156]. Similarly, the highest activity of a polysaccharide fraction of G. frondosa extract was ascribed to one with a molecule weight of over 800 kDa [157]. In contrast, low molecular weight polysaccharides can penetrate immune cells and exert stimulatory effects from within.
The number and lengths of short branched chains in mushroom polysaccharides can significantly influence their bioactivity [155]. In most cases, the bioactive immunomodulator polysaccharides are characterized by a main chain of 1,3-β-d-glucan with a small number of short branched chains with 1,6-β-linkage ( Figure 4). Studies have shown that immunologically active polysaccharides generally have a degree of branching number (DB) between 20% and 40%. For example, lentinan has a DB number of 40%, schizophyllan of 33%, and PSK of 20%. While a high DB number is generally correlated with a high activity, in some cases, debranching of polysaccharides can also increase their bioactivity. For example, the partially debranched form of pachymaran from Poria cocos showed greater activity than the original native form [158]. Even in the well-studied lentinan, its maximal immunomodulating and antitumor activities were achieved when the molecule had a DB of 32% [159], and there was a negative correlation between their biological activity and DB number between 32 and 40% [160]. Figure 4). Studies have shown that immunologically active polysaccharides generally have a degree of branching number (DB) between 20% and 40%. For example, lentinan has a DB number of 40%, schizophyllan of 33%, and PSK of 20%. While a high DB number is generally correlated with a high activity, in some cases, debranching of polysaccharides can also increase their bioactivity. For example, the partially debranched form of pachymaran from Poria cocos showed greater activity than the original native form [158]. Even in the well-studied lentinan, its maximal immunomodulating and antitumor activities were achieved when the molecule had a DB of 32% [159], and there was a negative correlation between their biological activity and DB number between 32 and 40% [160]. Aside from the main-chain structure and branching pattern, the conformation of polysaccharides can also impact their bioactivity, e.g., by influencing the stability of the structure. In polysaccharides, the triple helix conformation is usually more stable than other conformations and bears the cytokine stimulating activity of the β-D-glucan. Lentinan, schizophyllan, scleroglucan, and PSK all have a triple helix structure [161]. However, not all polysaccharide immunomodulators from mushrooms have the triple helix structure. For example, mushroom polysaccharides with a random coil conformation can also have potent immunomodulating and anticancer activity [56].
Chemical modification is an effective and common approach to increase biological activities of polysaccharides. This approach has been applied to develop a number of effective immunomodulators from mushroom polysaccharides. Those modifications include Aside from the main-chain structure and branching pattern, the conformation of polysaccharides can also impact their bioactivity, e.g., by influencing the stability of the structure. In polysaccharides, the triple helix conformation is usually more stable than other conformations and bears the cytokine stimulating activity of the β-d-glucan. Lentinan, schizophyllan, scleroglucan, and PSK all have a triple helix structure [161]. However, not all polysaccharide immunomodulators from mushrooms have the triple helix structure. For example, mushroom polysaccharides with a random coil conformation can also have potent immunomodulating and anticancer activity [56].
Chemical modification is an effective and common approach to increase biological activities of polysaccharides. This approach has been applied to develop a number of effective immunomodulators from mushroom polysaccharides. Those modifications include carboxymethylation, hydroxylation, formyl-methylation, amino-ethylation, or sulfation. The introduction of such chemical groups may increase the possible contacts between the modified polysaccharides and the immune cell receptors through hydrogen bonding and/or electrostatic attraction, and thus increase the immunological response. For example, the sulfated cell wall glucan from L. edodes exhibited higher immunomodulatory and anticancer activities compared to the native polysaccharides [162]. The increased effect from sulfation may be related to the increased solubility, as shown in the hyper branched β-glucan TM3b. Taken together, molecular weight, branching, chemical configuration, and chemical modification can all have strong influence on the bioactivity of polysaccharides from mushrooms.
Lectins
Lectins belong to a unique group of proteins that can recognize and interact with various cell surface carbohydrates/glycoproteins. Mushroom lectins have shown specific immunomodulatory, antiproliferative, and antitumor activities [163]. The diverse sources of mushroom lectins have different immunomodulatory mechanisms: some mediate their actions by activating the immune system while others produce potent cytotoxic effects towards cells [91]. For example, two lectins extracted from L. mongolica (syn. T. mongolicum), TML-1 and TML-2, show immunomodulatory and antitumor activities. These two lectins stimulate the production of nitrite and tumor necrosis factor (TNF)-α and inhibit the growth of mouse lymphoblast-like (p815) mastocytoma cells by the production of macrophage-activating factors. These factors include interferon (IFN)-γ and other cytokines, activated through upregulation of inducible nitric oxide synthase (NOS), interleukin (IL)-1β, and transforming growth factor-β [91]. G. frondosa lectin is reported as having a potent cytotoxic effect against HeLa cells in vitro, even at very low concentrations. A 15.9-kDa homodimeric, lactose-binding, ricin-B-like lectin (CNL) from Clitocybe nebularis exhibited antiproliferative activity against human leukemic T cells [78], which induces the maturation and activation of dendritic cells (DCs) and stimulates several proinflammatory cytokines such as IL-6, IL-8, and TNF-α [164]. The encoding gene of CNL from C. nebularis has been cloned and successfully expressed in Escherichia coli [165].
FIPs
The fungal immunomodulatory proteins are a group of proteins with highly similar amino acid sequences. They exist as dimers in a dumbbell-shaped structure similar to that of the variable region of immunoglobulin heavy chains [166]. The FIPs have shown diverse functions. Through binding to Toll-like receptors (TLRs), FIPs stimulate antigen presenting cells and release cytokines such as NO and IL-12. By activating phosphorylation of p38/MAPK and increasing the production of NF-κB, FIPs can promote the proliferation and differentiation of helper T cells (Th0) to form Th1 cells and Th2 cells, activate macrophages and B cells, produce a variety of cell factors ( Figure 5). For example, FIP-fve from Flammulina velutipes can upregulate the expression of intercellular adhesion molecules on the T cell surface by phosphorylation of p38/MAPK, and activate Th1 cells to produce IL-2, IFN-γ, to exert its immunomodulatory effect [99]. FIP-vvo can not only activate Th1 cells and enhance IL-2, TNF-α and IFN-γ transformations, but also induce Th2 cells to produce IL-4, B cell differentiation, and the transformation of immunoglobulin and production of antibody IgE. Several studies have also shown that by interacting with TLRs, FIP can activate other signaling pathways besides the p38/MAPK and NF-κ B. For example, FIP from Ganoderma tsugae (FIP-gts) can stimulate human peripheral blood monocyte to produce IFN-γ and activates the PI3K/Akt signaling pathway [104]. exert its immunomodulatory effect [99]. FIP-vvo can not only activate Th1 cells and enhance IL-2, TNF-α and IFN-γ transformations, but also induce Th2 cells to produce IL-4, B cell differentiation, and the transformation of immunoglobulin and production of antibody IgE. Several studies have also shown that by interacting with TLRs, FIP can activate other signaling pathways besides the p38/MAPK and NF-κ B. For example, FIP from Ganoderma tsugae (FIP-gts) can stimulate human peripheral blood monocyte to produce IFN-γ and activates the PI3K/Akt signaling pathway [104]. FIPs typically exist in low quantities in their native mushrooms. The low yield/production has been a major limitation of their research and application. Therefore, techniques are being rapidly developed to enhance the production of recombinant FIPs in other organisms such as the yeast Pichia FIPs typically exist in low quantities in their native mushrooms. The low yield/production has been a major limitation of their research and application. Therefore, techniques are being rapidly developed to enhance the production of recombinant FIPs in other organisms such as the yeast Pichia pastoris and the bacterium E. coli. For example, the LZ-8 gene of G. lucidum has been expressed in P. pastoris to produce a recombinant LZ-8 protein (rLZ-8). While the recombinant protein lacks the carbohydrate moiety of the native protein, it shows similar bioactivity for IL-2 induction as the native protein. The FIP-fve protein has also been successfully expressed in E. coli [1]. Interestingly, the recombinant FIPs showed higher immunomodulatory activity and induced greater expression of specific cytokines than that extracted from the mushrooms [167].
Terpenes and Terpenoids
Terpenes and terpenoids are widely distributed in mushrooms. They are a large and diversified group of organic compounds but share the core of isoprene five-carbon atom units of molecular formula (C 5 H 6 )n as the main building block ( Figure 6) [1,13]. Among this group of compounds, the best-known is probably the triterpenoids from G. lucidum and G. lingzhi. These triterpenoids can help reduce drug nephrotoxicity and minimize inflammation. Figure 6 shows a diversity of terpene derivatives in G. lucidum and G. lingzhi, including ganodermic and ganoderic acids, ganoderals, ganoderols, ganodermanontriol, lucidone, and ganodermanondiol. All these compounds have shown immunomodulating, antitumor, and/or anti-infective activities [168]. At present, their mechanisms of action and structure-activity relationships are little understood. However, their broad activities suggest significant potential for research and for clinical therapeutic applications. derivatives in G. lucidum and G. lingzhi, including ganodermic and ganoderic acids, ganoderals, ganoderols, ganodermanontriol, lucidone, and ganodermanondiol. All these compounds have shown immunomodulating, antitumor, and/or anti-infective activities [168]. At present, their mechanisms of action and structure-activity relationships are little understood. However, their broad activities suggest significant potential for research and for clinical therapeutic applications.
Genomes and Molecular Techniques in the Study of Immunomodulatory Compounds in Medicinal Mushrooms
Due to their environmental, agricultural, commercial, and/or medical interests, the genomes of a number of medicinal mushrooms have been sequenced and annotated. Table 6 lists the genomic features of 12 representative medicinal mushrooms. These species differ in genome size and/or gene content. Analyses of these genomes have revealed some of the genes related to the synthesis and production of immunomodulatory compounds in medicinal mushrooms. Not surprisingly, the most commonly identified genes related to immunomodulatory effects are those coding for FIPs (Table 6). However, in G. lucidum, genes involved in the synthesis of several other immunomodulators have also been identified. The identification and confirmation of those genes require genetic manipulation systems which are not available at present for most medicinal mushrooms. In G. lucidum, such a system is available.
Genomes and Molecular Techniques in the Study of Immunomodulatory Compounds in Medicinal Mushrooms
Due to their environmental, agricultural, commercial, and/or medical interests, the genomes of a number of medicinal mushrooms have been sequenced and annotated. Table 6 lists the genomic features of 12 representative medicinal mushrooms. These species differ in genome size and/or gene content. Analyses of these genomes have revealed some of the genes related to the synthesis and production of immunomodulatory compounds in medicinal mushrooms. Not surprisingly, the most commonly identified genes related to immunomodulatory effects are those coding for FIPs (Table 6). However, in G. lucidum, genes involved in the synthesis of several other immunomodulators have also been identified. The identification and confirmation of those genes require genetic manipulation systems which are not available at present for most medicinal mushrooms. In G. lucidum, such a system is available.
Ganoderic acids (GAs) are among the main active ingredients of G. lucidum with immunomodulatory effects. GAs belong to the triterpenoid secondary metabolites. Genome sequence analyses and functional studies showed that the terpenoids in G. lucidum are synthesized through the Mevalonate (MVA) pathway. Several genes in this pathway in G. lucidum have been cloned and their functional roles confirmed, including those encoding 3-Hydroxy-3-methylglutaryl-CoA reductase (HMGR) and Farnesyl diphosphate synthase (FPPs). Genome sequence data mining also identified the putative genes involved in the modification of the triterpene backbone, such as those involved in cyclization and glycosylation, which are very important for the synthesis of the diversity of GAs in G. lucidum.
At present, most of the genes and metabolic pathways involved in the synthesis of immunomodulators in the categories of polysaccharides, lectins, and terpenoids in medicinal mushrooms have not been identified or confirmed. However, the availability of increasing genomic resources coupled with the broad pharmacological activities and therapeutic effects of medicinal mushrooms should help facilitate the identification of genes and metabolic pathways involved in their biosynthesis. Such understandings could help future productions of those compounds through biotechnology using surrogate hosts. [193,194]
Conclusions and Perspectives
Many edible and medicinal mushrooms contain compounds with significant immunoregulatory activities. This paper attempts to provide a comprehensive review on the types of these compounds; their distributions, structures and functions; and their potential mechanisms of actions. These compounds have shown their activities through in vitro, ex vivo, tissue cultures, and/or in vivo studies. Some of these compounds have been commercialized and licensed for clinical use. Aside from the above described compounds, other compounds from edible and medicinal mushrooms may also have great potentials. One such group is chitin and chitin-related compounds. Chitin is the most common aminopolysaccharide polymer in nature and the main material that gives strength to the fungal cell walls (as well as to the exoskeletons of crustaceans and insects). Through deacetylation, either chemically or enzymatically, chitin can be converted to chitosan, a well-known derivative. Through hydrolysis, both chitin and chitosan can be converted to chito-oligosaccharides. Several fungal chitin, chitosan, and chito-oligosaccharides have shown promising benefits to humans and human health [195,196]. For example, chitins from filamentous molds such as Aspergillus niger and Mucor rouxii and other organisms have been used in plant protection and food processing; chitosan in diagnosis, drug delivery, infection control, molecular imaging, and wound healing; and chito-oligosaccharides in antimicrobial and antitumor activities [195,196]. At present, none of those tested chitin, chitosan, and chito-oligosaccharides for human effects have come from edible or medicinal mushrooms yet. However, due to the similar chemical structures of chitin from different groups of organisms, it's highly likely that this group of natural products from edible and medicinal mushrooms will have similar effects and they represent a promising area of future development for edible and medicinal mushrooms.
While the future looks bright, significant issues remain before the full potential of medicinal mushrooms can be reached. Specifically, during our review of the literature, we identified several significant gaps and areas for future research and development. In the first, there is an urgent need to identify the structures and mechanisms of action for active ingredients in many extracts and formulations from medicinal mushrooms. More rigorous chemical analyses as well as understanding the in vivo pharmacokinetics and pharmacodynamics of individual compounds are needed to fill this gap of knowledge. The second promising area of study is to identify the genes and metabolic pathways involved in producing these immunomodulators in medicinal mushrooms. As shown above, aside from the few FIPs where the specific encoding genes have been identified, we have little information about the genes and how they are regulated in producing the other types of mushroom immunomodulators. While the availability of high through-put technologies and genome sequences are facilitating the discoveries, experimental investigations are needed in order to confirm and identify the conditions for increased productions of these compounds. Fortunately, gene editing technologies and -omics tools are becoming increasingly accessible to the broader life sciences communities. In the third, most immunomodulators described above exist in low quantities in medicinal mushrooms and their extractions can take a long time and be costly. For efficient production, it is very important to develop alternative approaches, e.g., by cloning and expressing the relevant biosynthesis genes in alternative hosts, using industrial fermentations, and developing efficient extraction and purification protocols from such commercial cell cultures. Lastly, the potential interactions between immunomodulators from mushrooms and other medicines, foods, and food supplements need to be critically analyzed in order to establish guidelines for safe and effective use of immunomodulators from medicinal mushrooms [197]. Indeed, at present, safety data about many medicinal mushroom products are not available from controlled clinical trials and associated negative side effects have been reported in several cases for certain types of usages of medicinal mushroom products [198,199]. There is also a cultural difference between Oriental and Western cultures about the use of medicinal mushrooms [197], presenting both a challenge and an opportunity for researchers and policy makers on the broad implications of these products on human health. Author Contributions: S. Z. and J. X. conceived and prepared the manuscript, Q. G., C. R. and Z. Z. participated in the manuscript editing; S. W. and Y. L. provided the photos of mushrooms. All authors have read and agreed to the published vision of manuscript.
Funding: This work was supported by National Natural Science Foundation of China (NSFC31701975), by the Collaborative Innovation Center of Beijing Academy of Agricultural and Forestry Sciences (KJCX201915), and by Beijing Academy of Agriculture and Forestry Sciences (KJCX20200208).
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2020-11-12T09:10:04.641Z", "created": "2020-11-08T00:00:00.000", "id": "226304317", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/2309-608X/6/4/269/pdf", "pdf_hash": "b0a0297e1b8d9b5aa3849d3603727d0fb271193c", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:33076", "s2fieldsofstudy": ["Medicine", "Biology", "Environmental Science"], "sha1": "410f764d935058558131a16684cfd7d00132c5ec", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Spinal Cord Injury Reduces Serum Levels of Fibroblast Growth Factor-21 and Impairs Its Signaling Pathways in Liver and Adipose Tissue in Mice
Spinal cord injury (SCI) results in dysregulation of carbohydrate and lipid metabolism; the underlying cellular and physiological mechanisms remain unclear. Fibroblast growth factor 21 (FGF21) is a circulating protein primarily secreted by the liver that lowers blood glucose levels, corrects abnormal lipid profiles, and mitigates non-alcoholic fatty liver disease. FGF21 acts via activating FGF receptor 1 and ß-klotho in adipose tissue and stimulating release of adiponectin from adipose tissue which in turn signals in the liver and skeletal muscle. We examined FGF21/adiponectin signaling after spinal cord transection in mice fed a high fat diet (HFD) or a standard mouse chow. Tissues were collected at 84 days after spinal cord transection or a sham SCI surgery. SCI reduced serum FGF21 levels and hepatic FGF21 expression, as well as β-klotho and FGF receptor-1 (FGFR1) mRNA expression in adipose tissue. SCI also reduced serum levels and adipose tissue mRNA expression of adiponectin and leptin, two major adipokines. In addition, SCI suppressed hepatic type 2 adiponectin receptor (AdipoR2) mRNA expression and PPARα activation in the liver. Post-SCI mice fed a HFD had further suppression of serum FGF21 levels and hepatic FGF21 expression. Elevated serum free fatty acid (FFA) levels after HFD feeding were observed in post-SCI mice but not in sham-mice, suggesting defective FFA uptake after SCI. Moreover, after SCI several genes that are implicated in insulin’s action had reduced expression in tissues of interest. These findings suggest that downregulated FGF21/adiponectin signaling and impaired responsiveness of adipose tissues to FGF21 may, at least in part, contribute to the overall picture of metabolic dysfunction after SCI.
INTRODUCTION
Spinal cord injury (SCI) causes partial or total interruption of neural signal transmission between the brain and the periphery, thereby limiting physical activity. Chronic SCI, due to an extremely sedentary lifestyle, invariably results in adverse body composition changes characterized by decreased lean mass and increased adiposity. Such changes are associated with metabolic dysfunction (1)(2)(3)(4)(5) leading to an increased risk for cardiovascular disease, liver disorders, and type 2 diabetes (T2DM) compared to the general population (6)(7)(8)(9). However, the underlying mechanisms for the predisposition of persons with SCI to metabolic perturbations and the associated cardiac and hepatic disorders remain unclear. Several lines of evidence suggest that alterations in the liver or visceral adipose tissue including neutrophil infiltration, macrophage activation, increased expression of pro-inflammatory cytokines and hepatic lipid accumulation (10)(11)(12)(13) which often is associated with development of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH) (14,15). Furthermore, persons with SCI have decreased anabolic hormone levels (16) which are associated with abnormal lipid and metabolic profiles. A clear understanding of hepatic pathology that develops after SCI may provide new insight into the development of impairments of fat and carbohydrate metabolism. However, limited studies have addressed the details of pathophysiological events which occur in the liver after SCI. The mechanism underlying dyslipidemia and whether this change precedes, or is the cause of, systemic metabolic dysfunction after SCI is not known. In addition, the potential mediators, if any, that regulate the interaction between hepatic function and other tissues to achieve systemic metabolic homeostasis after SCI have yet to be investigated.
Fibroblast growth factor-21 (FGF21) is a metabolically active peptide primarily secreted by the liver in response to various nutritional, physiological and pathological stimuli (17). FGF21 is an integral component of the hormone network that modulates the action of insulin, leptin, glucagon-like peptide (GLP1), adiponectin, growth hormone, and insulin-like growth factor (18). FGF21 has been shown to regulate carbohydrate and lipid metabolism in the liver, adipose tissue, skeletal muscle, pancreas, and central nervous system (18)(19)(20). In animal models, FGF21 corrects multiple abnormalities of metabolism, including those present in persons with insulin resistance and T2DM (21)(22)(23). FGF21 reduces levels of serum triglycerides (TG), total cholesterol, and low-density lipoprotein (LDL)-cholesterol but raises the level of high-density lipoprotein (HDL)-cholesterol, all of which changes represent favorable alterations to the lipid profile (19,21). Transgenic mice overexpressing FGF21 exhibit resistance to the development of high fat diet (HFD)-induced obesity (24). In addition, FGF21 exerts hepato-protective effects by promoting hepatic lipid and free fatty acid (FFA) metabolism, reducing hepatic adiposity, and ameliorating NAFLD and NASH (19,22,(25)(26)(27). Moreover, FGF21 can act on cardiac muscle to protect against heart disease (28) and ameliorates obesity-related inflammation in a rat model reducing inflammatory cytokine production (29). The molecular mechanisms by which FGF21 produces its physiological effects are, however, still being elucidated. FGF21 transcription is induced by peroxisome proliferator-activated receptor (PPAR)-g and PPAR-a, which are the targets of agents prescribed to treat T2DM (thiazolidinediones) and hyperlipidemia (fibrates), respectively (23,30). FGF21 requires the obligate coreceptor b-klotho (KLB) for cellular signaling both in vivo and in vitro (31). The binding of FGF21 to KLB enables its interaction with the FGF receptor 1 (FGFR1), thus activating intracellular signaling pathways (32). FGF21 signaling is tissue-specific and is chiefly determined by tissue distribution of KLB. Adipose tissue has high levels of KLB expression and is a primary site of FGF21 action (32,33). In obese individuals, serum FGF21 levels positively correlate with the subcutaneous adipose tissue (SAT) area (34). FGF21 knockout mice show less SAT and are more insulin-resistant when fed a HFD (24,34). Mechanistically, FGF21 improves systemic insulin sensitivity by promoting the healthy expansion of SAT (34), thereby increasing serum levels of adiponectin, a major adipokine produced by adipose tissue, which mediates the beneficial effects of FGF21 on the liver and muscle (35,36). Adiponectin signaling is mediated by activation of two types of receptor: type 1 (AdipoR1) is predominantly expressed in skeletal muscle and activates AMP-kinase; the type 2 adiponectin receptor (AdipoR2) is predominantly expressed in the liver and activates the PPARa signaling pathway (37). Of note, the magnitude of FGF21-induced weight loss is reduced in leptin-deficient mice (33), implying a possible requirement of a functional leptin axis for FGF21 action on body weight regulation (38). While the beneficial effects of FGF21 on metabolic homeostasis have been extensively studied in ablebodied individuals, FGF21 expression and function after SCI has not been investigated.
Persons with SCI have been reported to have reduced energy expenditure resulting from a marked reduction in lean tissue mass and in the ability to ambulate and/or exercise (39,40). On the other hand, recent reports indicated increased total food intake in persons with SCI when compared to individuals who were immobilized for reasons other than SCI (41,42). In addition, analysis of the pattern of food intake of overweight or obese persons with SCI suggests that there is an increased proportion of total calories consumed from fat (43). These post-SCI dietary changes almost certainly accelerate the accumulation of fat in various body compartments. However, studies that investigated nutritional status in chronic SCI are limited, and knowledge with respect to how macronutrients impact metabolic gene expression and dysregulation of metabolic homeostasis after SCI is lacking. Furthermore, the relationship between serum FGF21 levels and SCI is unknown.
Using a spinal cord transection model, a high fat diet was recently reported to induce glucose intolerance without increasing fat mass or liver weights (44). To understand the possible role of FGF21 in altered fat, carbohydrate and energy metabolism following SCI, blood and tissue samples collected from SCI mice fed either a high fat diet (HFD) or control diet (ConD) were used to determine circulating FGF21 levels and expression of FGF21 mRNA. The cause of the potential perturbations in FGF21 levels and, if altered, the probable impact on lipid and carbohydrate metabolism was investigated by analyses of relevant signaling and molecular changes in fat, liver and muscle.
Animal Studies
Procedures with experimental animals were approved by the Institutional Animal Care and Use Committee at the James J. Peters VA Medical Center and were conducted in accordance with all applicable Federal standards for humane animal care and use of laboratory animals. A detailed description of the experimental protocol has been previously published (44). Briefly, three-month-old male C57BL/6 mice were purchased from Charles River. Animals were randomly assigned to 4 groups: two sham-laminectomy (Sham) groups in which the spinal cord was exposed but not transected, and two thoracic spinal cord transection (level T9) groups. At day 84 after surgery, the animals were euthanized. The liver, adipose tissue including inguinal fat, (iFAT) and omental fat (oFAT); and gastrocnemius muscles were carefully removed and immediately frozen in liquid nitrogen. Blood was collected by ventricular puncture, and serum was frozen for subsequent analysis.
High-Fat Diet (HFD)
A lard-based HFD was purchased from Research Diets [D12492; 60% fat, 20% protein, 20% carbohydrate, 5.21 kcal/g energy density]. A source and micro-nutrient control chow with a macronutrient content similar to standard rodent chow [D12450J; 10% fat, 20% protein, 70% carbohydrate, 3.82 kcal/g energy density] was fed to the control-diet (ConD) group. Animals were fed standard rodent chow pre-surgery with mice having ad libitum access to HFD and control chows beginning immediately post-surgery. Chow was replaced every 3d.
Reverse Transcriptase Quantitative Real-Time PCR (RT-qPCR)
Total RNA was extracted using RNasey mini kits (QIAGEN, Germantown, MD). Relative expression of mRNA was determined by qRt-PCR, as described previously (45) using a model ViiA7 thermocycler (Applied Biosystems, Foster City, CA). Specific primers and TagMan Universal Master Mixer were purchased from Applied Biosystems. TagMan Reaction Mixes contained 3 ml cDNA (10 ng/ml). For each sample, the determinations were performed in triplicate. Relative mRNA levels were expressed as fold-change using the 2 −DDCt method. Data were normalized relative to 18s RNA.
Preparation of Tissue Lysates and Western Blotting (WB)
The tissues were homogenized in 500 ml lysate buffer [150 mM sodium chloride, 3.2 mM Na 2 PO 4 , 0.8 mM K 2 PO 4 (pH 7.4), 1% NP-40, 0.5% sodium deoxycholate, 0.5% sodium dodecylsulfate] using a tissue grinder bead homogenizer (MP Biomedicals Inc., Solon, OH) followed by centrifugation at 14,000 rpm for 5 min. Western blot analysis was performed as described previously (45). Antibodies against phospho-ACC and total ACC were purchased from Cell Signaling Technology (Beverly, MA). Anti-phospho-PPARa, total PPARa and b-tubulin antibodies were obtained from AbCam (Cambridge, MA). Densitometry was determined using Image Lab software (Bio-Rad, Hercules, CA). b-tubulin was used as a loading control. All assays were performed in triplicate. Mean Ct values for the replicates were used in all subsequent calculations.
Preparation of Nuclear Extracts and PPARa DNA-Binding Activity Assay
Nuclear extracts were prepared from mouse liver using a commercial kit (Pierce; Rockford, IL). The binding capacity of PPARa to DNA was assayed using a non-radioactive assay kit from AbCam (Cambridge, MA), which uses a sensitive method for detecting specific transcription factor DNA binding activity in the nuclear extracts. The assays were performed according to the manufacturer's instructions in duplicate with mean values for the two determinations used in all subsequent calculations.
Serum Preparation and Enzyme-Linked Immunosorbent Assay (ELISA)
The blood samples were collected on ice and allowed to clot before being centrifugation for 20 min at 1000 g with the serum separated from the formed blood cell elements. The FGF21, adiponectin and leptin assays were performed using a commercial ELISA kit (R&D System; Minneapolis, MN); The high molecular weight (HMW) adiponectin assay was performed by a commercial kit (Creative Diagnostics; Shirley, NY). Alanine aminotransferase (ALT) and free fatty acid (FFA) assay kits were purchased from MyBioSource Inc. (Santa Ana, CA) and AbCam respectively. All assays were performed according to manufacturer's instructions. Assays were performed in duplicate with mean values for the two determinations used in all subsequent calculations.
Glycogen Content Assay
Tissue glycogen levels were measured using a colorimetric kit (AbCam). Briefly, mouse liver and skeletal muscle tissues were homogenized with 200 ml distilled water on ice. The homogenates were boiled for 10 min followed by centrifugation at 18,000 rpm for 10 min. Supernatant was collected and subjected to assay according to the manufacturer's instruction. Samples were assayed in duplicate and mean values for the duplicates were used in subsequent calculations.
Statistics
Data are expressed as mean values ± SEM. The significance of differences between sham and SCI groups, with or without HFD, was determined using Two-Way ANOVA with a Bonferroni test post-hoc. Statistical calculations were performed using Prism software (GraphPad, San Diego, CA). p < 0.05 was considered statistically significant.
SCI Reduced Serum FGF21 levels and Inhibited FGF21 Receptor Expression in Adipose Tissue
To determine whether FGF21 plays a role in SCI-induced dysregulation of metabolism, serum FGF21 levels were examined. There was a significant reduction in serum FGF21 levels after 84 days of SCI in mice fed a ConD (SCI-ConD) compared to sham-mice fed the same diet (sham-ConD) ( Figure 1A). Because circulating FGF21 levels correlate well with hepatic gene expression (46), FGF21 mRNA expression in mouse liver was determined. As shown in Figure 1B, SCI led to a significant downregulation of hepatic FGF21 mRNA expression. In addition, we observed that sham mice fed with HFD (sham-HFD) also had reduced FGF21 serum levels and hepatic FGF21 mRNA expression; serum FGF21 and hepatic FGF21 mRNA were further decreased in SCI-HFD mice compared to SCI-ConD mice ( Figure 1B). Adipose tissue expresses high levels of FGFR1 and KLB and is a primary site of FGF21 action in regulating fat and carbohydrate metabolism (32). Therefore, the effects of SCI with or without HFD on FGFR1 and KLB gene expression were examined in adipose tissues. Because increased intra-abdominal fat worsens insulin resistance, as opposed to extra-abdominal fat depots (47), both inguinal fat (iFAT) and omental fat (oFAT) were analyzed. There were significant reductions in both FGFR1 and KLB mRNA expression in iFAT ( Figures 1C, D) and oFAT of SCI-ConD mice compared to sham-ConD mice. While HFD-feeding did not change FGFR1 and KLB mRNA levels in iFAT of sham mice, a significant increase in mRNA expression was seen in oFAT of sham-HFD mice ( Figures 1C, D). In SCI-HFD mice, KLB mRNA levels were reduced compared to SCI-ConD mice in iFAT ( Figure 1C) but not oFAT ( Figure 1D).
SCI Reduced Expression and Serum Levels of Adiponectin and Leptin
It has been demonstrated that FGF21 increases insulin sensitivity through direct effects on fat to stimulate expression and secretion of adiponectin, a critical mediator of FGF21 actions in the liver and skeletal muscle which are two major sites of insulin-induced glucose uptake, and also by specific expansion of subcutaneous fat (34)(35)(36). Thus, we determined whether reduced serum FGF21 impacted the production of adiponectin and leptin as well as adipose tissue differentiation markers. Consistent with reduced levels of FGFR1 and KLB noted in both iFAT and oFAT of SCI-ConD mice, there was a significant decrease in mRNA expression of adiponectin in iFAT and oFAT of SCI-ConD mice as compared to sham-ConD mice ( Figure 2A). Serum levels of both total adiponectin and HMW adiponectin, the most potent form of the protein (48), were also reduced by SCI ( Figures 2B, C). Levels of serum total adiponectin in sham-HFD and SCI-HFD mice were further decreased as compared to sham-ConD and SCI-ConD groups, respectively ( Figure 2B). HMW adiponectin was decreased in sham-HFD mice when compared to sham-ConD mice but was not different when comparing SCI-HFD and SCI-ConD mice ( Figure 2C). Leptin, another major adipokine, has been implicated in one of the actions of FGF21 (33,38). The effects of SCI and HFD on leptin mRNA expression and protein secretion were therefore examined. Leptin mRNA in both iFAT and oFAT ( Figure 2D), and serum protein levels ( Figure 2E) were reduced in SCI-ConD mice as compared to sham-ConD mice. Feeding mice with a HFD induced a significant increase in iFAT and oFAT leptin mRNA and serum leptin protein in sham mice, but did not alter serum leptin or leptin mRNA in iFAT or oFAT of SCI-HFD mice when compared to SCI-ConD mice ( Figures 2D, E), suggesting to an impaired ability of adipose tissue to secrete leptin in response to a HFD. PPARg, a downstream target of FGF21 in adipocytes, is a master transcriptional regulator of adipocyte differentiation, and a determinant of adiponectin expression (23,49). There was a significant reduction in PPARg mRNA expression in iFAT and oFAT of SCI-ConD mice as compared to sham-ConD mice ( Figure 2F). Dramatically reduced mRNA levels of ASC-1, a marker of white adipose tissue (50), was also observed in iFAT and oFAT of SCI-ConD mice ( Figure 2G). Expression of PPARg and ASC were not significantly changed by HFD in sham or SCI mice ( Figures 2F, G). These data indicate that SCI decreased secretion of adiponectin and leptin, and inhibited adipocyte differentiation suggesting a linkage of these impairments to reduced FGF21 signaling.
SCI Reduced Hepatic Adiponectin Receptor 2 (AdipoR2) Expression and PPARa Activity
AdipoR2 is predominantly expressed in the liver and plays an important role in regulating hepatic lipid metabolism (37,51). To evaluate whether SCI impacts hepatic adiponectin signaling, AdipoR2 expression was examined in the liver of SCI mice. There was a significant reduction in AdipoR2 mRNA expression in SCI-ConD mice compared to sham-ConD mice and these levels were further decreased in SCI-HFD mice as compared to SCI-ConD mice ( Figure 3A). A HFD also decreased AdipoR2 mRNA in livers of sham-HFD mice compared to sham-ConD mice ( Figure 3A).
The expression and activity of PPARa, a nuclear receptor that serves as a downstream target of AdipoR2 (37,52), was also examined. Neither SCI nor HFD changed hepatic PPARa mRNA levels ( Figure 3B). PPARa transcriptional activity is decreased by phosphorylation at Ser12, (53). To understand how SCI altered PPARa activity, the effect of SCI on PPARa phosphorylation and nuclear levels were determined. While SCI had no significant effect on total PPARa protein ( Figures 3C, D). the levels of phospho-(Ser12)-PPARa was increased in SCI-ConD compared to sham-ConD groups ( Figures 3C, E). Phospho-(Ser12)-PPARa was lower in SCI-HFD mice compared to SCI-ConD mice and, by contrast, was higher in sham-HFD mice than sham-ConD mice ( Figures 3C, E). To determine nuclear PPARa activity in SCI vs. sham mice, a DNA-binding assay was performed using nuclear extracts of mouse liver. PPARa DNA-binding capacity was reduced in SCI-ConD mice compared to sham-ConD mice but was elevated in SCI-HFD compared to SCI-ConD mice; in contrast, sham-HFD demonstrated decreased nuclear PPARa compared to sham-ConD ( Figure 3F). These data are consistent with hepatic resistance to adiponectin after SCI due to lower expression of AdipoR2, lower nuclear PPARa and perturbed response of PPARa to dietary fats (52,54).
SCI Influenced Rate Limiting Steps in Hepatic Cholesterol and Fatty Acid Metabolism
Because FGF21 regulates hepatic lipid and fatty acid metabolism and reduced PPARa activity after SCI (17,19,(25)(26)(27), the effects of SCI on hydroxy-methyl-glutaryl Co-A reductase (HMGCR), the rate limiting enzyme in cholesterol synthesis, and ABCA1, the cholesterol efflux regulatory protein, were tested. HMGCR mRNA levels were markedly increased in livers from SCI-ConD mice compared to sham-ConD mice, whereas reduced mRNA levels were observed in SCI-HFD mice as compared to SCI-ConD mice. HMGCR mRNA levels were also reduced by HFD in sham mice ( Figure 4A). In contrast, neither SCI nor HFD had no effect on ABCA1 mRNA expression ( Figure 4B) suggesting that SCI may influence hepatic cholesterol synthesis but not efflux. We next examined the effect of SCI on acetyl-coenzyme A carboxylase (ACC), the rate limiting enzyme controlling fatty acid synthesis and metabolism. Expression of ACCa mRNA levels, the predominant isoform in the liver, was not changed in the SCI-ConD group compared to the sham-ConD group ( Figure 4C). ACCa mRNA was significantly reduced in the sham-HFD and SCI-HFD groups compared to sham-ConD and SCI-ConD groups, respectively ( Figure 4C). Expression of ACCa protein was upregulated in SCI-ConD mice compared to sham-ConD mice without a further change in the SCI-HFD mice when compared to SCI-ConD mice; HFD did not alter total ACCa protein in sham mice ( Figures 4D, E). To understand how SCI and HFD influenced ACCa activity, ACCa phosphorylation at Ser79, which inhibits ACC activity (55), was tested. There was a significant decrease in Ser79phosphorylated ACCa protein in SCI-ConD mice compared to sham-ConD mice ( Figures 4D, F). Ser79-phosphorylation was not altered by HFD in sham or SCI mice ( Figures 4D, F). These results suggest that SCI suppressed FGF21 signaling associated with increased activity of HMGCR and ACCa pathways that determine hepatic lipid synthesis and metabolism.
HFD Raised Serum FFA and Hepatic FABP4 Expression After SCI
To further determine the influence of SCI on lipid and fatty acid metabolism, serum FFA levels, which correlate with insulin resistance (56) were examined. When compared to sham-ConD mice, serum FFA levels were not significantly different in SCI-ConD mice although a trend toward lower values in SCI-ConD mice was noted ( Figure 5A). However, as compared to SCI-ConD mice, there was a significant elevation in serum FFA level in SCI-HFD mice, whereas HFD did not alter serum FFA in sham mice ( Figure 5A), suggesting defective FFA uptake by adipose or other tissues after SCI. In addition, the hepatic expression of FABP4 mRNA was examined, the level of which has been linked to insulin resistance, T2DM, hypertension and cardiac dysfunction (57). Although SCI-ConD mice had no change in FABP4 mRNA expression as compared to sham-ConD mice, SCI-HFD mice showed increased expression in hepatic FABP4 mRNA when compared to SCI-ConD mice; HFD did not change FABP4 mRNA levels in livers of sham mice ( Figure 5B). This finding further implies that HFD may worsen SCI-induced metabolic dysfunction.
SCI Induced Increased Inflammatory Gene Expression in Fat and Liver
Inflammatory changes in liver and fat have been linked to impaired metabolism of carbohydrates (58). To understand if SCI and/or HFD influenced inflammatory status in liver or fat, selected mRNA levels of inflammatory cytokines were determined. Expression of TNFa mRNA in iFAT and oFAT was increased in SCI-ConD mice as compared to sham-ConD mice; a further increase in TNFa mRNA levels were observed in SCI-HFD mice when compared to SCI-ConD mice ( Figure 6A). In sham mice, HFD increased TNFa mRNA expression in oFAT but not iFAT ( Figure 6A). The expression levels of IL-1b and IL-6, two major cytokines induced by inflammation, were also tested. As shown in Figure 6, SCI had no effect on IL-1b or IL-6 mRNA in iFAT, but significantly upregulated IL-1b levels in oFAT ( Figure 6B). HFD-feeding increased IL-1b and IL-6 mRNA levels in iFAT but not in oFAT ( Figures 6B, C). While SCI had no effect on TNFa or TNF-receptor mRNA expression in mouse liver (Supplemental Data, Figure 2), significant upregulation of expression of IL-1b ( Figure 6D), IL-6 ( Figure 6E) and CD11b mRNA, a macrophage marker, ( Figure 6F) were observed in the liver from SCI-ConD mice as compared to sham-ConD mice. The expression of IL-1b and IL-6 mRNA was not changed by HFD in sham or SCI mice. In contrast, while CD11b mRNA was not changed in sham-HFD mice, it was significantly elevated in SCI-HFD mice as compared to SCI-ConD mice ( Figure 6F).
To understand the potential impact of elevated CD11b on hepatocytes, serum and hepatic mRNA levels of ALT, a hepatic enzyme that has been used as a sensitive marker of liver injury, were determined. Elevated levels of both serum ALT ( Figure 6G) and expression of hepatic ALT ( Figure 6H) were observed in SCI-ConD mice as compared to sham-ConD mice. Serum ALT was increased in sham-HFD mice compared to sham-ConD mice but was not different between SCI-HFD and SCI-ConD mice ( Figure 6G). Expression of hepatic ALT mRNA was increased in sham-HFD mice compared to sham-ConD mice, but not changed in SCI-HFD mice when compared to SCI-ConD mice ( Figure 6H). Taken together, these findings suggest that SCI may trigger inflammatory responses marked by increased production of cytokines from adipose tissue and enhanced numbers of macrophages migrating to the liver, which could result in hepatocellular injury and liver dysfunction, as well as contribute to the general metabolic perturbations observed after SCI.
SCI Resulted in Tissue Specific Reduction of Insulin Signaling-Related Gene Expression
Because there are links among FGF21, adiponectin signaling, hepatic inflammation and insulin action (22,25,27,36,59), we determined whether the dysregulated FGF21 and adiponectin signaling observed after SCI were associated with altered mRNA expression of selected components of the insulin signaling pathway. As compared to sham-ConD, SCI-ConD mice demonstrated reduced expression of IRS-1 mRNA in iFAT and oFAT ( Figure 7A), as well as in liver ( Figure 7C). When comparing SCI-ConD and sham-ConD mice, decreased Glut4 mRNA expression was seen in iFAT and oFAT ( Figure 7B). It is appreciated that muscle is the primary site of glucose uptake and disposal. In gastrocnemius muscle, expression of insulin receptor mRNA level was reduced in SCI-ConD mice as compared to sham-ConD mice ( Figure 7E). However, there was no significant difference in Glut4 or IRS-1 mRNA in gastrocnemius muscle from SCI-ConD as compared to sham-ConD mice, and these parameters were not altered by HFD (Supplemental Data, Figure 1). To further evaluate how impaired FGF21 and adiponectin pathways impact insulin signaling, glycogen content was measured in mouse liver and skeletal muscle. There was a significant reduction in glycogen content in gastrocnemious muscle in SCI-ConD compared to sham-ConD mice ( Figure 7F); muscle glycogen levels were not altered by HFD in either sham or SCI mice. In contrast, SCI with or without HFD had no effect on glycogen levels in liver ( Figure 7D).
DISCUSSION
The central question addressed in this study was whether SCI per se has an impact to change serum FGF21 levels or activity of FGF21-dependent signaling, including that through adiponectin, in a mouse model of complete spinal cord transection. To better understand the meaning of changes in levels of these regulatory factors, the analysis employed a broad approach that examined signaling in liver, iFAT, oFAT and skeletal muscle which, collectively, are responsible for most effects of insulin on blood glucose levels, with adipose tissue and liver also playing key roles in synthesis, storage and release of fats. The principal conclusions of our work herein are that SCI reduced serum FGF21 and adiponectin levels, including total and the more bioactive HMW moiety, and the expression of hepatic FGF21 and adipocyte expression of adiponectin. The data suggest that SCI reduced signaling in adipocytes in response to FGF21 via KLB and FGFR1, and diminished adiponectin action on the liver through reduced hepatic AdipoR2 expression and decreased PPARa activation. In addition, SCI increased levels in liver of inflammatory cytokines, including IL-1b, IL-6 and CD11b, as well as in both iFAT and oFAT of TNFa mRNA, the changes were exacerbated by HFD suggesting that these adipose tissue deposits contribute to the state of low-grade chronic inflammation reported after SCI (58), which was associated with elevated liver expression of ALT mRNA and serum levels of ALT, indicative of hepatocellular injury. Moreover, SCI reduced insulin receptor in gastrocnemius muscle, and reduced mRNA levels for IRS-1 and Glut-4 in iFAT and oFAT, and IRS-1 in liver suggesting impaired insulin signaling pathway after SCI. Evidence of perturbations in regulation of synthesis of steroids, fatty acids and fatty acid metabolism after SCI included greatly elevated HMGCR mRNA expression and ACCa activity in liver after SCI. During administration of a HFD, serum FFA rose in SCI mice on HFD, but not control mice, suggesting impaired ability to store FFA as triglycerides in adipocytes or to use FFA as fuel in muscle and other tissues. FABP4 mRNA was elevated in liver of SCI-HFD mice which is notable because serum levels of the corresponding protein are linked to T2DM and other disorders (52). Taken together, the data indicate defects in metabolism of fats and carbohydrates in multiple tissues after SCI and implicate profound dysregulation of the FGF21adiponectin metabolic regulatory circuit as one key driver of these many perturbations. Hepatic FGF21 expression is regulated by many factors that include nutrients, physical activity and medications (17,59). Reasons that hepatic FGF21 expression is reduced by SCI are unclear. Because of the key role played by PPARa in activation of FGF21 transcription in liver, the reduced nuclear levels and transcriptional activity of PPARa provide insight with respect to one possible mechanism by which FGF21 mRNA levels are reduced in liver after SCI (60), but other possibilities exist. While it remains uncertain how exercise increases FGF21, the linkage between physical activity and FGF21 suggests that just as exercise raises levels of this hepatokine (61, 62), a sedentary lifestyle may lower them. The possibility should also be considered that cues from hepatic or adipocyte inflammation, which was present in both adipose tissues (iFAT and oFAT) and liver, as suggested by the elevated expression of mRNA levels for inflammatory cytokines, particularly for TNFa in the adipose tissue, and IL-1b, IL-6 and CD11b mRNA in the liver, could lead to reduced hepatic FGF21 expression. These data suggest that an increased focus on derangements in liver function as a key factor in impaired fat and carbohydrate metabolism after SCI should be considered. The reduction in adipocyte expression of adiponectin mRNA and serum adiponectin level observed in spinal cord transected mice in the current study may be linked in part to the lower level of serum FGF21 observed after SCI; the data also suggest resistance to FGF21 actions based on lower expression of FGFR1 and KLB in iFAT and oFAT cells. While controversial, some evidence for FGF21 resistance has been presented (63). Other explanations cannot be excluded from the data available and include reduced capacity for PPARg to activate adiponectin transcription as a result of local pro-inflammatory signals (49).
Whatever the mechanisms responsible for reduced serum adiponectin after SCI might be, they were coupled with several molecular changes in liver that might be interpreted as indirect evidence of resistance to adiponectin action. These include lower levels of AdipoR2 mRNA and PPARa activity. While not extensively studied, the concept of adiponectin resistance has been proposed previously and some experimental support for its existence has been presented (64).
Lower expression in iFAT and oFAT of leptin and reduced serum leptin levels were also observed. These changes are consistent with a prior report of low serum leptin after a contusion SCI in rats (65). Implications of these changes are unclear but it should be noted that leptin has been implicated in multiple biological responses including of FGF21 action (33,38). Regulation of leptin expression remains incompletely understood, but it is clear that leptin levels correspond to adipose tissue mass and are reduced by low calorie intake (66). Thus, the low mass of adipose tissue deposits in the SCI mice (44) observed in this investigation may explain, at least in partial, the low leptin levels observed.
Serum FFA levels were significantly elevated in HFD-fed SCI mice compared to SCI-ConD mice but not in sham-HFD mice. These observations suggest impaired ability of SCI mice to clear FFA from the circulation or to metabolize FFA such that they accumulate in blood. Elevated serum FFA levels have been shown to reduce tissue insulin sensitivity through multiple mechanisms that include reduced glucose uptake by skeletal muscle and increased TNFa release by liver Kupfer cells (67,68), Importantly, elevated FFA levels are a risk factor for insulin resistance that may be independent of total fat mass (56). Elevations in serum FFA in SCI-HFD mice were associated with increased hepatic expression of FABP4. FABP4 is thought to have numerous functions that include its ability to bind free fatty acids and retinoids within cells, to participate in uptake of FFA from the extracellular space, and to route FFA to nuclear PPAR receptors to appropriately regulate the gene expression programs they control (69,70); the implications of elevated hepatic FABP4 following SCI fed with a HFD are uncertain, but it is noteworthy that elevated serum levels of the protein encoded by this gene showed a positive correlation with high-sensitivity C-reactive protein (a measure of systemic inflammation), lipid parameters, and the homeostatic model assessment of insulin resistance (HOMA-IR) (56,(69)(70)(71). Taken together, these findings indicate that following SCI, mice were less tolerant of a HFD, and that this type of diet elicited several significant and deleterious changes. Effects of SCI on the serum lipid profile have been evaluated in several other studies. Elevated serum triglycerides were observed at one month after SCI in T3-transected rats associated with increased visceral adiposity (72). In a separate study, serum FFA and glucose were elevated 42 days after a 200 kdyne contusion injury in rats (73). While extensive study of serum lipid profiles has been performed in patients with SCI (1), to date these studies have not characterized serum FFA levels.
An unexpected finding was the upregulation in SCI-ConD mice of HMGCR, a rate-limiting enzyme in the synthesis of cholesterol, possibly reflecting increased tissue demand, which might be expected during reparative responses in the spinal cord or to support neuroplasticity. In SCI patients, abnormalities of lipid metabolism include reduced HDL cholesterol (1, 2) and elevated triglycerides in response to a high fat challenge when compared to data for historical controls (74). In rats that were studied 1 month after thoracic level-10 transection, serum triglycerides, total and HDL cholesterol levels tended to be lower as compared to sham-operated controls, although this difference was not significant (72). Further study is needed to understand how SCI impacts metabolism of fats or how SCI perturbs the networks of adipokines and hepatokines that regulate them.
The data presented suggest that at 84 days after spinal transection, there is accumulation of tissue macrophages in the liver, as reflected by increased expression of inflammatory cytokines, including IL-1b, IL-6 and CD11b. There is also some evidence of inflammation in fatty tissue based on elevated expression of TNFa mRNA in both iFAT and oFAT, and increased IL-1b mRNA in oFAT. These alterations would be expected to stress hepatocytes, impair insulin action on hepatocellular regulation of glucose uptake and glucose output and, possibly, injure hepatocytes as well. Evidence of hepatocellular impairment, and possibly injury, was observed in post-SCI mice based on elevated circulating levels of ALT, as well as increased hepatic ALT expression. It is noteworthy that Kupffer cells, the resident hepatic macrophages, have been linked to NAFLD, a risk factor for insulin resistance (75), thus indirectly implicating the changes in CD11b to liver disease and insulin resistance after SCI. Elevated cytokine expression in fat deposits is also thought to be a risk factor for impaired glucose homeostasis (76,77), suggesting that such alterations in adipose tissues observed in the current study may further contribute to impaired glucose handling and insulin action in SCI mice. This postulate is indirectly supported by findings of reduced expression of several key genes in insulin signaling in iFAT, oFAT, liver and gastrocnemius muscle. However, the data do not establish causal relationship between TNFa expression in adipose tissue and altered fat or carbohydrate metabolism. Also unclear is the cellular source of cytokines in adipose tissues, but prior studies indicate that infiltration of fat deposits by tissue macrophages is the primary explanation for such changes (76). Evidence of hepatic inflammation and hepatocellular injury that increases with time to 21 days after SCI has been observed in rats following a 200 kdyne contusion SCI. In the same study, elevations in serum ALT were observed in rats with cervical (C5), thoracic (T8) and lumbar (T12) contusion injuries (11). In that study, hepatic mRNA levels for IL-1ß increased at days 3 and 7 post-SCI while TNFa increased at 14 and 21 days, suggesting a time-dependent response. Hepatic lipids, assessed by Oil red O staining, were also increased at 14 days post injury (11). A more recent study from the same group found elevated hepatic CD11b at 6-weeks after a 200 kdyne injury at T8 that was associated with increased mRNA levels for several cytokines, including TNFa and IL-1ß, and increased hepatocellular lipids and iron levels (73). These findings are in agreement with the increased expression of CD11b in liver in the current study. However, no significant alteration in hepatic expression of cytokines was observed in the present study. Reasons for these discrepant findings are unclear but may include differences in time since injury, species and/or responses to contusion versus transection SCI.
It was previously reported that, in mice, spinal cord transection results in marked decreases in weights of iFAT and oFAT and that HFD did not increase weights of these fat deposits in SCI mice (44). Similarly, a reduction in total fat mass in rats after transection at thoracic level-4 has been reported and was associated with a small increase in metabolic rate and food intake (78). There is insufficient data to determine why the observed decreases in fat depots occurred. It has been argued that because body weights of the mice from which tissues were obtained for the current study increased over time after SCI that their net caloric intake exceeded metabolic demands. However, it remains unclear why gains in body weight were not associated with a return toward normal adipose tissue mass; possible explanations include impaired ability of adipocytes to take up and store lipids or, possibly, a loss of the normal replacement of aging adipocytes from the pool of adipocyte progenitors such that adipose tissues become depleted of healthy and fully functional adipocytes. Similarly, the reduced levels of PPARg and ASC-1 observed in the current study could simply reflect low fat mass. Adipocytes undergo turnover whereby there is constant replenishment of aging or damaged adipocytes ones formed from a committed pool of progenitors (79). Thus, failure of HFD to promote gains in adipose tissue mass may indicate damage to adipocytes and/or inadequate replenishment of adipocytes from pools of progenitors. Of note, it was reported that FGF21 knockout in mice displayed disorders of adipose tissue due to defects in PPARg signaling and PPARg-dependent gene expression (34,80). This report demonstrates that it is plausible that the reduced levels of FGFR1 and PPARg activity observed in iFAT and oFAT could result in dysfunction of adipose tissue, low adipose tissue mass and impaired adiponectin expression and release.
It is now accepted that fat deposits are highly specialized in their functions depending on their location. oFAT is particularly important when considering the pathogenesis of insulin resistance and diabetes, perhaps because adipokines and free fatty acids liberated by oFAT are sent to the liver by the portal circulation. Unexpectedly then, in the present study, gene expression changes in iFAT and oFAT from SCI mice, and responses in SCI mice of these fat deposits to HFD were similar in magnitude and direction suggesting that physiological adaptations to SCI in mice overcame phenotypic differences related to localization of these fat deposits.
It was previously reported that HFD induced glucose intolerance in spinal cord transected mice at 84 days after injury and that fasting glucose was modestly increased in SCI-ConD mice at this time point (44). When these findings are taken together with the ability of FGF21 to improve insulin action, and low FGF21 observed in SCI in mice, it was of interest to examine insulin signaling pathways in tissues from these SCI mice. Whereas no reduction in IRS-1 or Glut-4 mRNA was observed in skeletal muscle, reduced levels of IRS-1 mRNA were observed in iFAT and oFAT, as well as liver, and these changes were not influenced by administration of a HFD. IRS-1 is critical to coupling insulin receptor and downstream signaling pathways while Glut4 represents the primary mechanism for insulininduced glucose uptake. The findings suggest that, in the spinal cord transected mouse model at least, loss of insulin responsiveness of liver and fat deposits is an important mechanism leading to insulin resistance and diabetes. However, as discussed above, the data do not exclude a possible deterioration of function of adipose tissue as an explanation for the uniform decrease in insulin signaling pathways genes in adipose tissues. Further studies are necessary to better understand the respective roles of fat, liver and muscle in development of T2DM after SCI.
A limitation to be considered when interpreting our findings or considering their translation is that the approach employed did not directly test for physiological significance of lower FGF21 levels in the SCI mouse model used. Consequently, although it is attractive to posit that the changes observed downstream of serum FGF21 may be attributable to the lower serum FGF21 levels observed after spinal cord transection, this mechanistic linkage is yet to be proven. It should also be noted that whereas persons with SCI usually gain body fat mass, mice with spinal cord transection lost fat mass (44). This difference may limit how well changes in the spinal cord transection mouse model used in the current model reflect what occurs in individuals with SCI. Another limitation of these studies is that they evaluated only male mice; males were selected because male C57B6 mice are more sensitive to high-fat-diet-induced glucose intolerance than females. We expect that the pattern of changes in serum and hepatic FGF21 levels would be the same between males and females after SCI although some gender effects on the magnitude of changes cannot be excluded.
In summary, this study, the impact of SCI on expression of and signaling by FGF21 was investigated. Major findings of the study are that at 84 days after a complete spinal cord transection, hepatic FGF21 expression and serum FGF21 protein were reduced, associated with lower serum total and HMW adiponectin and leptin. A HFD raised serum FFA in SCI but not sham mice suggesting impaired ability for tissue to take up or metabolize fats. Prior studies demonstrated glucose intolerance in the SCI-HFD mice model that was used for these studies. Decreased expression for insulin signaling-related genes were observed in adipose tissue and liver but, surprisingly, not skeletal muscle. Taken together, reduced levels of FGF21 and adiponectin provide attractive mechanisms for the links between SCI and insulin resistance and T2DM. Moreover, the findings suggest key roles for perturbed function of adipose tissue and liver in the pathogenesis of T2DM after SCI. Further research will be needed to formally prove a causal relationship between abnormal levels of these proteins and carbohydrate and fat metabolism after SCI.
DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
ETHICS STATEMENT
The animal study was reviewed and approved by Institutional Animal Care and Use Committee at the James J. Peters VA Medical Center.
AUTHOR CONTRIBUTIONS
X-HL designed the study, carried out the research, analyzed, interpreted the results, and wrote the manuscript. ZG designed the study, carried out the research and edited the manuscript. LH carried out the research (animal studies). JP carried out the research. DA carried out the research. WB designed the study, analyzed, interpreted the results, edited the manuscript. JY analyzed, interpreted the results, edited the manuscript. CC designed the study, analyzed, interpreted the results, and wrote/ edited the manuscript. All authors contributed to the article and approved the submitted version.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2021-03-11T14:09:39.532Z", "created": "2021-04-20T00:00:00.000", "id": "232171388", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.frontiersin.org/articles/10.3389/fendo.2021.668984/pdf", "pdf_hash": "639e5dabe242c1a654b175fd3535b57e9ff7944a", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:31323", "s2fieldsofstudy": ["Medicine", "Biology", "Environmental Science"], "sha1": "8580576f77c08e93928fdd9a53c0c0108a380575", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Oncolytic effect of wild-type Newcastle disease virus isolates in cancer cell lines in vitro and in vivo on xenograft model
Oncolyic virotherapy is one of the modern experimental techniques to treat human cancers. Here we studied the antitumor activity of wild-type Newcastle disease virus (NDV) isolates from Russian migratory birds. We showed that NDV could selectively kill malignant cells without affecting healthy cells. We evaluated the oncolytic effect of 44 NDV isolates in 4 histogenetically different human cell lines (HCT116, HeLa, A549, MCF7). The safety of the isolates was also tested in normal peripheral blood mononuclear (PBMC) cells. The viability of tumor cell lines after incubation with NDV isolates was evaluated by MTT. All cell lines, except for normal PBMC primary cells, had different degrees of susceptibility to NDV infection. Seven NDV strains had the highest oncolytic activity, and some NDV strains demonstrated oncolytic selectivity for different cell lines. In vivo, we described the intratumoral activity of NDV/Altai/pigeon/770/2011 against subcutaneous non-small cell lung carcinoma using xenograft SCID mice model. All animals were responsive to therapy. Histology confirmed therapy-induced destructive changes and growing necrotic bulk density in tumor tissue. Our findings indicate that wild-type NDV strains selectively kill tumor cells with no effect on healthy PBMC cells, and intratumoral virotherapy with NDV suppresses the subcutaneous tumor growth in SCID mice.
The oncolytic potential of NDV strains circulating in wild migratory birds of Russia remains poorly understood. Here we describe oncolytic wild-type NDVs from natural reservoirs obtained in 2008-2014 in Russia. We report the rejection of excessive attenuations and the usage naturally occurring NDV strains. Oncolytic properties were determined using 4 tumor cell lines of various histogenesis. We demonstrate the in vitro ability of NDVs to influence the viability of tumor cells after infection and evaluate in vivo efficiency of NDV strain against non-small cell lung carcinoma.
Viruses
Newcastle disease virus was isolated from wild migratory birds in eight administrative regions of the Russian Federation: the Altai Territory, the Novosibirsk Region (Western Siberia), the Republic of Tyva (Eastern Siberia), the Amur Region, the Kamchatka Territory, the Republic of Sakha (Yakutia), the Sakhalin Region (Far East) and the Republic of Adygea (Southern Federal District). A total of 44 wild-type NDV isolates were collected in Siberia and the Far East of the Russian Federation in 2008-2014.
Cytotoxicity of NDV strains in human peripheral blood mononuclear cells (PBMC)
The selected strains represented different NDV pathotypes: NDV/Yakutiya/mallard/852/2011 (852)-mesogenic pathotype with the typical avirulent type F-gene sequence [27], NDV/Altai/ pigeon/770/2011 (770)-mesogenic pathotype with the typical virulent type F-gene sequence [28] and Adygea/duck/12/2008 (AD)-velogenic pathotype [29]. There were no changes in viability of suspended PBMCs even after 4 days of infection with different NDV strains. The viability ranged from 94% to 110% of that of controls. NDV-infected PBMC cell culture had no visible morphological disorders compared to controls after an hour of viral exposure and on the following days of cultivation. MTT assay also shows that the strains have no toxic effect on PBMCs because of unchanged cell viability after infection (Fig 1). Thus, we demonstrated that viral strains were safe for human cells regardless of pathotype.
In vitro cytotoxicity
The cytotoxic properties of Newcastle disease virus isolates were assessed using MTT assay at 540 nm in four tumor cell lines: A549, MCF7, HeLa and HCT116. The MTT assay gives an idea of the metabolic activity of the cells being studied, which allows one to estimate the specific cell death after infection with the virus. Cells were grown on plates for one day and infected with viral dilutions of 2, 8 and 16 HAU per 10.000 cells. The assay was performed on 4 th day after cell monolayer infection. Fig 2 show MTT assay results for HCT116, HeLa, A549 and MCF7, respectively.
The following NDV strains had the most pronounced oncolytic effect (less than 30% of viable cells) on the 4 th day after infection with a dose of 16 (Fig 3B).
The cytotoxic effect of NDV isolates on A549 cell line (human non-small cell lung cancer). A total of 17 NDV isolates effectively eliminated tumor cells (up to 60% of tumor cells remained viable). These isolates included the above mentioned 11 and 6 additional isolates: The cytotoxic effect of NDV isolates on MCF7 cell line (human breast adenocarcinoma). Human mammary adenocarcinoma cells MCF7 proved the most resistant to all NDV isolates. MTT assay showed a relatively small decrease in cell viability after incubation with viruses (from 80% to 100% of cells remained viable) (Fig 2D).
The following isolates showed the greatest antitumor activity in MCF7 cells on the 4 th day after infection with a dose of 16
In vivo virotherapy
NDV/Altai/pigeon/770/2011 was chosen for virotherapy as a strain with the highest oncolytic activity in vitro and mesogenic pathotype. Previous studies suggest that more virulent NDV Oncolytic effect of wild-type NDV isolates strains may have improved oncolytic efficacy [30], which makes NDV/Altai/pigeon/770/2011 a promising agent.
Cytotoxicity experiments with tumor cell lines demonstrated a wide range of cell susceptibility to NDV infection. The number of viable non-small-cell lung carcinoma A549 cells decreased to less than 40% after NDV/Altai/pigeon/770/2011 infection, according in vitro test.
Here we hypothesize that mesogenic strain may have high anticancer efficiency in vivo in a xenograft model of A549 cells.
To determine the efficacy of NDV virotherapy in vivo, SCID mice were injected subcutaneously with 3.5 × 10 6 A549 tumor cells. After 20 days, when an average tumor diameter reached 4 mm, tumor-bearing mice received a course of intratumoral injections of NDV/Altai/pigeon/ 770/2011 during 4 days. Control mice received a series of PBS injections during 4 days.
Therapeutic efficacy of pigeon NDV strain in tumor-bearing mice. A549 tumor-bearing mice were infected with NDV/Altai/pigeon/770/2011 when tumor size reached 4 mm in diameter. Fig 5 shows that weight dynamics was the same in three groups of mice and controls (+) with an increase and a slight decrease of body weight. There was therefore no effect of viral injections on weight dynamics. Controls (+) had a lower weight due to the absence of tumors. Tumor nodules in PBS group were larger than those of virotherapy group (Fig 6).
On 20 th day (after A549 tumor inoculation), the average tumor size in experimental and PBS groups was 53.21±10.45 mm 3 and 59.12±8.45 mm 3 , respectively. Mice received NDVbased virotherapy and PBS on 20-23 th days of the experiment. The Fig 6 shows the increasing size of tumor node during the virotherapy and PBS injections, which may be related to the beginning of active tumor growth and inflammatory response to mechanical damage of tumor by injections. However, NDV/Altai/pigeon/770/2011-based virotherapy resulted in significant inhibition of tumor node growth throughout the experiment compared to control group. As soon as on 5 th day after virotherapy (28 th day of experiment), the average tumor size in the experimental group was 2.5 times less than in the control group (175.27±40.38 mm 3 and 439.43±35.62 mm 3 , respectively). On 10 th day (33 rd day of the experiment), there was a 4.5-fold difference in the average size of tumor nodes in the experimental and control groups (of 122.01±32.38 mm 3 and 560.37±53.42 mm 3 , respectively), and by the end of the experiment (15 th day after the course of virotherapy and 38 th day of the experiment), there was a 4.2-fold difference (247.70±63.43 mm 3 and 1032.72±100.38 mm 3 , respectively) ( Fig 7) (P <0.05).
NDV induces significant tumor necrosis in non-small cell lung cancer in vivo.
To determine the oncolytic potential of NDV/Altai/pigeon/770/2011 in vivo, we sacrificed animals at various time points after virotherapy (5, 10, and 15 days (28, 33 and 38 days of tumor growth, respectively)) and measured tumor virotherapy response and structural changes in tumor tissue. Histological examination of tumor node sections demonstrated increasing tumor volume and hemorrhages. The tumor nodes were restricted to connective tissue capsule. By 5 th day post-therapy, there were almost no necrotic areas in tumor tissue. Necrotic changes appeared later. Morphometric analysis of necrotic areas revealed increasing tumor necrosis, which peaked on 10 th day after therapy (Fig 8). On 10 th and 15 th days, necrotic areas with destructive swelling were often found. Structurally, destructive edema sites and necrotic foci accounted for 54.13% and 73.89% of tumor bulk density on 10 th and 15 th day after virotherapy, respectively.
Discussion
NDVs are not pathogenic for mammals and have natural oncolytic properties, which makes them promising anticancer agents. Wild-type NDV strains are safer and more cost-effective than modified recombinant strains requiring additional measures for genetic control of viral products. Promising in vitro and in vivo findings on several attenuated wild-type NDV strains such as MTH-68/H [31] and HUJ [32,33] were followed by encouraging results of clinical trials [34][35][36].
Here we investigate the oncolytic properties of 44 wild-type NDV isolates obtained from migratory birds in Russia in Western Siberia and the Far East from 2008 to 2014. Most of the Oncolytic effect of wild-type NDV isolates viruses were isolated from Anseriformes and proved avirulent and lentogenic except one velogenic and several mesogenic isolates including NDV/Altai/pigeon/770/2011 [28].
MTT-based viability testing of tumor cells after incubation with NDV isolates demonstrated the oncolytic properties of the isolates in cell lines of different etiology and histogenesis. Tumor cell lines had different sensitivities to NDVs. The following cell lines were most sensitive: non-small-cell lung cancer A549 cells (17 isolates had anticancer effect) and cervical cancer HeLa cells (11 isolates had anticancer effect). Mammary adenocarcinoma MCF7 cells were least sensitive, and cell viability after incubation with the viruses remained high.
Among the 44 wild-type NDV isolates, tested for oncolytic potential in the present work, The oncolytic properties of natural NDV isolates were dose-dependant. The viability of tumor cell lines decreased averagely by 5-10% with increasing virus dose (from 2 to 16 HAU per 10.000 cells). The oncolytic effect of the virus was independent from the etiology of tumor cells and was the same for colorectal cancer, cervical cancer or non-small-cell lung carcinoma.
However, specific NDV strains demonstrated different oncolytic properties with specific tumor cell lines. Several NDV strains were effective against one cell line and failed to eliminate others. For example, NDV/Yakutiya/mallard/852/2011 and NDV/teal/Novosibirsk_region/ 320/2010 were effective against A549 cells but had a poor performance with HeLa and HCT116 cells.
Oncolytic potential of all the 7 isolates is slightly different in relation to different tumor cell lines, however, the most pronounced antitumor effect observed for all studied cell lines, was observed for strain NDV/Altai/pigeon/770/2011 -from 20 to 40% of viable cells HCT116, A549 and HELA after incubation with virus dose 16 HAU per 10.000 cell, and about 80% for the resistant lines MCF7.
Numerous studies focused on investigations of mechanisms of viral-induced death in tumor cells. There is data that NDV can trigger not only apoptosis in the transformed cells, but also autophagy [37] and necrosis [15]. In the A549 cells, when infected with strains of different virulence (LaSota-lentogenic, Beaudette C-mesogenic, FMW-velogenic), activation of caspase-9 and caspase-8 was shown, at the same time interval and regardless of the pathotype of strain [38]. There is also evidence of the activation of the MARK signaling pathway during NDV-induced apoptosis in A549 cells, in which pathways are activated through ERK, JNK and p38 kinases, but the activation of p38-MAPK kinase plays a major role, regardless of the pathogenesis of the NDV strain [38]. Apoptosis can be induced by the matrix (M) protein of the Newcastle disease virus, located on the inner surface of the viral envelope. The proapoptotic effect of the M protein was demonstrated by the infection of the HeLa [39] and HCT116 and HT29 [40] cells with the Malaysian strain AF2240. The Newcastle disease virus can lead to the destruction of tumor cells by the mechanism of necrotic death [14,15].
For a more detailed study of differences in the implementation of the oncolytic properties of natural isolates NDV further work is needed to explore possible differences in the mechanisms of penetration of viruses in tumor cells of different etiology and histogenesis, replication and release of infectious progeny, to evaluate the influence of genetic alterations of tumor cells on the program of oncolytic properties in virotherapy. For example, it was shown that the Raspathways activation is required for a successful replication of NDV, namely, its branches, related small GTP-Rac1 basics [41]. However, in the present study, despite the activation of Ras-pathways in the cells of tumor lines MCF7 (amplification of the Her), the implementation of anticancer potential in this line was not fully achieved with any studied strain.
In next part of study we have focused our attention on the ability of the virus to suppress tumor growth in vivo. To evaluate the effect of the virotherapy on the tumor progression and on the changes in the structure of tumor tissue after the viral treatment, we investigated the direct oncolytic influence of the course of intratumoral injections on the xenograft model of human non-small cell lung A549 carcinoma. Referring to the experience of carrying out the in vivo experiments described in the scientific literature on different tumor models and different strains of viruses, an amount of A549 carcinoma cells, dose of virus, volume of injection and method of administration were selected.
As was described in studies, virulent (mesogenic and velogenic) strains are able to induce more powerful apoptotic response in tumor cells and at very early stages of virus-cell interaction, in contrast to apatogenic strains and strains with reduced virulence [38,42]. To choose one oncolytic strain to carry out in vivo experiment we decided to use mesogenic NDV/Altai/ pigeon/770/2011. The usage of much more virulent strains in virotherapy can be more effective strategy for successful tumor treatment. For example, Ghrici M. et al. showed activation of caspase-8 already in 2 hours after infection of MCF7 cells with strain AF2240, in contrast to the data presented by Elankumaran S. et al. [43] and Ravindra P. [44], where activation of caspase-8 was demonstrated only after 48 and 24 hours, respectively. Such a temporal spread of caspase-8 activation could be associated not with the difference in the order of activation of the external and internal pathways of apoptosis, but probably with the dependence of the rate of apoptosis on the virulence of the strain. This time difference in activation of death program can be important for suppression tumor progression.
According to the analysis of the growth dynamics of the tumor node, we showed that in the control group of mice receiving injections of the PBS, the tumor process developed and progressed, while in the group that received virotherapy with NDV/Altai/pigeon/770/2011, the tumor growth rate decreased. Moreover, virus contributed to a decline of tumor growth at the initial stage after virotherapy (day 5) with an increase of the antitumor effect by 10 days. It is worth noting that, at the end of the experiment, there is an insignificant increase in the average volume of the tumor node. However, this growth is insignificant in comparison with the growth rates in the control group.
Tumor nodes from animals treated with virotherapy were much denser than the nodes from the control group of animals. These results suggest that there is the prevalence of fibrosis in experimental tumors after virotherapy. However, such a relationship was not revealed, because the bulk density of the connective tissue of experimental tumors was higher only on the 5 th day. At subsequent time points (10 and 15 days), the bulk density of the connective tissue was more than 3 times lower than in the control.
On the other hand, the dynamics of changes in the bulk density of connective tissue could indirectly suggest the increase in the volume density of necrosis with areas of destructive swelling, the numerical density of which increases from 5 to 15 days. With the intensification of necrotic processes, the volume density of the connective tissue also decreases. However, it should be noted that necrotic processes in the control group receiving injections of the PBS also increase in dynamics, that is probably due to the rapid growth of the tumor node and the formation of foci of ischemia.
Despite the fact that necrosis is considered to be an unfavorable development of destructive changes in tissue, necrosis can also pass under the control of cellular signaling pathways, being, like apoptosis, a process of regulated cell death. In vivo study with recombinant rNDV/ F3aa (L289A) showed that morphometric analysis in the tumor tissue of hepatocellular carcinoma after NDV administration through the hepatic artery revealed the increase of necrotic changes in tumor tissue. However histological studies of healthy tissue surrounding the tumor showed no hepatotoxicity-there was no syncytium formation (typical for NDV infection) in the surrounding tissue and no inflammatory infiltrates [15].
Obviously, the above results suggest the prospect of using the virus of Newcastle disease as an antitumor agent affecting tumor cells and tissues, causing their death. The data obtained suggest the NDV/Altai/pigeon/770/2011 strain for further investigation as an oncolytic agent whose antitumor mechanism of action requires further research.
In summary, we have characterized the ability of wild-type Newcastle disease virus to demonstrate strong oncolytic effects in direct lysis of human tumor cell lines of various histogenesis. Normal PBMC culture seems to be resistant. The efficacy of virotherapy of subcutaneous non-small cell lung A549 carcinoma in SCID-mice was demonstrated. Intratumoral course of NDV leads to inhibition of tumor progression, which is reflected in the dynamics of tumor nodule size by direct lytic effect bypassing the activation of immune system. Further work is necessary to investigate mechanisms of virus-induced cell death and the wild-type NDV/Altai/ pigeon/770/2011 as the anticancer agent.
Conclusion
We demonstrate possibility to use wild-type Newcastle disease virus isolated from birds in approach for cancer therapy. There may be lots of different recombinant oncolytic strains under investigation but we believe that it is important to propose a perspective for study the wild-type NDV strains for developing oncolytic anticancer agent. Based on the our results, such strains can be promising biological agents for future cancer therapy.
Viruses
Newcastle disease virus isolates were gathered from wild birds within their migratory routes in the territory of Siberia and the Far East, Russia (Table 1) No specific permission was required for sample collection from wild birds killed by local hunters in compliance with the Russian Federation hunting laws. Also no permit was needed for sampling from wild birds captured alive during ringing activities in Biostations with collaboration with other research institute as part of the national avian influenza surveillance.
The original stocks were amplified in the first passage using 10-day-old embryonated eggs from non-vaccinated specific-pathogen free chicken according to the WHO recommendations [45]. After 72 hours post inoculation, the virus containing allantoic fluid was harvested and partially purified through 0.45 μm porous filters and stored at -80˚C. The presence of Newcastle disease virus in the allantoic fluid was determined by the hemagglutination reaction and RT-PCR with specific primers [46]. Based on the obtained pathogenicity test results-mean death time (MDT) and intracerebral pathogenic index (ICPI), it can be concluded that the 44 NDV investigated isolates are avirulent, with the exception of the several lentogenic strains and virulent strains-NDV/Altai/pigeon/770/2011 [28], NDV/Altai/pigeon/777/2010 [29] and NDV/Yakutiya/mallard/852/2011 [27], which belongs to the group of mesogenic strains, and the strain NDV/Agydea/Duck/12/2008 [29], belonging to the group of highly pathogenic velogenic strains.
Cell lines and culture conditions
We studied 4 human cancer cell lines of various etiologies. Human colon carcinoma cell line HCT116 was obtained from State Research Center of Virology and Biotechnology «Vector» (Novosibirsk region, Koltsovo, Russia), human breast adenocarcinoma cell line MCF7 was obtained from Russian collection of cell lines of vertebrates of the Institute of Cytology Russian Academy of Sciences (Saint-Petersburg, Russia), human epitheloid cervix carcinoma HeLa was kindly donated by Institute of Clinical and Experimental Lymphology (Novosibirsk, Russia), human non-small lung carcinoma cell line A549 was a kind gift from Belogorodstev S., Research institute of fundamental and clinical immunology (Novosibirsk, Russia). All cell lines were grown in Dulbecco's modified eagle's media (DMEM) (Gibco Inc., UK) with 1% Oncolytic effect of wild-type NDV isolates antibiotic (penicillin, streptomycin, gentamicin) solution and 10% heat-inactivated fetal bovine serum (Gibco Inc., South America) and were maintained at 37˚C in an incubator with 5% CO 2 .
Human peripheral blood mononuclear cells (PBMC) were selected as control healthy nontransformed cells. PBMCs were isolated from blood of conditionally healthy donors by the method of sedimentation in a ficoll density gradient. The blood sampling protocol was approved by the Committee on Biomedical Ethics of Research Institute of Experimental and Clinical Medicine, Novosibirsk (Permit Number: 15). Written informed consent was obtained prior to the study from the human volunteer. Peripheral blood mononuclear cells were cultured in RPMI medium with 10% heat-inactivated fetal bovine serum (Gibco Inc., South America) and with 1% antibiotic (penicillin, streptomycin, gentamicin) solution at 37˚C and in an atmosphere with 5% CO 2 .
Cytotoxicity assay
Tumor cells were plated at 3 × 10 4 cell per well in 96-well plates in 0.2 ml of MEM supplemented with 1% heat-inactivated FBS with 1% antibiotic (penicillin, streptomycin, gentamicin) solution. After 24 hours, tumor cells were infected with NDV isolates at different doses-2, 8 and 16 HAU per 10.000 cells in 1% FBS maintenance medium. After incubation with virus for 1 hour, virus was removed and 0.2 ml fresh maintenance medium was added. Mock cells were incubated with 1% FBS MEM medium. Cell viability was analyzed using MTT assays at time point 96 hours after NDV infection. The principle of the method is based on the ability of mitochondrial succinate dehydrogenase to transfer the water-soluble yellow MTT salt into violet crystals of formazan, which accumulate in the cytoplasm of cells. On the level of accumulation intensity of these crystals, one can judge the viability of cells in a cellular monolayer. First, cells were washed two times with Hanks solution. Stock MTT solution was prepared in concentration 5mg/ml in phosphate buffered saline (PBS). 10%-MTT working solution was prepared on the basis of a maintenance MEM medium. 0,1 ml of 10%-MTT was added to each well. Tumor cells were incubated with MTT for 4 hours at 37˚C in an incubator with 5% CO 2 and lysed in DMSO (0.15 ml/well) for 1 hour at room temperature in dark place to release crystals of formazan. These result in changing the color of medium and are due to activity. The optical density was measured at a wavelength of 540 nm and 630 nm (background) on a Lon-zaBiotek ELX808 Absorbance Microplate Reader (USA).
To evaluate safety for PBCM equal volumes of the cell suspension in the growth medium were distributed over the tubes. The cells were concentrated by centrifugation 1500g for 5 minutes, the supernatant was removed and incubated with a different dilution of the virus in 1%-FBS RPMI medium with 1% antibiotic from 2 to 32 HAU per 10.000 cells. The virus concentration was increased in 2-fold increments. Cells with the virus were incubated for 1 hour at 37˚C in an incubator with 5% CO 2 , then the cells were concentrated by centrifugation 1500 g for 5 minutes, the supernatant was removed and cells were resuspended in fresh medium and then plated on 96-well plate at 10,000 cells in 100 μl per hole. Control tumor cells were incubated in 1%-FBS RPMI medium with antibiotics without virus for 1 hour.
Three days after infection of the cell lines with NDV isolates, an analysis of the viability of PBCM-cells was carried out using the commercially available CellTiter 961AQueous One Solution Cell Proliferation Assay (Promega, USA). This reagent contains the tetrazolium compound MTS, which in living cells with enzymatic activity is reduced to formazan, staining the culture medium. To each well, 20 μl of reagent was added to 100 μl of the culture medium and incubated for 2 hours at 37˚C in an atmosphere with 5% CO 2 . The optical density was measured at a wavelength of 490 nm.
The percentage of living cells was calculated by the formula: where E is the indices obtained from the virus infected wells, and K is the parameters of the control wells. All viral isolates were analyzed on each tumor cell line in triplicate. The results of measured cell viability are expressed as the percentage of surviving of infected cells versus uninfected mock cells which were considered to be 100% of viability.
Animals and xenograft human non-small lung carcinoma model
Male 7-8-weeks-old SCID immunodeficient mice were purchased from Institute of Cytology and Genetics SB RAS and were housed under special conditions (in sterile, filtered cages) with food and water ad libitum. Animal studies were conducted at the Center for Genetic Resources of Laboratory Animals at the Institute of Cytology and Genetics, Siberian Branch, Russian Academy of Sciences. The number of A549 non-small lung carcinoma cells injected to induce xenograft was chosen after trial experiments with different concentrations of cells. Concentration of 3.5 × 10 6 A549 cells in 100 μl of PBS per mouse proved to be preferable in terms of the developmental and growth dynamics of the tumor node, which makes possible to use tumor-bearing animals in the experiment according to the rules of the bioethical commission. All efforts were made to minimize suffering. During the experiment with SCID-mice we did not find dead mice. Also no mouse was euthanized because of inactivity, significant weight loss, inappetence, piloerection and other symptoms. Experiments protocols were approved by the Committees on Biomedical Ethics of Research Institute of Experimental and Clinical Medicine, Novosibirsk (Permit Number: 15).
Mice were injected subcutaneously into the right flank with 3.5 × 10 6 in 100 μl PBS. Subcutaneously solid tumor node is formed without metastasis. Tumor nodules became visible in place of the A549 on the 7th day among 100% of inoculated animals. The design of experiment ensured efficiency as it is possible with the object of obtaining consistent results with the minimum number of animals and the minimum degree of pain and suffering, according to the European convention for the protection of vertebrate animals used for experimental and other scientific purposes (Strasbourg, 18. III.1986). On 10 th day nodules began to measure metrically. The nodes reached the desired diameter (4 mm) for initiation of the course of virotherapy on 20 th day after inoculation of tumor cells. The dynamic of tumor growth allows to conduct a course of viral treatment and to observe the animals for 20 days after virotherapy.
Newcastle disease virotherapy
Viral treatment was performed with NDV/Altai/pigeon/770/2011 strain. SCID mice were divided into three groups: group1 -control tumor-bearing animals receiving 4 viral dose injections every other day starting 20 day after inoculation of tumor cells. NDV was administrated intratumorally into solid node at a dose of 7 lgTCID50/100μl. Virus titers were determined as TCID50/ml on Vero cell line, obtained from State Research Center of Virology and Biotechnology «Vector» (Novosibirsk region, Koltsovo, Russia) [48]. Group 2-experimental tumorbearing animals receiving 4 injections of 100 μl PBS every other day starting 20 day after inoculation of tumor cells. All volume of 100 μl was injected into the solid tumor by 2-3 several injections from different sides of tumor node to spread the virus into the tissue more effectively. Group 3 -control animals without tumor but receiving injections of the PBS instead of introducing tumor cells and injection of the virus subcutaneously.
On the 10 th after tumor cell inoculation, during virotherapy and after treatment mice were regularly weighed in 2-3 days and observed for tumor growth by measuring of dynamics of changes in the size of the tumor nodes. Five animals were randomly chosen for killing at each of the following time points: day after completion of therapy 5, 10, and 15 (28, 33 and 38 days of tumor growth, respectively). Animals from the control and experimental groups were removed from the experiment by dislocation of the cervical vertebrae. Upon killing, biomaterial of subcutaneous tumor was taken for histological diagnosis. Animals from the group 3 were observed during the whole time of the work and were withdrawn from the experiment on 38 th day of tumor growth.
Tumor growth was monitored by callipers measuring each 2-3 days from the first day of virotherapy till the end of experiment. The measurements of the length and width of the subcutaneous tumor node were calculated by the standard formula and the received tumor volume The animals were placed under observation every 2 days with a check of changes in the state of motor activity, changes in body weight, skin and mucous membranes, food intake and water. Endpoint of the experiment and the last point of biomaterial sampling (38 days of tumor growth) was chosen in according with statement that tumor size must not exceed 15 mm (1.5 cm) at the largest diameter in an adult mouse, because larger tumor may include mobility restriction, the inability to access food and water, pressure on internal organs or sensitive regions of the body.
Histology
In tumor-bearing animals subcutaneous tumor samples were harvested. Tumor tissue were fixed in 4% paraformaldehyde and paraffin embedded. Sections of 4 μm thickness were subjected to hematoxylin-eosin staining (H&E) for histological analysis and Van Gieson staining.
Statistical analysis
For comparison of individual data points, the student t-test was applied to determine statistical significance. P values of <0.05 were considered statistically significant. The results of the evaluation of the viability of tumor cell lines are presented in the form of histograms in percent as the average relative values of the proportion of living cells on the fourth day after treatment with virus to the proportion of control cells not treated with virus, taking into account the standard deviation (mean relative value ± standard deviation). In animal experiment, dynamic of tumor growth is presented by graphic. Statistical data were obtained using Statistica 6.0 software.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2018-04-26T19:49:05.178Z", "created": "2018-04-05T00:00:00.000", "id": "4972806", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0195425&type=printable", "pdf_hash": "bdb77f0b5aaa14f533f968403b3cf5c6fe0e0445", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:34535", "s2fieldsofstudy": ["Medicine", "Biology", "Environmental Science"], "sha1": "bdb77f0b5aaa14f533f968403b3cf5c6fe0e0445", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Transcriptional Effects of Psychoactive Drugs on Genes Involved in Neurogenesis
Although neurogenesis is affected in several psychiatric diseases, the effects and mechanisms of action of psychoactive drugs on neurogenesis remain unknown and/or controversial. This study aims to evaluate the effects of psychoactive drugs on the expression of genes involved in neurogenesis. Neuronal-like cells (NT2-N) were treated with amisulpride (10 µM), aripiprazole (0.1 µM), clozapine (10 µM), lamotrigine (50 µM), lithium (2.5 mM), quetiapine (50 µM), risperidone (0.1 µM), or valproate (0.5 mM) for 24 h. Genome wide mRNA expression was quantified and analysed using gene set enrichment analysis, with the neurogenesis gene set retrieved from the Gene Ontology database and the Mammalian Adult Neurogenesis Gene Ontology (MANGO) database. Transcription factors that are more likely to regulate these genes were investigated to better understand the biological processes driving neurogenesis. Targeted metabolomics were performed using gas chromatography-mass spectrometry. Six of the eight drugs decreased the expression of genes involved in neurogenesis in both databases. This suggests that acute treatment with these psychoactive drugs negatively regulates the expression of genes involved in neurogenesis in vitro. SOX2 and three of its target genes (CCND1, BMP4, and DKK1) were also decreased after treatment with quetiapine. This can, at least in part, explain the mechanisms by which these drugs decrease neurogenesis at a transcriptional level in vitro. These results were supported by the finding of increased metabolite markers of mature neurons following treatment with most of the drugs tested, suggesting increased proportions of mature relative to immature neurons consistent with reduced neurogenesis.
Introduction
Neurogenesis is the development of new neurons from neural stem/precursor cells (NSCs). In the adult human brain, the NSCs reside primarily in two regions: The subgranular zone in the dentate gyrus of the hippocampus, and the subventricular zone of the lateral ventricles [1].
The NSCs are a diverse population of cells with the capacity to self-renew and differentiate into neurons in response to stimuli in order to maintain central nervous system homeostasis. Local stimuli from the niche where these cells are found, as well as extrinsic factors such as neurotransmitters, growth factors, cytokines, and adhesion molecules, can positively or negatively affect the NSCs' state and differentiation potential, thereby modulating neurogenesis in the adult brain [1]. For example, microglial activation and secretion of pro-inflammatory cytokines are processes involved with healthy aging in the brain, and this pro-inflammatory environment is deleterious to NSCs, thereby decreasing neurogenesis [2]. Age-related changes in the cell cycle due to accumulation of damaged proteins can cause a reduction in the proliferation rate of NSCs [3] and cause the cells to become quiescent, but they can be reactivated upon stimulation, such as by physical exercise or epileptic seizures, through Notch signalling activation [4].
This form of circuit plasticity may be altered in different diseases and targeted with pharmacological therapies [5]. Cognitive deficits, mood dysregulation, and declines in hippocampal volume have been correlated with mental disorders that present decreased neurogenesis, such as major depression, post-traumatic stress disorder, schizophrenia, and Alzheimer's disease [6]. Neurogenesis in the hippocampus is necessary for the therapeutic effects of antidepressants [7], but in the subventricular zone of the lateral ventricles, for example, chronic treatment with fluoxetine decreased neurogenesis [8]. However, if chronic stress was first induced (known to be both a contributor to the development of mood disorders [9] and decreased neurogenesis), the treatment with fluoxetine and imipramine can revert the decline in neurogenesis by increasing the NSCs pool and survival [10].
While the effects of antidepressants acting through serotonergic receptors is widely investigated and thus better understood, the effects and mechanisms of action of other psychoactive drugs in neurogenesis remain unknown and/or controversial, with studies in mice showing positive effects on neurogenesis. Chronic treatment with atypical antipsychotics (olanzapine, quetiapine, clozapine, risperidone, and aripiprazole) increased neurogenesis in the hippocampus of adult mice [11]. In contrast, haloperidol treatment resulted in decreased neurogenesis [11]. On the other hand, in humans, the use of atypical antipsychotics (risperidone, olanzapine, paliperidone, amisulpride, and aripiprazole) was associated with a reduction in grey matter volume in first-episode schizophrenia patients [12], suggesting a reduction in neurogenesis.
The changes that occur in the NSCs in response to environmental cues such as stress, psychiatric disorders, and aging, together with the evidence of NSCs responsiveness to drugs used in treating psychiatric disorders, supports the idea of neurogenesis being a potential therapeutic target for these diseases. Therefore, this study aims to evaluate the effects of common psychoactive drugs, used in the treatment of affective disorders (bipolar disorder and schizophrenia), on the expression of genes involved in neurogenesis in a human neuronal model in cell culture.
Gene Ontology (GO) Database
The effects of the eight individual psychoactive drugs on the expression level of genes in the GO database classified as involved in neurogenesis were investigated. The effects were quantified and expressed as the enrichment score (ES) and the normalised ES (NES) of all genes following gene set enrichment analysis (GSEA). The results are summarised in Table 1, and the GSEA plot for each drug is shown in Supplementary Figure S1. Abbreviations: ES = Enrichment score; NES = Normalised enrichment score.
As shown in Table 1, 7 out of the 8 drugs of interest appeared to cause a general decrease in the expression of genes involved in neurogenesis (as indicated by negative ES and NES). Among the 7 drugs associated with reduced gene expression, 6 showed significant p values (p < 0.05). Risperidone was the only drug that led to an increase in the expression of genes involved in neurogenesis.
Having established that the drug treatments generally decreased the expression of genes involved in neurogenesis based on the GO database, we further identified the number of genes that were differentially expressed after the drug treatments.
As illustrated in Figure 1, risperidone and lamotrigine were shown to regulate the least number of differentially expressed genes individually (n = 0 and n = 1, respectively) whilst quetiapine appeared to regulate the most (n = 586). Interestingly, a number of differentially expressed genes were commonly regulated by more than 2 and up to 5 of the drugs.
Mammalian Adult Neurogenesis Gene Ontology (MANGO) Database
The GO database contains 1818 genes that are annotated as being involved in neurogenesis. However, it is likely that a number of these genes are only peripherally involved, many of which are not supported by empirical evidence, and may have more important roles in other biological processes. Therefore, we sought to identify an alternative ontological database that was more specific for genes that play key roles in neurogenesis. The MANGO database classifies approximately 250 genes as involved in neurogenesis and provides experimental evidence to accurately define the role of each gene. Effects of the drugs on MANGO neurogenesis genes are shown in Table 2. Because the data were not normally distributed, the change in gene expression level for each gene was quantified and expressed as a median log fold change (logFC) with 95% confidence intervals for the median value.
The results showed that 6 of the 8 drugs significantly down-regulated the expression of genes involved in neurogenesis as classified in MANGO (amisulpride, aripiprazole, clozapine, lamotrigine, lithium, and quetiapine). Among the various drugs, quetiapine (median logFC = −0.053), clozapine (−0.028), and lithium (−0.023) showed the strongest effects. Overall, these findings are consistent with the previous results generated using the GO database.
To compare the effects on the expression levels of genes involved in the various processes of neurogenesis across 6 of the 8 drugs (lamotrigine and risperidone were excluded from the analysis due to empty counts), a χ 2 test was performed. The result showed no statistically significant differences between the drugs of interest (p = 0.87). As depicted in Figure 2, the lack of statistical significance is likely due to the fact that the drugs showed similar patterns of proportions for the various processes that comprise neurogenesis, as defined by MANGO. The complete list of these unique genes and the processes they are categorised in are shown in Supplementary Table S1. involved in neurogenesis based on the GO database, we further identified the number of genes that were differentially expressed after the drug treatments.
As illustrated in Figure 1, risperidone and lamotrigine were shown to regulate the least number of differentially expressed genes individually (n = 0 and n = 1, respectively) whilst quetiapine appeared to regulate the most (n = 586). Interestingly, a number of differentially expressed genes were commonly regulated by more than 2 and up to 5 of the drugs. To visualise the number of unique genes transcriptionally regulated by the different drug treatments, both individually and collectively, Figure 3 was plotted using the MANGO database neurogenesis genes with the FDR cut-off set at <0.05. In order to focus on the commonalities between various drugs, we applied a more stringent standard for differentially expressed genes to be considered as overlapping, i.e., genes must be differentially regulated in the same direction (negative or positive trend) across multiple drugs. neurogenesis across 6 of the 8 drugs (lamotrigine and risperidone were excluded from the analysis due to empty counts), a χ 2 test was performed. The result showed no statistically significant differences between the drugs of interest (p = 0.87). As depicted in Figure 2, the lack of statistical significance is likely due to the fact that the drugs showed similar patterns of proportions for the various processes that comprise neurogenesis, as defined by MANGO. The complete list of these unique genes and the processes they are categorised in are shown in Supplementary Table S1. To visualise the number of unique genes transcriptionally regulated by the different drug treatments, both individually and collectively, Figure 3 was plotted using the MANGO database neurogenesis genes with the FDR cut-off set at <0.05. In order to focus on the commonalities between various drugs, we applied a more stringent standard for differentially expressed genes to be considered as overlapping, i.e., genes must be differentially regulated in the same direction (negative or positive trend) across multiple drugs. Similar to the findings reported from using the GO database, risperidone and lamotrigine individually regulated the least number of genes (n = 0), whilst quetiapine appeared to regulate the most (n = 143). Some genes were shown to be transcriptionally regulated in the same direction by up to 4 drugs. We looked closely at the neurogenesis genes that were differentially regulated by 3 or more of the drugs, as these may represent key targets for future drug development. Table 3 below summarises the data for the 15 genes that are regulated in the same direction by 3 or more of the drugs. For several of these genes that were differentially expressed following drug treatment (FDR < Similar to the findings reported from using the GO database, risperidone and lamotrigine individually regulated the least number of genes (n = 0), whilst quetiapine appeared to regulate the most (n = 143). Some genes were shown to be transcriptionally regulated in the same direction by up to 4 drugs. We looked closely at the neurogenesis genes that were differentially regulated by 3 or more of the drugs, as these may represent key targets for future drug development. Table 3 below summarises the data for the 15 genes that are regulated in the same direction by 3 or more of the drugs. For several of these genes that were differentially expressed following drug treatment (FDR < 0.05), we conducted a quantitative reverse transcription polymerase chain reaction (RT-qPCR) to validate these effects. In most cases, we observed a dose-dependent effect of the drugs on the expression of these genes, which confirmed the results obtained from the next generation sequencing dataset ( Figure 4). Supplementary Table S2 contains the entire list of transcription factors that could affect the expression of genes in the MANGO database as assessed using TRRUST.
SOX2
Among the genes regulated by multiple drugs was SOX2 (Sry-related HMG box 2), a transcription factor previously implicated in neural development and neurogenesis. As shown in Figure 5, SOX2 gene expression was decreased following treatment with quetiapine (p = 1.75 × 10 −8 , FDR q = 2.76 × 10 −7 ), clozapine (p = 0.00021, FDR q = 0.010), and valproate (p = 0.0036, FDR q = 0.022), and tended to be decreased by amisulpride (p = 0.0027, FDR q = 0.088) and aripiprazole (p = 0.0072, FDR q = 0.11). To validate these effects, RT-qPCR was performed, and the results were compared to the next generation sequencing data. As shown in Figure 6, we observed a dose-dependent effect of clozapine and quetiapine on SOX2 gene expression. This confirms the next generation sequencing results for these two drugs.
To further investigate the effects of reducing SOX2 gene expression on neurogenesis genes, we first verified the impact of SOX2 on the MANGO neurogenesis genes using the TRRUST database. To achieve this, the list of MANGO neurogenesis genes was submitted to and analysed using the TRRUST database, which generated a list of transcription factors that could affect the expression of the genes submitted for query. Fisher's exact test was then performed based on the contingency table of gene counts. The results obtained for SOX2 revealed a p value of 0.00059, and an FDR q-value of 0.00096, indicating that SOX2 was likely a major regulator of these neurogenesis genes.
As SOX2 is a transcription factor, we then investigated whether the expression of down-stream target genes of SOX2 were affected by the psychotropic agents. We focused on data obtained from the quetiapine, clozapine, and valproate treatment arms as they showed the strongest evidence of negatively regulating SOX2 gene expression compared to the other drugs.
Among the list of genes submitted for query in TRRUST, 3 target genes were identified as most strongly associated with SOX2: Dickkopf WNT signalling pathway inhibitor 1 (DKK1), Bone morphogenetic protein 4 (BMP4), and Cyclin D1 (CCND1). Table 4 below shows the change in next generation sequencing expression levels in these 3 genes following drug treatment. The findings suggest that quetiapine had an overall negative effect on the expression of these genes (logFC range = −0.74 to −0.41) and by implication, neurogenesis. RT-qPCR was used to confirm the effects of the drugs on the expression of these genes (Figure 7). In most cases, the differential expression observed in the next generation sequencing data set were confirmed and showed a dose-dependent effect of the drugs on these genes. FDR q = 2.76 × 10 −7 ), clozapine (p = 0.00021, FDR q = 0.010), and valproate (p = 0.0036, FDR q = 0.022), and tended to be decreased by amisulpride (p = 0.0027, FDR q = 0.088) and aripiprazole (p = 0.0072, FDR q = 0.11). To validate these effects, RT-qPCR was performed, and the results were compared to the next generation sequencing data. As shown in Figure 6, we observed a dose-dependent effect of clozapine and quetiapine on SOX2 gene expression. This confirms the next generation sequencing results for these two drugs. To further investigate the effects of reducing SOX2 gene expression on neurogenesis genes, we first verified the impact of SOX2 on the MANGO neurogenesis genes using the TRRUST database. To achieve this, the list of MANGO neurogenesis genes was submitted to and analysed using the TRRUST database, which generated a list of transcription factors that could affect the expression of the genes submitted for query. Fisher's exact test was then performed based on the contingency table of gene counts. The results obtained for SOX2 revealed a p value of 0.00059, and an FDR q-value of 0.00096, indicating that SOX2 was likely a major regulator of these neurogenesis genes.
As SOX2 is a transcription factor, we then investigated whether the expression of down-stream target genes of SOX2 were affected by the psychotropic agents. We focused on data obtained from the quetiapine, clozapine, and valproate treatment arms as they showed the strongest evidence of negatively regulating SOX2 gene expression compared to the other drugs.
Among the list of genes submitted for query in TRRUST, 3 target genes were identified as most strongly associated with SOX2: Dickkopf WNT signalling pathway inhibitor 1 (DKK1), Bone morphogenetic protein 4 (BMP4), and Cyclin D1 (CCND1). Table 4 below shows the change in next generation sequencing expression levels in these 3 genes following drug treatment. The findings suggest that quetiapine had an overall negative effect on the expression of these genes (logFC range = −0.74 to −0.41) and by implication, neurogenesis. RT-qPCR was used to confirm the effects of the drugs on the expression of these genes (Figure 7). In most cases, the differential expression observed in the next generation sequencing data set were confirmed and showed a dose-dependent effect of the drugs on these genes. Figure 6. The effects of clozapine, quetiapine, and valproate at various doses on the regulation of SOX2 as revealed by next generation sequencing and RT-qPCR data. * p < 0.05. Figure 7. The effects of the drugs at various doses on the regulation of 3 target genes of SOX2 as revealed by next generation sequencing and RT-qPCR data. * p < 0.05.
Cell Culture Metabolite Profiling
The six drugs with effects on genes involved in neurogenesis were tested for their effects on the levels of metabolites suggested to either have a role in neurogenesis, or as markers of neuronal maturity (Table 5). Generally, the drugs led to an increase in the levels of most of the metabolites evaluated in NT2-N cells, particularly N-acetyl-L-aspartic acid (NAA; increased by amisulpride, aripiprazole, clozapine, and lithium), L-glutamic acid (increased by amisulpride, aripiprazole, and lithium), Gamma-Aminobutyric acid (GABA; increased by amisulpride, aripiprazole, clozapine, and quetiapine) and L-glutamine (increased by amisulpride, aripiprazole, and lithium). The exceptions were amisulpride and L-lactic acid (logFC = −0.35, p = 0.042), quetiapine and glutathione (logFC = −2.94, p = 0.002), and lithium and GABA (logFC = −0.23, p = 0.024).
Cell Culture Metabolite Profiling
The six drugs with effects on genes involved in neurogenesis were tested for their effects on the levels of metabolites suggested to either have a role in neurogenesis, or as markers of neuronal maturity (Table 5). Generally, the drugs led to an increase in the levels of most of the metabolites evaluated in NT2-N cells, particularly N-acetyl-l-aspartic acid (NAA; increased by amisulpride, aripiprazole, clozapine, and lithium), l-glutamic acid (increased by amisulpride, aripiprazole, and lithium), Gamma-Aminobutyric acid (GABA; increased by amisulpride, aripiprazole, clozapine, and quetiapine) and l-glutamine (increased by amisulpride, aripiprazole, and lithium). The exceptions were amisulpride and l-lactic acid (logFC = −0.35, p = 0.042), quetiapine and glutathione (logFC = −2.94, p = 0.002), and lithium and GABA (logFC = −0.23, p = 0.024).
Discussion
The generation of functional neurons from progenitor cells, neurogenesis, is a continuing process in the hippocampus throughout life, from embryonic development until adulthood [13]. Altered neurogenesis is implicated in diverse neuropsychiatric disorders including depression, schizophrenia, and bipolar disorder [14]. This study showed the effects of psychoactive drugs on genes involved in neurogenesis. Clozapine, amisulpride, aripiprazole, lithium, quetiapine, and lamotrigine all significantly decreased the expression of genes involved in neurogenesis, while risperidone significantly increased the expression of genes involved in neurogenesis. Apart from risperidone, the findings were confirmed when we investigated the effects of these drugs on the expression of neurogenesis genes using a more curated database (MANGO).
The literature around the effects of atypical antipsychotics in neurogenesis is controversial. In rodent models, chronic administration of aripiprazole, quetiapine, and clozapine significantly increased neurogenesis, cell proliferation, and survival [11,[15][16][17]. Kim et al. (2015) suggested a reversal of depressive behaviours through preventing degeneration of dopaminergic neuronal cells and enhancing neurogenesis after treatment with aripiprazole [18]. Aripiprazole treatment increased neurite branches in primary cortical neurons derived from mice with dopamine D2 receptor hyperactivity and disrupter in schizophrenia 1 (DISC1) [19]. Clozapine treatment in male Wistar rats subjected to chronic mild stress upregulated adult neurogenesis and neuronal survival, reversing the behavioural effects of chronic stress [20]. Quetiapine augmentation treatment increased hippocampal cell proliferation and neuronal differentiation and improved depressive-like behaviours in treated rats [21].
In humans, some studies of antipsychotic treatment were associated with grey matter loss as shown in imaging studies in patients with schizophrenia, but studies have yielded inconsistent results as explored in a meta-analysis by Fusar-Poli [22]. Wang et al. (2018) showed a significant decrease in grey matter volume in the left parahippocampal gyrus/hippocampus, right temporal pole mid/superior temporal gyrus, right parahippocampal/hippocampus, and right insula after four weeks of treatment with antipsychotics [23]. Guo et al. (2019) also showed a decrease in grey matter volume in the bilateral frontal, temporal, and left parietal brain regions associated with antipsychotic treatment [12]. In both studies, the majority of the treatment group were using several antipsychotics at the same time, and therefore conclusions on the specific effects of individuals drugs could not be made.
Antipsychotics are also prescribed for patients with bipolar disorder, together with mood stabilisers and antidepressants. These drugs have also been associated with structural brain differences in these patients, but data is still inconclusive and difficult to parse from illness effects. In bipolar disorder, total grey matter volume was reduced in individuals treated with atypical antipsychotics [24]. The ENIGMA study reported significantly reduced cortical surface areas of large prefrontal areas in individuals with bipolar disorder taking atypical antipsychotics [25]. However, lithium, considered the first line treatment for bipolar disorder, has been demonstrated to increase neurogenesis in the dentate gyrus of the rodent hippocampus [26,27]. Another study also demonstrated that lithium could promote neuronal differentiation of hippocampal neural progenitor cells both in vitro and in vivo [28]. Yucel et al. (2008) also demonstrated a bilateral increase in hippocampal volume in bipolar disorder patients treated with lithium, but not lamotrigine [29]. Similarly, in other human imaging studies, grey matter volume increased after four weeks of lithium treatment [30] and appeared to be independent of long-term treatment response [31]. Lithium increases brain-derived neurotrophic factor (BDNF) expression and genes associated with neuroprotection, such as Bcl2 and Bcl-XL, and also decreases the expression of pro-apoptotic genes Bax, Bad, and caspases in rat hippocampal neurons [32]. This effect on apoptosis could, at least in part, be the mechanism involved in the increase of grey matter volume in vivo, rather than effects on neurogenesis.
Overall, drugs analysed in our study appear to increase neurogenesis in rodent models, which is in contrast to the findings from human imaging studies, where the drugs appear to decrease grey matter volume, suggestive of reduced neurogenesis, although it is unclear in some designs whether this relates to illness effects or other lifestyle or medical confounders. Our data is at transcriptional levels. The increases in grey matter demonstrated in imaging studies probably occurred because of neurotrophic effects and could explain the differences in results. To the best of our knowledge, there is no study with a similar approach to ours.
We further investigated possible mechanisms by which these drugs decrease genes involved in neurogenesis in vitro, identifying a candidate target molecule in SOX2. SOX2 is a transcription factor and a marker of the nervous system from the beginning of development in many species, and acts to co-ordinate widespread transcriptional regulation of genes involved in neurogenesis. SOX2 mutations in humans cause defects in the brain, particularly in the hippocampus, involving cognition, movement control, and vision [33]. As identified by TRRUST analysis, three major gene targets of SOX2 in the neurogenesis pathway are CCND1, BMP4, and DKK1. CCND1 promotes neurogenesis in vivo, a role that is not linked to its cell cycle function [34]. Both in vivo and in vitro studies present evidence of the role of BMP4 in the modulation of adult neurogenesis in the hippocampus [35]. DKK1 is a suppressor of neural stem cell proliferation, and studies demonstrate increased self-renewal of neural progenitors and increased generation of immature neurons after the deletion of DKK1 in adult brains [36]. The downregulation of BMP4, CCDN1, and DKK1 by quetiapine could represent a molecular mechanism for the reduction in expression of neurogenesis genes observed in this study after treatment with quetiapine.
The drugs appeared to cause a general increase in a number of metabolites in NT2-N cells that have previously been suggested to be related to the process of neurogenesis. While it may appear that this suggests a positive effect on neurogenesis, the increased levels of these particular sets of metabolites are related to the presence of more mature neurons, with active neuritogenesis, and not the generation of new immature neurons.
For example, decreased levels of NAA is used as a marker of neuronal integrity and is reportedly decreased after whole-brain radiation in rodents, associated with changes in neurotransmission and loss of neuronal viability [37]. The pathological pruning of dendrites has also been associated with a reduction of NAA, and this could contribute to the finding of reduced brain volume associated with reduced levels of NAA in schizophrenia, bipolar disorder, post-traumatic stress disorder, and obsessive-compulsive disorders [38]. The effects seen in this study corroborate previous studies showing that antipsychotics and mood stabilizers increase the levels of NAA, suggesting a therapeutic response acting on increasing neuronal viability [39].
GABA is another example, with four of the drugs increasing the levels of this metabolite in NT2-N cells. GABA agonists reportedly regulate synaptic integration by increasing the numbers and extension of neurites and promoting the survival of existing neurons [40,41].
In summary, the present study demonstrated that psychoactive drugs decrease the expression of genes involved in neurogenesis in neuronal cell culture after 24 h treatment, partially through inhibition of SOX2 and its targets genes BMP4, CCDN1, and DKK1. It also showed the alterations in the levels of metabolites involved in neuronal health, particularly NAA and GABA.
This study has some limitations that need to be acknowledged. Only one dose of each drug was tested. In addition, we explored acute effects of these drugs, and there is a possibility of a different effect during chronic administration. Our study did not evaluate the effects of these drugs in models of any particular disease state, therefore the nature of the pathophysiologic process of specific diseases might interfere in the effect of these drugs in patients.
Cell Culture
NTera2/cloneD1 (NT2) is a pluripotent cell line used as a model of human neurons due to its ability to differentiate into post-mitotic neurons (NT2-N) following retinoic acid (RA) treatment [42,43]. NT2 cells (CVCL_0034, ATCC, Manassas, VA, USA) were cultured and differentiated into NT2-N cells as previously described [44]. In summary, cells were cultured and expanded in Dulbecco's modified Eagle's Medium (DMEM; Life Technologies, Melbourne, Australia) supplemented with 10% foetal bovine serum (FBS; Thermo Fisher Scientific, Melbourne, Australia) and 1% antibiotic/antimycotic solution (Life Technologies). NT2 cells were treated with 10 −5 M RA (Sigma-Aldrich, Sydney, Australia) for 28 days, with media refreshed every 2-3 days. For experiments, cells (2 × 10 5 cells/well) were seeded onto 24-well plates coated with 10 µg/mL poly-d-lysine (Sigma-Aldrich) and 10 µg/mL laminin (Sigma-Aldrich). To enrich for a culture of differentiated neuronal cells, mitotic inhibitors (1 µM cytosine and 10 µM uridine; Sigma-Aldrich) were added to the media every 2-3 days for a total of 7 days. The expression of the neuronal markers NeuroD, Tau, and GluR were evaluated using polymerase chain reaction and agarose gel electrophoresis (data not shown). This was used to validate the generation of differentiated cells with a neuron-like phenotype.
Genome-Wide Gene Expression Quantification
Following the 24-h drug treatment, cells were harvested using Trizol and total RNA was extracted using RNeasy ® mini kits (Qiagen, Melbourne, Australia). The quality of the extracted RNA was evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies, Melbourne, Australia). The RNA quantity was determined using a NanoDrop 1000 (Thermo Fisher Scientific, Waltham, MA, USA).
RNAseq libraries were prepared for all samples from 1 µg total RNA using a TruSeq RNA Sample Preparation Kit (Illumina, Victoria, Australia) as per the manufacturer's instructions. Samples were run on an Illumina HiSeq platform (HiSeq 2500 rapid 50bpSE; 1 flow cell, 2 lanes) to quantify genome wide mRNA expression.
Genome-Wide Gene Expression Analysis
The raw data were obtained in fastq format and processed using the Deakin Genomics Centre RNA-Seq alignment and expression quantification pipeline ( [URL]). In summary, this involves: Raw read quality filtering and adapter trimming (ILLUMINACLIP:2:30:10:4, SLIDINGWINDOW:5:20, AVGQUAL:20 MINLEN:36) with Trimmomatic v35 [45], and alignment to the reference genome using STAR v2.5 in 2-pass mode (Human genome version GRCh38) [46]. The expression was quantified at the gene level, and individual sample counts were collated into a m × n matrix for differential abundance testing. Normalisation (TMM) and removal of low expressed gene were performed using edgeR [47] in R [48] following the edgeR manual (<1 cpm in n samples, where n is the number of samples in the smallest group for comparison).
Differential gene expression analysis was assessed using edgeR in R, and statistical significance was corrected for multiple testing using FDR by applying the Benjamini-Hochberg method on the p-values. Genes with FDR q-values of <0.05 were considered to be differentially expressed.
GO Database
GSEA was deployed using the R package fgsea from Bioconductor [49,50], with gene lists ranked based on log fold change (logFC) and the neurogenesis gene set retrieved from the GO database by the AMIGO web application [51,52]. The following filters were set for neurogenesis GO annotations: "GO:0022008" as accession identification, "Homo sapiens" as organism, and "experimental evidence" as evidence. The resulting tables had enrichment scores and p-values calculated from 1000 permutations. The UpSet plot was generated using the UpSetR package in R to summarize the overlap of differentially expressed GO neurogenesis genes (FDR cutoff < 0.05) across the 8 drugs [53,54].
MANGO Database
For the purpose of focusing on genes with well-established evidence of involvement in the neurogenesis process, the MANGO database was utilised to identify genes regulated by the drugs of interest. MANGO version 3.2 contains 397 genes curated and classified according to their effects and associated subprocesses in adult neurogenesis [55]. Differential expression of the MANGO neurogenesis genes by each drug were then tested for overall statistical significance. Distribution of logFC data was checked for normality using Kolmogorov-Smirnov tests. Drug treatment groups were compared against their respective controls using independent samples t-tests for normally distributed data and Mann-Whitney U tests for data not normally distributed. The 95% confidence interval for the median was obtained by bootstrapping 1000 samples using R [48]. χ 2 test of homogeneity was used to test if the proportions for subprocesses of neurogenesis (e.g., proliferation, axonogenesis) were homogenous across different drugs. Aiming to unravel the biological processes driving neurogenesis, we further investigated transcription factors that are more likely to regulate MANGO neurogenesis genes using the prediction tool of key transcription factors from TRRUST-a manually curated database for transcriptional regulatory networks [56].
Validation of Genome-Wide Gene Expression Using RT-qPCR
Following the 24-h treatment, cells were harvested, and RNA was extracted using RNeasy ® mini kits (Qiagen) and reverse-transcribed to produce cDNA using Maxima H Minus first strand cDNA synthesis kit (Thermo Fisher Scientific) following the manufacturer's instructions. RT-qPCR was used to measure the expression of specific genes as listed on Table 6. The experiments were carried out in a QuantStudio 3 Real-time PCR system (Thermo Fisher Scientific) using the following protocol: 95 • C for 7 min, 4 cycles of 95 • C for 30 s, and 60 • C for 1 min and then data acquisition, 60 • C for 30 s, 55-95 • C, data acquisition and 20 • C for 10 s. Resultant melt curves were used as an indicator of amplification specificity. The Quant-iT™ OliGreen ® ssDNA Assay Kit (Life Technologies) was used to quantify the cDNA concentration in each sample as per the manufacturer's instructions. Gene expression data was quantified using the ∆∆C t method normalised to the derived cDNA concentration of each sample. Kolmogorov-Smirnov test was used to check data sets for normality of distribution. Levene's Test was used to determine whether or not equal variances could be assumed between groups. Drug treatment groups were compared against their respective controls using independent sample t-tests for normally distributed data and Mann-Whitney U tests for data not normally distributed. Statistical analysis was performed using Statistical Package for the Social Sciences version 22 (SPSS) software. Differences were considered statistically significant when p ≤ 0.05.
Cell Culture Metabolite Extraction and Profiling
Cells were treated as described on Section 4.2. After 24 h of treatment, cells were washed twice with 1× phosphate buffered saline (PBS) at 37 • C then snap-frozen by covering the plate in liquid nitrogen. Metabolites were extracted on ice by addition of 250 µL/well of methanol:chloroform (9:1 v/v), containing the internal standards, U-13C-sorbitol (4 µM) and 13C5, 15N-valine (4 µM). Cells were scraped and incubated on ice for 10 min. Samples were then centrifuged (5 min, 14,000 rpm, 4 • C) to pellet precipitated proteins, and the supernatants were transferred to fresh 1.5 mL tubes. Briefly, 200 µL of the aqueous extract was dried for analysis in vacuo. An aliquot of each extracted sample was also taken and pooled to create a pooled biological quality control (PBQC) sample. This pooled sample was split into equivalent loadings as for the biological samples and also dried for analysis. These PBQC samples were run at regular intervals throughout the sample sequence. Gas chromatography-mass spectrometry (GC-MS) analysis was performed following methoximation and trimethylsilylation using a Shimadzu GC/MS-TQ8050NX system and analysed in multiple reaction monitoring (MRM) mode using the Shimadzu Smart Metabolites Database containing 521 MRM metabolite targets. Subsequent data analysis was performed in a targeted manner using Shimadzu LabSolutions Insight software (version 3.6). Briefly, target ion areas for polar metabolites contained within the Shimadzu Smart Metabolites database were integrated and output as a data matrix for downstream data analysis. Each detected metabolite was visually inspected and manually integrated if required. This resulted in a highly curated matrix representing the detected metabolites in each sample.
==
Domain: Psychology Biology Medicine
|
[{"added": "2020-11-12T09:06:12.465Z", "created": "2020-11-01T00:00:00.000", "id": "226303704", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/1422-0067/21/21/8333/pdf", "pdf_hash": "84d28a919f15c719781770075d3b4ee0b2a21b52", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:32491", "s2fieldsofstudy": ["Medicine", "Biology", "Psychology"], "sha1": "70cf55d040fbbd548bb4e310379c7bfdb86d3e99", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Metabolomics in Toxicology and Preclinical Research
Tzutzuy Ramirez 1§, Mardas Daneshian 2§, Hennicke Kamp 1, Frederic Y. Bois 3, Malcolm R. Clench 4, Muireann Coen 5, Beth Donley 6, Steven M. Fischer 7, Drew R. Ekman 8, Eric Fabian 1, Claude Guillou 9, Joachim Heuer 10, Helena T. Hogberg 11, Harald Jungnickel 12, Hector C. Keun 5, Gerhard Krennrich 13, Eckart Krupp 14, Andreas Luch 12, Fozia Noor 15, Erik Peter 16, Bjoern Riefke , Mark Seymour 18, Nigel Skinner 19, Lena Smirnova 11,12, Elwin Verheij 20, Silvia Wagner 16, Thomas Hartung 2,11§, Bennard van Ravenzwaay 1§, and Marcel Leist 2,21§ BASF Se, experimental toxicology and ecology, ludwigshafen, Germany; Center for Alternatives to Animal testing – europe (CAAt-europe), University of Konstanz, Konstanz, Germany; Royallieu Research Center, technological University of Compiegne, Compiegne, France; Biomedical Research Centre, Sheffield Hallam University, Sheffield, UK; Imperial College london, london, UK; Stemina Biomarker Discovery Inc., Madison, USA; Agilent life Sciences Group, Agilent technologies, Santa Clara, USA; U. S. ePA, National exposure Research laboratory, ecosystems Research Division, Athens, GA, USA; Joint Research Centre of the European Commission, Institute for Health & Consumer Protection, Ispra, Italy; Federal Institute for Risk Assessment (BfR), Department of Scientific Services, Berlin, Germany; Johns Hopkins University, Bloomberg School of Public Health, Center for Alternatives to Animal testing (CAAt), Baltimore, USA; Federal Institute for Risk Assessment (BfR), Department of Product Safety, Berlin, Germany; BASF SE, GVC/S Scientific Computing, Ludwigshafen, Germany; Sanofi-Aventis Deutschland GmbH, Genetic & Investigative Toxicology, Frankfurt, Germany; Biochemical engineering, Saarland University, Saarbruecken, Germany; metanomics GmbH, Berlin, Germany; Bayer Pharma AG, Investigational toxicology, Berlin, Germany; Syngenta, Jealott’s Hill International Research Centre, Berkshire, UK; Agilent technologies, london, UK; tNO Quality of life, Zeist, the Netherlands ; Doerenkamp-Zbinden chair for in vitro toxicology and biomedicine, Konstanz, Germany
of major importance. This is the basis for non-invasive or minimally-invasive sampling of body fluids (blood, urine, etc.), and metabolomics analysis on such samples to gain information on target organ toxicities that would otherwise only be identifiable by highly invasive (histopathological) methods (Ebbels et al., 2007;Lindon et al., 2003). Also, time course studies within one study subject or animal are greatly facilitated by this particular advantage of the metabolomics approach (Ebbels et al., 2007;van Ravenzwaay et al., 2007van Ravenzwaay et al., , 2012. In addition to providing information for a large number of metabolites in one measurement, either from body fluids, tissues, or whole organisms (i.e., fungi, aquatic organisms, etc.), metabolomics has been applied to in vitro cell systems for understanding drug effects (Balcke et al., 2011;Strigun et al., 2011a,b). First pilot studies show that future applications of the metabolomics approaches are high throughput chemical screening applications ( [URL]). Finally, new imaging techniques are not only capable of locating environmental toxicants within biological systems but can be used in combination with metabolomics approaches to describe specific toxicological effects within cells (Haase et al., 2011;Tentschert et al., in press).
Due to the increasing use of metabolomics in toxicology and safety sciences, a workshop was organized in Berlin on February 14-15, 2012. Scientists from academia, industry, and regulatory bodies discussed the current status of this approach and its present/future applicability. One day prior to the workshop, an international symposium was organized by BASF/CAAT-Europe to present the state of the art regarding the use of metabolomics for addressing a variety of pertinent toxicological questions. Participants identified several hurdles in the wider application of metabolomics in safety assessments and for in vitro compound screening. This paved the way for in-depth discussions on these issues in the workshop that followed. Here, we summarize the result of these discussions and offer solutions for successfully moving forward with this important area of research.
Metabolomics in vitro
The application of metabolomics in vitro is an emerging theme that has been driven mostly by two major factors: (1) a better understanding of the biochemical changes provoked by a toxic insult in a defined and controllable experimental system and (2) the increasing need to move towards the use of human-relevant non-animal alternatives in toxicology in accordance with policies endorsing the 3Rs concept (Reduction, Refinement, and Replacement of animal testing). Special challenges for the application of metabolomics in vitro can be summarized as 1) different requirements of models, 2) quality criteria and quality control, 3) application areas, 4) investigation strategies, 5) technical challenges of the analysis, and 6) extrapolation to the in vivo situation.
Apart from the evident benefits of reducing animal testing and getting better insights into the molecular targets of xenobiotics and their mode of action (MoA), the application of
Introduction
Metabolomics is the comprehensive analysis of hundreds of metabolites in a biological sample; it provides detailed information on the physiological status of a living organism, a cell, or a subcellular compartment at any given moment. The analytes of interest are the small endogenous molecules, such as carbohydrates, amino acids, nucleotides, phospholipids, steroids, or fatty acids and their derivatives, which are produced and/or transformed by cells as a result of cellular metabolism (Lindon et al., 2004;Patti et al., 2012). Since these metabolites directly reflect the biochemical processes of the system under investigation, their analysis offers the opportunity not only to gain insight into the activity of biochemical pathways giving a particular metabolite profile, but also into the alteration of such pathways. As such, metabolomics can be used to study human physiology not only under normal conditions but also in pathological situations.
This opens up the possibility for its application in clinical settings; as such an approach allows the monitoring of treatment success at a very early stage. This is essential given the rise of combinatory multi-drug treatment scenarios. In this context, metabolomics has been expanding in scope from a basic research approach to an applied science, not only in medicine but also in the fields of biotechnology and toxicology (Bouhifd et al., 2013;Llorach et al., 2012;Nguyen et al., 2012;van Ravenzwaay et al., 2012;Zhang et al., 2012b). For instance, in biotechnology, metabolomics offers the possibility of assessing the relationship of a genetic modification(s) to a specific desired phenotype in an effort to determine the critical biochemical pathways involved. For example, it allows identification of increased activation of certain metabolic pathways for improved yield or production (Kim et al., 2012). In clinical medicine and pharmacology, metabolomics is becoming an established tool for the identification of pathologies through the use of more relevant biomarkers (Patti et al., 2012;Rhee and Gerszten, 2012).
Metabolomics is ideally positioned to address the challenges of toxicology in the 21 st century (Tox-21c). It represents a powerful tool for collecting rich mechanistic information indicating not only the extent of a toxic insult but also its underlying mechanisms.
From the currently available data it seems that metabolomics information can be more easily related to classical toxicological endpoints used in animal studies than, e.g., transcriptomics data. One reason for this may be that the changes in the metabolic profile are often "downstream" of those initial changes that occur at the level of the genome, transcriptome, and proteome (van Ravenzwaay et al., 2007). In addition, the relatively small number of metabolites (i.e., hundreds/thousands) present in a tissue or bio-fluid, compared to the tens of thousands of transcripts, or hundreds of thousands of proteins can be advantageous when working to determine meaningful changes associated with a toxic effect (Strauss et al., 2012;van Ravenzwaay et al., 2007). For toxicological studies, the fact that extracellular metabolites somehow reflect the intracellular situation is with reference standards for metabolomics are essential for research purposes and for the credibility of this approach (Holmes et al., 2010). Therefore, the most important issues include the definition and the availability of negative and positive test controls. Notably, standards will allow a clear communication of results and integration of metabolomics data with other -omics approaches, e.g., proteomics and transcriptomics (Holmes et al., 2010).
Moreover, the definition of the test and its respective acceptance criteria, sterility issues, assurance of the identity and the condition of the cells (e.g., cell aging or spontaneous mutations), measures for the ratio of cell types and differentiation stages, adequate measures for viability assessment and availability of in vitro bio-kinetics data, determination of free concentrations of test components, information on the cellular concentrations of compounds, and the contribution of metabolism in case of metabolically active systems.
Apart from the biological variability, analytics and sample processing also are potential sources of variability. Therefore, the group agreed that there is an urgent need for harmonization of metabolomics protocols, allowing integration of quality criteria not only at the biological level, but also at the analytical level. Moreover, it is also relevant to have a quality control for measurement and data processing bias.
Metabolomics-specific quality criteria
All information about technical details, such as the extraction methods and the storage methods should be compiled in standard operating procedures. Ideally, other responses of the model will be evaluated in parallel with metabolomics data (e.g., cell morphology and/or viability parameters). This allows the anchoring of the metabolomics data set to physiologic or functional responses. In this respect, verification of the presence of essential targets, signaling pathways, and response features needs to be verified. In this way it can be ensured that the changes in the metabolic profile are related to specific physiological conditions. Another important but frequently neglected criterion is whether the chosen in vitro model can be positioned reasonably within a decision tree, i.e., how the information obtained from the model can contribute to an overall evaluation or strategy.
Quantification and inter-experiment comparability
This is still an area of active development, and generally applicable solutions are not available. Differences exist between the analysis of the metabolome in medium (cell supernatant) and in cells/tissues. In the first case, leakage from cells needs to be controlled, but sample spiking with isotope-labeled reference standards is easier than for the analysis of intracellular metabolites. Sets of standards covering different pathways may be used (Roede et al., 2012). Different normalization approaches have been tried for measurements within cells.
These include the use of "house-keeping" metabolites or combinations thereof, or the introduction of isotope labeled standards. The availability of a "housekeeping" metabolite or any other form of normalization standard is of utmost importance to correct for errors and variation in the cell number, in cell harvesting, during the extraction procedure, during sample metabolomics to in vitro systems allows the application of this approach at a high throughput level. Due to the increasing interest in in vitro systems combined with metabolomics, a working group specifically discussed the current uses, overall strengths, and pitfalls of in vitro metabolomics. The major topics that were discussed are: -Adaptation to the diversity of in vitro models; -Guidance for experimental design, with special consideration of quality criteria; -Technical challenges in terms of sample processing; -Guidance for data analysis and interpretation. The subchapters below give a summary of the discussion.
Use of diverse in vitro models
As for other methods, standardization is essential for comparability and reproducibility of results. The in vitro metabolomics approach faces difficulties similar to other in vitro approaches with respect to the heterogeneity and special requirements of the experimental models (Hartung, 2007). These concerns can be classified according to their level of complexity and the handling requirements as: -Cell lines -Primary cells and stem cell derivatives -Co-cultures in 2D or 3D format -Tissues in vitro or ex vivo In addition, model organisms, such as zebrafish (D. rerio) and nematodes (C. elegans) often are used in an in vitro manner, but with the advantage of a complete living organism also endowed with metabolic capacity. The impact of handling and special requirements of the in vitro systems should be subjected to routine evaluation. For instance, changes in central metabolism also can be due to changes in the cell culture conditions rather than to the chemical insult. Therefore, profound knowledge of the system used and performance of controls for all potential influence factors of the in vitro model are crucial (see also Good Cell Culture Practice guidance, Coecke et al., 2005).
Quality criteria for in vitro systems for metabolomics
The relevance and reproducibility of in vitro data depends strongly upon the quality of the test system and its analytical endpoint (Hartung, 2009a;Kadereit et al., 2012;Leist et al., 2010Leist et al., , 2012a. Therefore, establishment of a complete set of quality criteria and guidelines is crucial for the acceptance and further use of in vitro metabolomics approaches. Such criteria can be classified into three categories, (1) general requirements (as also defined for Good Cell Culture Practice) (Coecke et al., 2005;Hartung et al., 2002), (2) criteria specific for the use of model systems for the purpose of metabolomics analysis, and (3) criteria referring to the quantification of in vitro metabolomics data.
General requirements
For metabolomics, i.e., an approach that aims at a simultaneous determination of hundreds of metabolites, quality criteria are crucial in order 1) to avoid artifacts and 2) to facilitate the comparison of generated data. Establishment of reference systems Prediction models A major goal of the field is the use of metabolomics information in comparison to known standards to predict actions of unknown chemicals in biological systems (Rusyn et al., 2012). To develop such models, metabolite patterns related to well known training compounds would be used to develop classification schemes. These would then be applied to the metabolome patterns triggered by unknown compounds in order to predict their toxicological hazard (Fig. 1).
Lead prioritization in screening: A slightly less ambitious application would be relative ranking, i.e., metabolomics would provide relative information within a group of compounds to rank them, e.g., according to toxicity and to facilitate decisions on further development.
Mode of Action: This approach can help in understanding the effect of chemicals on the complexity of the metabolic networks (Roede et al., 2012). Furthermore, this is expected to lead to the discovery of the metabolic pathways that are per-processing, in detection sensitivity, and in many other steps related to the overall analysis of the metabolome. Further features required for a robust and reproducible quantification are a stable baseline metabolite pattern and a reproducible response to positive controls.
Potential applications of metabolomics approaches in vitro
Metabolomics is a versatile approach with multiple potential applications in drug discovery and safety profiling. The in vivo metabolomics approach could be proven advantageous already, not only in clinical applications but also in toxicology. Examples are: for identification of toxic modes of action (van Ravenzwaay et al., 2012) or toxicological screenings providing insights into the potential toxic effects of chemicals under development, which lead to accelerating the decision-making processes (Kolle et al., 2012). Potential uses of the in vitro metabolomics approach include the following: The comparison of metabolite patterns of known reference compounds with their in vivo toxicological profile will yield "toxicity patterns," i.e., metabolite patterns that are correlated to defined toxicological endpoints. Alignment of metabolite patterns of unknown compounds with these toxicity patterns will allow calculation of the degree of overlap. This information is then used for toxicological predictions.
Idiosyncratic toxicity in vitro: Metabolomics approaches also may open the door to human-relevant research on idiosyncratic toxicity; such toxicities occur when a convergence of risk factors (disease, age, gender, co-medications, nutritional status, physiological status, microbiome, and genetic predispositions) disturbs the otherwise stable homeostasis and allows adverse chemical effects at otherwise innocuous concentrations (Clayton et al., 2006;Coen et al., 2009). Metabolomics studies allow insight into the cellular homeostasis under different experimental conditions, and the data may explain the conditions under which unexpected toxicities would occur.
Flux-analysis (Fluxomics): Standard metabolomics methods measure concentrations of metabolites -"frozen" at a certain time point -but not the speed of their turnover. Knowledge of the complete set of metabolites is not enough to predict the phenotype, especially for higher cells in which the distinct metabolic processes involved in their production and degradation are finely regulated and interconnected. In these cases, quantitative knowledge of intracellular fluxes is required for a comprehensive characterization of metabolic networks and their functional operation (Cascante and Marin, 2008). Under given homeostatic conditions, for instance, a glycolysis metabolite may show similar concentrations but, e.g., low turnover when mitochondria are functioning and high turnover when mitochondria are unable to meet energy requirements. For instance, depletion of ATP in livers due to high fructose exposure does not yield information on whether glycolysis or mitochondria are affected, while fluxomics would deliver clear results (Latta et al., 2000). Metabolic flux analysis has been used, e.g., to study drug effects on the metabolome of HepG2 cells (Niklas et al., 2009). By using isotope-labeled substrates of metabolism in combination with time series experiments, information on the turnover -fluxes -in different pathways can be obtained. The use of isotope labeling in metabolomics and fluxomics has been recently reviewed (Klein and Heinzle, 2012). This type of metabolomics data covers an important aspect relevant to chemical hazards, which is necessary for a systems biology type of modeling but does not yet represent a routine approach in toxicology (Hartung et al., 2012).
Sensitivity and speed: Compared to the sequential measurement of individual endpoints, large increases in the speed of data acquisition are to be expected. The simultaneous availability of data on a large number of metabolites also is likely to increase the sensitivity.
Biomarker discovery: The qualitative and quantitative analysis of the metabolome in vitro opens the opportunity for discovering biomarkers, which could be used for diagnostic purposes; other metabolites may be useful as biomarkers for the efficacy of drugs, and/or they may help to quantify the progression of human relevant diseases or the extent of organ damage.
Contribution of cell-cell interactions: a still unsolved large challenge of in vitro toxicology is the understanding of communication between cells that contributes to adverse effects. This is particularly important for interactions involving inflammatory and non-parenchymal cells. For instance, interaction between neurons and glial cells (Falsig et al., 2004;Henn et al., 2009, turbed by the chemical. Such information would help to pinpoint potential targets of the chemical and drugs and to predict their mode-of-action, as demonstrated by recent studies (Strigun et al., 2011a(Strigun et al., , 2012. Recently, simple metabolome analysis was shown to be useful to classify drugs into MoA classes (Strigun et al., 2011b).
Pathways of Toxicity (PoT):
In an extension of the mode-ofaction approach, metabolomics can be used for the mapping of toxicity-related pathways (Hartung and McBride, 2011;Hartung et al., 2012;Shintu et al., 2012). The essential challenge is the identification of the critical pathways that lead to toxicity, as opposed to other chemical-induced changes that are only adaptations, counter-regulations, or epiphenomena (Andersen et al., 2011;Bhattacharya et al., 2011;Boekelheide and Andersen, 2010;Hartung et al., 2012). The pathways of toxicity (PoT) may be specific for the cell types and model systems, but some may allow general predictions from in vitro to in vivo.
Organ specificity of toxicity: One of the great expectations is that information from different in vitro models will allow predictions of potential target organs of toxicity in vivo (Zidek et al., 2007). This would require the identification of responses and of the activation of PoT in a concentration-dependent manner and in different systems predictive of processes in various organs. Moreover, background information on the relevant metabolite changes or activation of PoT in vivo would be required for various target organs, as well as for various modes of toxicity that may affect them (e.g., classifying hepatotoxicants as producing cholestasis, hyperlipidosis, or necrosis in the liver).
Points of departure (PoD): The PoD is a concentration of a test chemical that results in a significant change in the in vitro system, which is considered predictive for the in vivo situation. The PoD is used for in vitro-in vivo-extrapolation calculation (IVIVE) to determine the relevant in vivo dose or exposure. Upon exposure of a model system to a chemical, multiple changes take place. The shift in metabolite patterns will depend on the test concentration and exposure time. It will be important to identify for each model the type of change (i.e., the combination of metabolites and the extent of their change) that predicts toxicity. The conditions leading to these changes (i.e., the free concentration of the chemical or its key metabolite) will be used as points of departure (PoD) to extrapolate findings mathematically to in vivo doses.
Xenobiotics-metabolism: Not only may endogenous metabolites be detected, but other questions may be addressed as well: What are the concentrations of different xenobiotics' metabolites, and how do they change over time?
Intraspecies variability: Human cells with different genetic backgrounds may vary in their responses to toxicants (Ingelman-Sundberg, 2001). Metabolomics may be useful to identify such differences. Information gained therefrom will be useful to model subpopulations with different susceptibilities.
Species differences: In the case of distinct reactions of cells from different species, metabolomics may help identify the reasons and consequences of inter-species variability. Such knowledge would improve species extrapolations, e.g., from rodents to man. endogenous metabolome) or extracellularly (the cell secretome). For instance, leakage of metabolites from cells can be a problem. Large differences exist between the analysis of the intracellular and extracellular metabolites. For intracellular metabolites, leakage from cells or cell debris can be controlled by gentle centrifugation of the samples to avoid contamination of the supernatants. In the second case, which is actually the most challenging, the processing must be carried out with high speed in order to avoid changes of the intracellular metabolite concentrations but also contamination with exogenous metabolites from cell culture media. Therefore, washing steps have to be included, which can delay the processing of the samples. Different washing procedures could already have different effects on cellular metabolites. The choice of sampling time points poses particular challenges, and the cellular reaction to toxicants changes over time by the activation of counter-regulatory pathways.
Speed of sample processing prior to quenching also is a crucial step, since metabolite concentrations can drastically change in a very short period of time. Therefore, the fast "freezing" of the biochemical processes by sample quenching is of high relevance for obtaining reliable data. In addition, depending on the in vitro system, more technical steps could be included, affecting the quality and consistency of the analyzed metabolites.
Sample throughput: The strength of in vitro systems is the control and easy variation of parameters. To fully profit from these features, a large number of samples need to be measured. Increase in sample throughput (with respect to sample preparation, measurement, and analysis) is a major factor determining future widespread use of in vitro metabolomics.
Sample size: As for other -omics approaches, undersampling (too low sample number) leads to overfitting. Simply put, statistical tables contain too many columns (endpoints) compared to rows (sample replicates). The choice of the right sample size is essential for a conclusion regarding whether a marker behaves differently from the controls or not. The sample size should be determined from preliminary experiments in which different sample replicates are set and the internal variability among samples is used to estimate the number of replicates to achieve statistically significant results. Statistical rules suggest that the sample size needs to be matched to the number of metabolites and to the required statistical power. For the estimation of the right sample size for metabolomics approaches, some in silico tools can be recruited, e.g., the programs nemaed samr, ssize, and ssize.fdr. Notably, the variables for these estimation tools have to be well chosen, e.g., number of measured metabolites and the relative abundance of the metabolite concentrations. In practice, it often will not be possible to adhere completely to the stringent rules of statistics. Compromises need to be found that still allow technical feasibility.
Reproducibility: In order to identify the robustness and reproducibility of the system it is essential to understand the baseline metabolite pattern of the cell system under the standard conditions of culturing and prior to a toxic insult. Only by doing so can reliable parameters be defined for comparison 2011; Kremer et al., 2010;Schildknecht et al., 2012;Hirt et al., 2000) involve multiple metabolite exchanges. Communication in co-cultures may be bidirectional, and the overall response would not be understood from the reaction of single cells alone (Gantner et al., 1996). Metabolomics approaches may help elucidate chemical communication between cells in the context of adverse reactions.
3Rs aspects (Reduction, Refinement and Replacement of animal testing): The combination of metabolomics with good in vitro models has great potential for the field of 3Rs (Hartung and Leist, 2008;Leist et al., 2008a,b). In vivo metabolomics already substantially contributes to the 3Rs principle by reducing animal testing through refinement (Kolle et al., 2012). In vitro predictions may lead directly to the replacement of animals, as well as to the improvement of the chemical risk assessment of pharmaceuticals and environmental toxicants. The rich information also would help to optimize in vivo testing. For instance, relevant endpoints could be chosen and the study design optimized. This would lead to a reduction in the use of animals. The use of early biomarkers also would shorten studies and thus lead to a refinement.
Investigation strategy
Two different strategies may be followed for the use of metabolomics in safety evaluations. The more long-term perspective is that a large set of rich data, comprising metabolomics and transcriptomics information, would be sufficient on its own to predict potential hazard. One requirement would be broad background knowledge of systems toxicology and the human toxome. This would be realistic in the more distant future.
In the immediate future the strategy will not be based on vast biological background data but rather on pattern comparison. Reference compounds will be tested in an in vitro model battery. The metabolome analysis of unknown compounds then will be aligned with the known pattern of the reference material. Here also, data from other -omics approaches can be implemented into the alignment pattern ( Fig. 1).
Technical challenges
To promote the widespread application of in vitro metabolomics, several technical challenges need to be solved.
Quality control: Metabolomics is particularly challenging with respect to quality control, as the data set obtained is the result of a multi-step process. Each of these steps can create potential artifacts. It also should be noted that the in silico handling of large amounts of data requires a defined quality assured workflow. Also, sample preparation steps are critical, as the desired metabolites are usually embedded in biological matrices. Thus, metabolites have to be extracted without compromising their structure and concentration. Some metabolic processes are so fast that the metabolite pattern may change during sample preparation. Every in vitro model comes with its own particular issues concerning quality control. Thus, guidelines and SOP both require continuous adaptation.
Sampling: Different challenges apply depending on whether samples are collected intracellularly (cell lysates reflecting the ly constant, but the connections originating from them may change (altered metabolic flux).
In vitro-in vivo extrapolation (IVIVE)
The ultimate challenge of the in vitro metabolomics approach is the extrapolation of the in vitro data to obtain relevant information for the in vivo situation. This will require further advances in the field of physiology-based pharmacokinetic modeling (Blaauboer et al., 2012;Leist et al., 2012b;Louisse et al., 2012;Prot and Leclerc, 2012).
More immediate goals will be to provide qualitative information on, for instance, what a potential target organ may be or whether developmental toxicity is to be expected (Kleinstreuer et al., 2011). The overall vision is that in vitro metabolomics facilitates qualitative in vivo predictions. For instance, key metabolites (or groups thereof) may be selected that predict in vivo toxicity (Yoon et al., 2012). Their concentrations would be used to define benchmark concentrations to be used as the point-ofdeparture for IVIVE. With the definition of the points-of-departure and the employment of PBPK-modeling (ADME), the NOAEL for the in vivo situation can be calculated.
Identification of PoT by using metabolomics
The challenge of information-rich technologies (high-throughput and high-content, for overview see (van Vliet, 2011)) is to make sense of extremely large datasets. This requires the integration of data, likely from different technologies and test systems (Leist et al., 2012b). Systems biology proposes to make use of our increasing understanding of the biological systems, i.e., how the different endpoints are physiologically interconnected. In the end, it attempts the modeling of the dynamics of the biological system (especially on a biochemical and molecular biology level) and of its response to perturbations such as disease. For toxicology, an analogous approach, i.e., a "systems toxicology" could be envisaged (Hartung et al., 2012) where the impact of an agent on the biological system is modeled. This concept represents an extension of moving from black-box models of effects (from apical endpoints), where effects are recorded without understanding the underlying mechanisms, to an approach based on knowledge of the MoA. The 2007 report of the US National Research Council has called for exactly this (NRC, 2007). The buzzword "Toxicology for the 21 st Century" (Tox-21c) or similarly "Toxicity Testing for the 21 st Century," has been taken up to describe the variety of activities implementing the report. Among them an NIH funded initiative to map the human toxome by systems toxicology is attempting to create a process for pathway-of-toxicity annotation and sharing ( [URL]). To enable a systems toxicology approach and to allow quantitative modeling, we have to move beyond a rather qualitative MoA knowledge, and rather describe molecularly defined pathways. The abbreviation PoT for a pathway-of-toxicity has been coined when applying a toxic compound and changing the metabolomics profile.
Sensitivity: Cellular metabolites can be present at concentrations spanning at least 6 orders of magnitude, but they cannot be amplified like DNA, and they are chemically much more diverse than proteins. This poses a particular analytical challenge for metabolomics, and the sensitivity of the method to cover low abundance metabolites needs to be increased considerably.
Normalization: Normalization allows reduction of the potential variability among replicates or experimental samples, e.g., due to viability changes of cells. Normalization will correct for slight changes in cell harvesting, during the extraction procedure, during sample processing, or in detection sensitivity. Normalization parameters are essential for analysis and comparison of in vitro metabolomics data. Different options in this regard include the use of external standards, such as protein concentrations, quantification of cell number, as well as the use of internal references, such as "housekeeping" metabolites (Ruiz-Aracama et al., 2011). Instead of external standards or internal "housekeeping" metabolites, intra-sample normalization based on overall metabolite quantity could be performed. For this procedure, exclusion of contaminants and artifacts is crucial. For instance, plasticizers may be present in varying amounts in the "metabolite" spectrum. Such contaminants would spoil normalization to a sum of total metabolites.
Metabolome coverage: Metabolites are a chemically extremely diverse group of compounds. They range from highly charged phosphoesters or organic cations to extremely hydrophobic lipid constituents. Moreover, many metabolites exist as isomers or epimers that need to be separated. At present, a combination of analytical approaches is used to target metabolites with different physiochemical properties. This requires different analytical technologies, e.g., NMR, GC-MS, UPLC-ESI-MS/MS, TLC/GC-FID, DFI MS/MS. A big technical challenge is to optimize the technology in a way to allow analysis of all metabolites with only few methods.
Database for metabolites: A metabolomics-specific database is still lacking, but some technology providers, such as Agilent and Bruker, have started providing first solutions. These need to meet several challenges, i.e. (a) identification of metabolites: It is still common that the analysis yields a large number of metabolite peaks that cannot be unequivocally assigned to a chemical structure; (b) assignment of identified metabolites to known metabolic pathways or PoT; (c) combination of metabolite information with other data, e.g., transcriptomics.
Pathway analysis: The traditional perception of a metabolic pathway is a sequence of steps leading from an educt to a product. The analysis of metabolic pathways aims to determine the concentration and the fate of the relevant molecules at every stage of the procedure. The challenge arises from the fact that pathways are not linear, one-way roads but rather should be seen as parts of an intricate metabolic network. In this sense, each analyzed molecule is a node of such a network. In cells exposed to toxicants such nodes may change (altered metabolite concentration). Alternatively, nodes may remain relative-on pathways. The complete measurement of many endpoints represents, rather, a challenge to this preconception (not to say prejudice) opening up opportunities for new PoT identification or for balancing the relative importance of different PoT.
Workshop participants felt that current metabolomics technologies are largely fit for the purpose of Tox-21c, while there is a tremendous need to (1) define standard procedures for quality control and data reporting, (2) annotation of metabolites and pathways, and (3) quantification of metabolites required for biological modeling.
Technological challenges
A number of technological challenges were identified: Statistical pathway integration: Available methods appear straightforward but current pathway databases may not reflect reality.
Need for flux assessment: Routine metabolomics does not directly report on metabolic fluxes, which are necessary for modeling as discussed above.
Mathematical modeling of cell (patho)physiology: Obviously, this is the holy grail of systems biology, which is only emerging as a discipline. It is still difficult to obtain the required data (forerunners are, for example, the Metabolights repository (EBI at [URL]) or DIXA (at [URL]. dixa-fp7.eu) for toxicogenomics and metabolomics data). It represents a major challenge at a computational level, for which bioinformatics resources need to evolve. The good news is that toxicology is not alone: The entire field of biomedicine is embracing systems approaches, and each discipline benefits from cross-fertilization.
Validation: We have to distinguish here between (a) compound screening (typically based on signatures), which should allow an early-on regulatory use of metabolomics, as discussed in the previous chapters and (b) validating causal pathways for the purpose of Tox-21c. The validation of the former screening approach would be based on gathering data on lots of compounds that are historically well understood and looking for similarity of signatures/predictivity/anecdotal evidence of mechanistic relevance. However, we will not necessarily understand how changes in these biochemical pathways actually cause disease/ pathology. The mechanistic approach of Tox-21c, however, requires interfering with critical points in pathways identified in order to prove causality. This is difficult and laborious, but more convincing than deduction from phenotypic changes. Modeling strategies might bridge the gap, as they would allow virtual experiments to check the plausibility of suggested PoT before validation of a causal role is undertaken.
Translation of PoT findings: Current work on PoT identification is focused on in vitro systems. Therefore, the relevance for in vivo situations and correspondence of PoT will have to be established. Currently, multiple species often are a prerequisite for regulatory acceptance; the translation of PoT between species needs to be established. A similarity of signatures argues for predictivity and mechanistic relevance, but stability of signatures under various experimental conditions and their relevance to humans need to be established. to differentiate PoT from MoA/toxicity pathways, which are typically defined in a narrative way (Blaauboer et al., 2012;Hartung and McBride, 2011). The opportunities lying in such a systems toxicology approach were discussed intensively in the consensus process to a roadmap for replacing systemic toxicological animal testing (Basketter et al., 2012).
The established networks within an organism, which form the basis for modeling in systems toxicology, are based on molecular biology and biochemistry. Transcriptomics in all of its variants, including the increasing use of deep sequencing technologies, is the key approach for the molecular biology part, with a minor additional contribution by proteomics studies. Metabolomics is the core approach for the biochemistry part of this modeling. In this sense the advent of metabolomics in toxicology represents a "kick-start" into systems toxicology.
This can be initially viewed as a mostly academic exercise aimed at the generation of new knowledge that is not aimed at a specific regulatory purpose. However, society has large expectations of toxicology: this science has the potential to identify potential hazards of chemicals, and to provide improved safety to the consumer (Hartung, 2009b;Leist et al., 2008a). This situation calls for the exploitation of new powerful technologies such as metabolomics, and the goal of making regulatory use of this approach should be kept in mind. Early-stage uses, before definitive regulatory decisions are made on the basis of systems toxicology information, could be the screening for high risk compounds. This means that the right questions must be asked early in the process, i.e., to focus testing on substances with a higher likelihood of being identified as a problem.
The concept of PoT is key to the Tox-21c and systems toxicology concept. Ironically, even after some years of discussion, no definition of PoT has been agreed upon, though various such initiatives are on the way. Two very different views prevail at present, as discussed elsewhere (Hartung et al., 2012): (a) PoT represent the cascade of events leading to the perturbation of a system; (b) PoT represent the downstream signaling triggered by perturbed physiology (Fig. 2). Intuitively, PoT is understood as the initiating event (Fig. 3). However, neither metabolomics nor transcriptomics is currently used to assess these early events, instead we typically seek to assess the established new homeostasis under stress (Hartung et al., 2012).
For Tox-21c and systems toxicology we need high resolution sampling to capture time-dependent changes (dynamics) and the dose-response behavior of systems challenged with toxicants. We need to monitor a wide range of phenotypes (hazards). Metabolomics is especially well suited for this purpose as it (1) is most closely related to phenotypic changes representing functional endpoints, (2) assesses many such processes simultaneously, (3) has some protocols that are already broadly fitfor-purpose in terms of throughput, cost, sensitivity, coverage of the metabolome, and reproducibility, (4) achieves the sample throughput required for detailed dynamic / dose response studies, (5) can sometimes be non-invasive (especially NMR and secretome technologies), and (6) is future-proof since an untargeted approach can be employed. The latter means that we do not necessarily remain influenced by established knowledge grate metabolomics information with classical endpoints from clinics, pathology, histology, etc. This poses some difficulties with regard to the time point of sampling. Classic endpoints represent "late" events (Fig. 3). Sampling at time points when these become evident may not be optimal for metabolomics endpoints and the identification of activated PoT (these are rather early events). Actually, we might need to control for the occurrence of late, generally-degenerative events as confounders for PoT identification, "taking them out of the equation" by measuring, for example, at subtoxic concentrations or early, before functional manifestations.
Taken together, metabolomics is core to the implementation of the Tox-21c concept. It will be a workhorse for PoT identification and possibly later for the testing of PoT activation/ perturbation as it is multiplexing information on various PoT.
Identification versus application of PoT
We should keep in mind that the future use of PoT may be much simpler than the methods used to find them: Ultimately, identified PoT should allow the design of rapid and targeted assays, e.g., for high throughput platforms. Metabolomics will not be the stand-alone approach to identify PoT. The combination with transcriptomics can help resolve relevant pathways, as metabolites typically play a role in several pathways. Metabolomics could be used to screen for candidate PoT, which are targeted in subsequent assays. The question arises whether metabolomics should be prioritized over other -omics for PoT identification. Many of the aspects discussed above argue in favor of this, including the low costs and high throughput once the method is established and the relative ease of interpreting metabolite changes. However, there is a strong need to inte- Fig. 2: The role of in vitro metabolomics in identification, mapping, and use of pathways-of-toxicity (PoT) and hazard prediction Xenob ot cs (X) can be metabo zed/metabo ca y act vated (X*), transported nto d fferent ce compartments, and nteract as parent compounds or as metabo tes w th endogenous targets (T). The nteract on w th some targets forms the mo ecu ar n t at ng events (MIE) that tr gger mmed ate ce u ar changes re ated to metabo sm, s gna ng, and/or transcr pt on. These very n t a steps are somet mes c rcumscr bed as upstream PoT. In an attempt to re-estab sh homeostas s, and as consequence of the n t a d sturbance, severa weconserved ce u ar react ons are tr ggered that dec de on the overa ce fate. For nstance, severa stress response pathways (SRP) are act vated. In add t on, ce death programs are act vated and/or funct ona oss s observed (e.g., uncoup ng of m tochondr a, oss of ATP, nab ty to buffer ntrace u ar ca c um). The atter two events favor/augment tox c ty (TOX). TOX and SRP can nteract n many ways, e.g., the p53 pathway s n t a y a typ ca SRP but can a so tr gger apoptos s when over-act vated. The d fferent nputs from SRP and TOX pathways dec de whether the ce can buffer the damage (re-estab sh a new homeostas s) or whether t oses funct on/v ab ty rrevers b y. B omarkers-of-tox c ty (BoT) dea y corre ate w th the ce fate sw tch. In many cases, they are not s ng e mo ecu ar events, but react ons of a network that requ res mode ng. The concentrat on of a xenob ot c that eads to a break ng of ce homeostas s can be used as po nt-of-departure (PoD) for quant tat ve r sk assessment. The network of events that enta s the ce u ar react on to nsu t and eads to the ce fate dec s on can be termed "downstream PoT." A metabo om cs approach, f part cu ar y we su ted to dent fy and measure the who e metabo te network re ated to downstream PoT.
Regardless of input from regulators, scientists using metabolomics should strive towards: 1) a high quality study design, 2) the development of appropriate standard operating procedures (SOP), and 3) a high level of standardization. Once the method used is well described, it is important to follow the developed SOP strictly in order to minimize changes over time and to ensure comparability of results. Thus, overall attention to quality management will be one of the essential features for laboratories using metabolomics, and it will lead to increased confidence in the approach from risk assessors.
The participants of the working group felt that it would be useful to obtain validation for metabolomics, but given the lack of standardization, for the time being it may not be useful to try to achieve complete validation of all elements of MoA identification with the metabolomics approach. The evidence-based toxicology (EBT) initiative may provide alternative ways to evaluate test performance (Stephens et al., 2013). For instance, procedures have been suggested for high throughput screens that may be used as models for the evaluation of the usefulness and robustness of metabolomics approaches (Judson et al., 2013). Participants were of the opinion that, as regulators The major step to convert metabolomics information into high throughput test systems is the transitioning from a largely untargeted PoT identification to the targeted measurement of the predictive metabolite changes that are characteristic for known PoT.
The road to regulatory acceptance of metabolomics approaches and data
The transition of "omics" technologies from basic to applied research may yield approaches that drastically improve our ability to conduct both predictive and diagnostic assessments of chemical toxicity and increase the efficiency for development of new drugs. In addition, information from omics technologies can improve the regulatory assessment of the safety profile of new compounds. However, regulators need to be convinced about the validity of such data. Here, policy makers play an essential role in speeding up the acceptance of these approaches for regulatory purposes. In order to achieve this, a major effort should be undertaken to design validation strategies tailored for omics technologies.
Today, a key challenge for the regulatory framework is to adapt more flexibly to rapidly-emerging technologies while at the same time ensuring safety for humans and the environment. However, the onus for the integration of these new data also rests with the researchers, who have a responsibility to objectively convey the strengths and weaknesses of the underlying techniques and to work in conjunction with regulators for the validation of these new methods. The main issues discussed at the workshop are summarized below:
Is the current state of the art sufficient to identify modes of action?
In all case studies presented at the workshop symposium, metabolomics analysis was able to reliably identify toxicological MoA. This was independent of the technological platform (e.g., mass spectrometric or nuclear magnetic resonance spectroscopic identification of the metabolites). Therefore, the question of whether metabolomics is suitable for MoA identification was affirmed.
There is a need to discuss with regulators on a case-by-case basis as to whether the evidence obtained with metabolomics is sufficient for identification of mode of action. One issue may be that, currently, neither standardization of metabolomics methods nor guidance on how this could be achieved is available. As the identification of MoA is not a mandatory regulatory ("standard") requirement and also not a toxicological endpoint, it is, at present, included in studies only on a voluntary basis. However, MoA identification is becoming more important in regulatory frameworks. For instance, the identification of endocrine disruptors is one of the targets of both EU and US legislation. Therefore, it is clear that knowing the MoA of a chemical will result in a better interpretation of the toxicological data, and it is likely to contribute positively to the entire risk assessment process (van Ravenzwaay et al., 2012), for example, by addressing species-specificity (Forgacs et al., 2012).
Fig. 3: Activation of pathways-of-toxicity (PoT) as part of the cell injury response
The response of ce s/t ssues to tox c nsu t may be d v ded nto d fferent phases. F rst (1), the network of d sturbances and stress response pathways (co ect ve y termed PoT) s tr ggered and dec des on the ce fate. After a ce has reached the po nt-ofno-return towards death, st many b ochem ca react ons are act vated (2). These are respons b e for degradat ve events and the response to njury that eads to the c ass ca (ap ca ) endpo nts n tox co ogy (e.g., nflammat on). Thus, c ass c endpo nts represent " ate" events. A th rd phase enta s most y pass ve processes, such as d s ntegrat on dr ven by many types of proteases and pases. Metabo om cs approaches wou d measure the act vat on of PoT and a ow pred ct on of tox c ty ndependent of the unspec fic ate changes determ ned by events n phase 2 and 3. of parameters being changed randomly. Bayesian statistics and considerations of biological relevance based on existing background knowledge are required to define meaningful endpoints. Experience already has shown that some MoA can be detected with a relatively low number of consistently altered parameters. To better evaluate the consistency of an effect it would be desirable to investigate multiple time points, but this is not always possible. If a series of parameters is found to be changed consistently, and if these parameters are known to be associated with a known pathway, then this set of changes would constitute a metabolomics effect useful for consideration as a toxicological endpoint. However, not all such metabolomics patterns need to result in pathological conditions or adverse effects in general. For instance, liver enzyme induction correlates with a specific metabolome pattern but does not necessarily result in functional or structural damage. With increased knowledge of the significance of metabolomics pathways, compensatory reactions may also become visible. These may represent good (in vitro) biomarkers of toxicity -BoT (Blaauboer et al., 2012), and they need to be taken into account for systems biology models of overall adverse outcomes (Hartung and McBride, 2011;Hartung et al., 2012). In addition, metabolomics could add to our understanding of the difference between compensatory reactions (adaptation) and those changes that are linked to cell fate decisions. Indeed, metabolomics patterns might constitute useful BoT (Blaauboer et al., 2012) that can help defining appropriate NOAEL.
In summary, the identification of altered metabolic pathways by metabolomics approaches does not necessarily mean that they lead to an adverse outcome. Consequently, for the time being, metabolomics is not a stand-alone approach in toxicology; it needs and can be matched with other toxicological data. An interesting and fruitful approach is to correlate metabolomics effects (patterns of change) with adverse toxicological outcomes, and to develop prediction models. Moreover, the relevance of reversibility is not yet clear for metabolomics parameters, and requires further studies.
In contrast to metabolomics data, the relevance of transcriptomics findings is often less clear, as these changes rarely can be linked directly to phenotypic changes. From a statistical point of view, transcriptomics is also more problematic than metabolomics because there are many more parameters relative to the sample size. But the combination of both transcriptomics and metabolomics may significantly enhance data interpretation, especially when results from time series experiments are available.
The participants of the working group recommended building a data base using metabolomics data from regulatory studies in order to validate its use for predicting adverse effects and/or identifying MoA. Using samples from regulatory studies would provide the necessary standards to correlate changes of metabolomics data with classical toxicological parameters.
ECETOC's guidance to derive a meaningful NOAEL recommended that (1) only specific patterns of change (in any type of -omics study) should be used to conclude that a potentially relevant biological effect is taking place, (2) as changes in -omics pathways do not necessarily implicate that changes at become more familiar with metabolomics, they are likely to recognize the value and advantages of this approach. They might then request its more frequent use (as has happened for markers of kidney toxicity). One of the additional advantages of metabolomics would be that it could put species comparisons (e.g., rat, mouse, human) on a more solid data basis. Metabolomics also can contribute to the assessment of additive or synergistic effects in co-exposure scenarios for both pharmaceuticals and environmental toxicants, which are more the rule than the exception in daily life.
A future perspective might be deduced from knowledge about other related (-omics) technologies. First examples for the use of transcriptomics can be found in the development of new pharmaceuticals, as well as in the safety evaluation of genetically modified crops (EC, 2011). The value (credibility) of MoA determined by fingerprints or biomarkers can be confirmed if the changes observed can be causally linked to toxicological pathways. It should also be noted that -omics data could be obtained routinely from regulatory studies, thus reducing the need for additional experiments and providing a highly standardized experimental setup of the biological study. To further enhance the acceptance of metabolomics, a careful design of biological experiments and high quality data are essential (e.g., appropriate biological model, treatment regime, and sampling method). In addition, proper controls, reference compounds, phenotypic anchoring as well as appropriate validation procedures should be used to ensure the quality of the generated data. Overall, metabolomics appears to be ready to be incorporated into regulatory testing as an additional robust source of relevant information, in a toxicological weight of evidence approach.
Definition of metabolomics no adverse effect level (NOAEL)
One of the critical elements of any regulatory study is the determination of a NOAEL. Sufficient guidance is available for experienced toxicologists to consistently determine a NOAEL based on the classical parameters observed in standard toxicological studies. However, for new technologies, such as metabolomics, there is very little guidance available. The absence of guidance criteria on how to determine a NOAEL in metabolomics is a hurdle for introducing such studies within a regulatory context. Therefore, defining criteria or providing guidance on metabolomics NOAEL setting is of utmost importance. It would reduce planning insecurity, especially for management decisions driven by financial factors and considerations of the time-to-market. Currently, there is only general guidance on how to determine a NOAEL in -omics studies from two ECETOC workshop reports (ECETOC, 2008) -see note on "ECETOC guidance" at the end of this section.
For scientists involved in metabolomics, or any other -omics approach, it is clear that with hundreds or even thousands of parameters measured, a single parameter cannot be used to determine a NOAEL. Classical stochastic-based statistical methods would result in an overly high false discovery rate, and a large degree of unreliability. Refined statistics can resolve this problem to some extent. For example, the use of false discovery rate corrections can be introduced to estimate the probability higher than the biological variability encountered in controlled animal experiments. Therefore, much larger sample sizes and enhanced sub-grouping of the population are needed. Human variability also will be an important factor to be taken into account when translating metabolomics findings from animal studies to humans. Again, standardization will be very important, but factors such as lifestyle, diet, disease state, etc. will inevitably introduce elements of variability.
For in vitro studies, the situation often is more complex than might be expected. As cell culture procedures can involve many steps, variability introduced by the experimental setup may be quite high. Initial experiences of several participants suggest that variability associated with in vivo systems may be less than that associated with in vitro systems.
The participants concluded that, due to lack of standardization, currently no general guidance can be provided for evaluation of the variability and that each individual researcher needs to assess the variability of their system/procedure. Guidance for adequate study design (e.g., how to determine adequate group sizes) based on statistical considerations (e.g., strength of the effect, prevalence, etc.) would be helpful. Such adaptations of the study design are not easily possible for regulatory studies that follow a strict protocol and build on historical background data. With a more mid-term or long-term perspective, regulatory study designs from already existing protocols may need to be changed to allow incorporation of modern endpoints such as those from the metabolomics approach.
Validation of metabolomics for toxicological regulatory purposes
In view of the many opportunities that metabolomics has to offer for toxicology, particularly in terms of identifying MoA, it would be desirable to make the metabolomics approach acceptable for regulatory purposes (Fig. 4). This would require some sort of validation process, as it is common for any other new method.
Assuming that the regulatory use of metabolomics for the time being would concentrate on MoA identification it was recommended by the workshop that each individual metabolome pattern, indicating a particular MoA, should be validated. In order to ensure that adverse outcome pathways would be addressed in such an exercise, it would be necessary to clearly demonstrate a good correlation with toxicological effects such as pathology. Plausibility of the metabolomics changes and observed toxicological effects should be one of the key elements for validation. Before the start of any type of regulatory validation, it seemed advisable first to consult with regulators to explain the usefulness of metabolomics in a regulatory context and to ensure that, following validation, such data would also be acceptable for regulatory purposes (Fig. 4). This requires first of all more communication with regulators and the publication/communication of success stories. It also will become more important to reach out to those more involved in regulatory and risk assessment aspects of toxicology. One example for regulatory acceptance is the altered metabolomics biomarker pattern for the detection of certain types of kidney damage (Dieterle et al., 2010;Fuchs and Hewitt, 2011). cellular, individual, or population levels will necessarily occur, these pathways need to be correlated to observable histological changes at the microscopic or macroscopic level, and (3) to use changes in an -omics pattern for NOAEL purposes, it must be assured that the pathway identified is related to an adverse effect (ECETOC, 2010).
Dealing with inherent variability during the use of metabolomics for toxicological purposes
There are two major sources, namely, technical and biological, that contribute to the overall variability, and they need to be handled separately. Technical variability results from the analytical process, starting with sample preparation and ranging to the separation and detection of metabolites. Optimization of procedures, the use of quality control samples, as well as compliance with SOP and the exact monitoring and documentation of observed deviations from SOP protocols, can help to reduce this variability. Randomization of samples, and quality control also are important measures to reduce variability. The second source of variability is the one inherent to the biological system used. Here also, standardization and the development of SOP will help to reduce variability. Moreover, it has been noted that each additional step in the experimental protocol will introduce more variability. Therefore, reducing complexity is essential. The risk of high variability is that it can mask subtle but important effects and thus reduce the sensitivity of the technology in obtaining biologically relevant data. As indicated above, variability is associated with the protocols and SOP used, therefore variability needs to be determined and defined for each individual "test system," and only then is it possible to decide how the test system can be used, i.e., which questions can be addressed and which cannot in terms of signal to noise ratios. At high noise levels (= high variability), only large signals can be studied. For example, in a study using different rat strains, the metabolome patterns and MoA induced by 2-methyl-4-chlorophenoxyacetic acid were still clearly visible, despite the additional noise introduced by using different rat strains. Weak changes, as those associated with anemia, were less clear when using different rat strains (Strauss et al., 2009). With increasing knowledge of how metabolites respond to different confounding factors such as reduced food consumption, dietary changes, age, etc. such effects can be recognized and compensated for. New statistical methods also allow the identification of outliers in -omics studies and thus help to reduce variability in experimental groups in which, e.g., one animal behaved quite differently from the rest. Therefore, statistical models need to be developed that have the capability of "learning." This means that recursive cycles of new data generation and improved analysis will improve the already existing model and make it more and more accurate.
The use of metabolomics in human samples is highly attractive because relevant body fluids such as blood or urine can be easily obtained. One example is collection of specimens in national bio-monitoring banks, such as the German environmental specimen bank, where sample specimens are stored and can be re-analyzed in the future. However, human variability is much outcome of the studies by regulatory agencies. Certainly the latter would require that regulatory agencies be more familiar with the metabolomics approach. Ideally, they would use it themselves to better understand the strengths and weaknesses of this approach and to build confidence in the data obtained. The working group believed that regulators would hardly accept metabolomics data unless they have gained their own experience with this technology.
The question was asked whether the regulatory framework for metabolomics should be different for pharmaceutical active ingredients, pesticides, or industrial chemicals. There was agreement that the regulatory framework should be identical for all sectors, as far as identification of the MoA in toxicology is addressed. For some special modes of action, e.g., endocrine disruption, there is a regulatory demand for identifying them (Hecker and Hollert, 2011). Consequently, MoA identification by means of metabolomics should be attractive, as this could be done without additional/animal studies (Fig. 4) by using various biomatrices (e.g., blood and urine) from regulatory studies (van Ravenzwaay et al., 2010;Zhang et al., 2012a). For the time being, the integration of metabolomics data into a regulatory decision-making framework may be limited to MoA identification for the three sectors. It was noted by some participants that, in the absence of any toxicological findings, which is not uncommon for certain classes of industrial chemicals (evaluated under REACH), there is no merit in MoA identification by metabolomics (or any other approach). For pharmaceutical compounds, there could be more (regulatory) options for the use of metabolomics, particularly with respect to human relevance and the comparison of metabolite responses in different species. A further practical application of metabolomics in a regulatory context is its use in diagnostics and food quality evaluation (Shepherd et al., 2011).
An aspect of metabolomics that has not received much attention, but could be very attractive for both research and regulatory purposes, is the fact that metabolomics data include information on both normal constituents of the organisms tested and on the test substance and its metabolites. Additionally, by integrating imaging techniques in metabolomics studies, the obtained results give insights into the actual distribution of metabolite patterns and pharmaceuticals or environmental toxicants and their metabolites within tissue or single cells. Thus, metabolomics could simultaneously provide information on the chemical exposure in the organisms/cells tested and on the perturbation triggered thereby. Although this may require adaptation of the technical equipment, tracking exposure and analyzing internal dose-response relationships is highly attractive. Overall, this will add to the weight of evidence concerning toxicological effects following chemical exposure.
For metabolomics information obtained from in vitro data, an important aspect is the translation to the in vivo situation. This has to be demonstrated before such data can be used in a regulatory framework. Concerning the combination of metabolomics data with information obtained from other -omics technologies (often referred to as systems biology), integration of transcriptomics and metabolomics data has already been shown to be The participants concluded that some guidance needs to be provided on how validation of metabolomics methods could be achieved. This guidance should be developed jointly by multiple stakeholders together with regulators and risk assessment institutions.
How can metabolomics data be integrated into a regulatory framework?
Currently, metabolomics is used mainly for academic research purposes, and only a few companies have started to use this approach for the early identification of toxicological effects. The use and application of metabolomics in toxicology would advance more rapidly if it was also used for regulatory purposes. This would require at least some type of validation protocols (see aforementioned questions) and acceptance of the
Fig. 4: Use of metabolomics in a regulatory context
Two d fferent uses can be env saged for metabo om cs approaches. The first s the dent ficat on of the mode-of-act on (MoA) of compounds. Th s s a potent a stand-a one approach, not necessar y requ r ng add t ona techno og es. A major techn ca cha enge s to keep the var ab ty of the techno ogy and of the samp e preparat on ow. For use n a regu atory context, va dat on of spec fic standard operat on procedures wou d be requ red. At present, nformat on on the MoA usua y s not requ red n the regu atory process and wou d be regarded as supp ementary nformat on. The MoA s a mandatory requ rement on y for few types of compounds (e.g., endocr ne d srupters). The second use of metabo om cs wou d enta the defin t on of the "no adverse effects eve " (NOAEL). Th s wou d requ re add t ona nformat on from other techn ca approaches. At present, th s s not a rea st c use of metabo om cs approaches n the regu atory context, and generat on of more data and ga n ng of exper ence w be requ red to judge the va d ty of the approach. The future acceptance of the method for e ther app cat on w depend on the ntroduct on and rout ne use of str ngent qua ty assurance procedures.
The rapidly emerging use of metabolomics analysis as endpoint for in vitro test systems requires special attention. Often such test systems allow a high throughput and a large degree of control of the experimental conditions. However, the extrapolation of in vitro data to the in vivo situation is still a substantial scientific challenge. In many cases, information from multiple systems may need to be combined to account for tissue effects such as cell-cell interactions, compensatory regulations, and communication between different organs. The interpretation of data from individual systems often requires ample experience. A second major issue is the susceptibility of in vitro systems to experimental artifacts due to poor study design or small variations of the experimental conditions. Therefore, now more than ever, the quality control of study design as well as all the conditions crucial for the good performance of the system must be taken very seriously.
Implementation of metabolomics in the regulatory context will require an intense collaboration among the different stakeholders, whether they belong to academia, industry, or regulatory bodies. It will be crucial to jointly investigate and define the relevance of the changes observed. If this is achieved, then the innovative methodology of metabolomics can be rapidly integrated into the regulatory process to provide more complete information on chemical effects on the physiological/cellular levels, information about the spatial distribution not only of the toxicants but also of specific marker metabolites within whole tissues and single cells, as well as on the safety of humans and the environment.
helpful for increasing certainty in the identification of a specific effect/MoA (Bundy et al., 2008) or for identifying pathways influencing susceptibility to toxicity (Cavill et al., 2011). However, there is a need for development of better tools for data integration and to further advance computer software. This would enable the use of combined -omics data to achieve a more comprehensive interpretation. Emerging technologies include mass spectrometry-based imaging metabolomics approaches like matrix-assisted laser desorption/ionization (MALDI) and time of flight secondary ion mass spectrometry (ToF SIMS). They would provide new insights into intra-tissue and intra-cellular metabolite distribution changes at sub-organ or sub-cellular levels, but data handling and standardization for quantitative analysis may be even more complex.
At present, there is considerably more experience in applying transcriptomics to toxicological investigations. In the future, the use of metabolomics is likely to increase, as the information obtained with this approach is closer to classical toxicological endpoints, and therefore easier to interpret.
Associations that may potentially help with the design of guidance for regulatory use and validations are: WHO, OECD, ILSI, and ECETOC.
Conclusions and recommendations
The use of metabolomics for a better understanding of the pathways and regulations relevant for the toxicity and efficacy of compounds (e.g., drugs, pesticides, industrial chemicals) will bring significant benefits for consumers/patients. However, some challenges still need to be overcome. Dealing with the large amounts of data is still not a trivial task. Moreover, more efforts are required with respect to the standardization of protocols, study designs, data processing, and analyses. Altogether, this will ensure the generation of reliable information that can be compared among laboratories and that can be used to create meaningful databases for their application in many fields of life sciences.
Regarding the application of metabolomics for safety evaluations, two different strategies may be followed. The more long-term perspective is that a large ("information rich") set of data from simple model systems (e.g., human cell cultures) would be sufficient on its own to predict potential hazard. The data would comprise metabolomics and transcriptomics information, and an additional requirement would be broad background knowledge of systems toxicology and the human toxome. This would become realistic only in the more distant future. In the immediate future, the strategy will not require vast systems biology background knowledge, but it will be based, rather, on pattern comparisons ("signatures"). Reference compounds will be tested in an in vitro or in vivo model battery. The metabolome analysis of unknown compounds will then be aligned with the known pattern of the reference materials. Within this approach, data from other -omics technologies also can be implemented in order to refine the predictive value of this strategy.
==
Domain: Environmental Science Biology Chemistry Medicine
|
[{"added": "2016-10-31T15:45:48.767Z", "created": "2013-01-01T00:00:00.000", "id": "10436157", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.altex.org/index.php/altex/article/download/387/383", "pdf_hash": "663a0521f6fe417705f41ae6392114290b2b8680", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:34961", "s2fieldsofstudy": ["Medicine", "Chemistry", "Environmental Science"], "sha1": "663a0521f6fe417705f41ae6392114290b2b8680", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Evolving geographic diversity in SARS-CoV2 and in silico analysis of replicating enzyme 3CLpro targeting repurposed drug candidates
Background Severe acute respiratory syndrome (SARS) has been initiating pandemics since the beginning of the century. In December 2019, the world was hit again by a devastating SARS episode that has so far infected almost four million individuals worldwide, with over 200,000 fatalities having already occurred by mid-April 2020, and the infection rate continues to grow exponentially. SARS coronavirus 2 (SARS-CoV-2) is a single stranded RNA pathogen which is characterised by a high mutation rate. It is vital to explore the mutagenic capability of the viral genome that enables SARS-CoV-2 to rapidly jump from one host immunity to another and adapt to the genetic pool of local populations. Methods For this study, we analysed 2301 complete viral sequences reported from SARS-CoV-2 infected patients. SARS-CoV-2 host genomes were collected from The Global Initiative on Sharing All Influenza Data (GISAID) database containing 9 genomes from pangolin-CoV origin and 3 genomes from bat-CoV origin, Wuhan SARS-CoV2 reference genome was collected from GeneBank database. The Multiple sequence alignment tool, Clustal Omega was used for genomic sequence alignment. The viral replicating enzyme, 3-chymotrypsin-like cysteine protease (3CLpro) that plays a key role in its pathogenicity was used to assess its affinity with pharmacological inhibitors and repurposed drugs such as anti-viral flavones, biflavanoids, anti-malarial drugs and vitamin supplements. Results Our results demonstrate that bat-CoV shares > 96% similar identity, while pangolin-CoV shares 85.98% identity with Wuhan SARS-CoV-2 genome. This in-depth analysis has identified 12 novel recurrent mutations in South American and African viral genomes out of which 3 were unique in South America, 4 unique in Africa and 5 were present in-patient isolates from both populations. Using state of the art in silico approaches, this study further investigates the interaction of repurposed drugs with the SARS-CoV-2 3CLpro enzyme, which regulates viral replication machinery. Conclusions Overall, this study provides insights into the evolving mutations, with implications to understand viral pathogenicity and possible new strategies for repurposing compounds to combat the nCovid-19 pandemic.
Background
In early January 2020, the World Health Organisation (WHO) reported cases of pneumonia of an unknown cause in Wuhan City, Hubei Province of China, and by 30 January 2020, WHO escalated the warning to public health emergency of international concern. By 12 March 2020, the novel coronavirus (nCoV) outbreak achieved a global pandemic status and was recognised as novel
Open Access
Journal of Translational Medicine *Correspondence1 Faculty of Medicine, Health and Human Sciences, Macquarie University, F10A, 2 Technology Place, North Ryde, NSW 2109, Australia Full list of author information is available at the end of the article Page 2 of 15 Chitranshi et al. J Transl Med (2020) 18:278 Covid-19 disease (nCovid-19) [1]. The present coronavirus outbreak is associated with severe acute respiratory syndrome 2 (SARS-CoV-2), phylogeny and taxonomy designated [2]. Worldometer reported the total SARS-CoV-2 infected cases on 31 May 2020 as 6,238,550 and deaths 374,374 worldwide ( https ://www.world omete rs.info/coron aviru s/#count ries). The pathogen has been established to transmit from human to human contact and has quickly spread to more than 187 countries across the globe ( https ://gisan ddata .maps.arcgi s.com/). Coronaviruses are single and positive stranded RNA viruses belonging to the genus Coronavirus of the family Coronaviridae that can cause acute and chronic respiratory and central nervous system illnesses in animals, including in humans [3,4]. The infection can also cause mild episodes of follicular conjunctivitis in certain patients. In animal models, the infection has been shown to induce anterior uveitis, retinitis, and optic neuritis like symptoms [5]. Recent study has shown formation of hyper-reflective lesions in the ganglion cell and inner plexiform layers of the retina particularly around the papillomacular bundles [6]. The disease has also been shown to affect sense of smell and taste bud sensitivity in patients [7]. All coronaviruses have a minimum of 3 basic viral proteins (i) an envelope protein (E), which is a highly hydrophobic protein involved in several aspects of the virus life cycle such as assembly and envelope formation [8] (ii) a spike protein (S), a glycoprotein involved in receptor recognition and membrane fusion [9] and (iii) a membrane protein (M), which plays a key role in virion assembly [10] (Fig. 1). The viral genome also encodes two open reading frames (ORF), ORFa and ORFb that activate intracellular pathways and triggers the host innate immune response [11]. The polyprotein encoded by the virus are initially processed by two main viral proteases, which include a papain-like cysteine protease (PL pro ) and 26,191) and Nucleocapsid (N, nt 28,274-29,533) proteins in green. ORF1a gene encodes papain-like protease and 3CL protease, ORF1b gene encodes RNA-dependent RNA polymerase, helicase and endo ribo-nuclease, S, E, M and N gene encodes spike, membrane glycoprotein and nucleocapsid phosphoprotein respectively. Three-dimensional crystal structure of 3CL-protese, endoribonuclease and SARS-Cov-2 spike protein receptor binding domain (RBD) engaged human angiotensin converting enzyme 2 (ACE2) receptor were collected from protein data bank chymotrypsin-like cysteine protease, known as 3C-like protease (3CL pro ), into intermediate and mature nonstructural proteins [12]. The main proteinase 3CL pro , is one of the primary targets for development in an antiviral drug therapies, as it plays a critical role in the viral replication [13]. K11777, camostat and EST, are cysteine protease inhibitors, which have been shown to inhibit SARS-CoV 3CL pro replication in cell culture conditions [14,15]. Recent release of the high-resolution crystal structure for the main proteinase 3CL pro (Protein Data Bank, PDB ID: 6Y2G), describing an additional amide bond with the α-ketoamide inhibitor pyridone ring to enhance the half-life of the compound in plasma [16] is suggested to accelerate the targeted drug discovery efforts. Two HIV-1 proteinase inhibitors, lopinavir and ritonavir, have been considered to target SARS-CoV [17]. Interestingly, the substrate binding cleft is located between domains I and II of both SARS-CoV 3CL pro and SARS-CoV-2 3CL pro enzymes [16,18].
Since the initial stages of the SARS-CoV-2 outbreak, laboratories and hospitals around the world have sequenced viral genome data with unprecedented speed, enabling real-time understanding of this novel disease process, which will hopefully contribute to the development of novel candidate drugs. The complete genomes of SARS-Cov-2 from all over the world have been deposited at The Global Initiative on Sharing Avian Influenza Data (GISAID) [19] database and more sequences continue to be deposited with the passage of time. Development of a novel vaccine against SARS-CoV-2 so far remains elusive and requires a thorough understanding of molecular changes in viral genetics. This may be attained by freely accessing the GISAID database and processing the data to enhance our understanding of the fine biochemical and genetic differences that differentiate this virus from the previously known strains [20].
It is well known that viruses are non-living and that they require host cells to survive and to reproduce, with the sole aim to perpetuate themselves. When a virus jumps from animal to human, it is termed a zoonotic virus. This occurred during the SARS outbreak of 2002, when a new coronavirus spread around the world and resulted in death of hundreds of people [21]. In 2012, another novel coronavirus outbreak, termed Middle East respiratory syndrome (MERS), caused over 400 fatalities and spread to over 20 different countries [22]. There are currently many circulating viruses, but why SARS-CoV-2 has achieved such a devastating pandemic status and whether this pandemic will subside remain unanswered.
The purpose of this study is to characterise known viral variants that have spread across different countries, especially hot-spot regions, with a focus on recurrent mutations in South American and African geographical regions. We also focused on the SARS-CoV-2 main proteinase, 3CL pro which is highly conserved in most of the coronaviruses and has been suggested to be a potential drug target to fight against nCovid-19. Repurposed drugs such as flavonoids and biflavanoids, known anti-malarial and anti-viral drugs and the inhibitory effects of vitamins could selectively inhibit this enzyme and can be used either alone or in combination with other disease management approaches to suppress the virulence of SARS-CoV-2. These bioinformatics, computational modelling and molecular docking approaches using repurposed drugs could be particularly useful in the current nCovid-19 outbreak.
Collection of SARS-Cov-2 genome
The Global Initiative on Sharing Avian Influenza Data (GISAID) is headquartered in Munich, Germany and is a public-private partnership project between German government and the non-profit organization founded by leading medical researchers in 2006. Since December 2019, GISAID has become a repository storage database for nCovid-19 genome. The genome analysis was carried out for data deposited up to 31 May 2020 ( https ://www. gisai d.org/). Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Wuhan genome was collected from NCBI, NC_045512.2.
Multiple sequence alignment and Phylogenetic tree construction
Multiple sequence alignment (MSA) of all nucleotide sequences were carried out in the EMBL-EBI Clustal Omega server to investigate sequence conservation [23,24]. The Newick format for the multiple align sequence was used to generate phylogeny [25]. The phylogenetic tree was constructed in the Interactive Tree of Life (iTOL) online tool [26]. The iTOL server generate phylogeny trees in a circular (radial) and normal standard trees. The circular trees can be rooted and displayed in different arc sizes [27][28][29].
Structure analysis SARS and SARS-CoV-2 3CL pro
Crystal structure of SARS and SARS-CoV-2 3CL pro with bound inhibitors were collected from the protein data bank (PDB) [30]. PDB ID: 3TNT, SARS main protease was selected as reference to analyse the variants in SARS-CoV-2 3CL pro (PDB ID: 6Y2G). All the PDBs were visualised using UCSF Chimera software [31]. Multiple alignment, ribbon, surface and superimposition module in Chimera software were used for analysis and image generation [24,32].
The dataset comprises of flavones and biflavanoids, anti-viral, anti-malarial and vitamins as SARS-CoV-2 3CL pro inhibitors [16]. In total 17 repurposed drugs were collected from the Pubchem database [33]. Twodimensional (2D) structures were downloaded from the Pubchem database in.sdf format. The inhibitor energies were minimized using the Austin Model-1 (AM1) until the root mean square (RMS) gradient value became smaller than 0.100 kcal/mol Å and later re-optimization was done by MOPAC (Molecular Orbital Package) method [34,35]. Later, all the inhibitors were converted to.pdb format in Open Babel software [36] and submitted to molecular docking studies.
Selection and preparation of SARS-Cov-2 main protease protein (3CL pro )
Crystal structure of the SARS-CoV-2 3CL pro was retrieved from PDB (PDB ID: 6Y2G). The protein macromolecule (SARS-CoV-2 3CL pro ) optimization was carried out in UCSF Chimera software [31,37,38] by adding polar hydrogen atoms, removing water molecules, implying amber parameters, followed by minimization with the MMTK method in 500 steps with a step size of 0.02 Å. SARS-CoV-2 3CL pro contained chain A and B of 306 amino acids sequence length. Chain A of PDB ID: 6Y2G containing alpha-ketoamide (O6K) inhibitor was used for identification of substrate binding site.
SARS-Cov-2 main protease protein (3CL pro ) inhibitors docking studies
The docking of SARS-CoV-2 3CL pro specific pharmacological inhibitors into the catalytic site was performed by the AutoDock 4.2 program [39]. The alpha-ketoamide (O6K) inhibitor was extracted from the SARS-CoV-2 3CL pro protein. The polar hydrogen atoms were added, the non-polar hydrogen atoms were merged, Gasteiger charges were assigned and solvation parameters were added to the protease, SARS-CoV-2 3CL pro protein. The protonation state for all inhibitors and O6K were set to physiological pH and rotatable bonds of the ligands were set to be free. The AutoGrid program was also used to generate grid maps. Cys145 residue in the SARS-CoV-2 3CL pro protein was selected with grid box dimensions of 40 × 40 × 40 Å formed around the Cys145 protease residue, which is present in the substrate binding site. Protein rigid docking was performed using the empirical free energy function together with the Lamarckian genetic algorithm (LGA) [40].
LGA default parameters were used in each docking procedure and 10 different poses were calculated. Chimera and Discovery Studio (DS) Visualizer2.5 [31] software were used for visualisation and calculation of protein-ligand interactions.
Sequences from bat-SARS-CoV and Pangolin-SARS-CoV were aligned and compared to the Wuhan SARS-CoV-2 (NC_045512.2) as a reference genome. To determine the evolutionary relationship among bat-CoV, Pangolin-CoV and SARS-CoV-2, we estimated a phylogenetic tree based on the nucleotide sequences of the whole-genome sequence. Bat-SARS-CoV and SARS-CoV-2 were grouped together and were observed to share > 96% similarity, whereas the Pangolin-SARS-CoV was closest evolutionary ancestor (Fig. 2d). Isolate of human Wuhan SARS-CoV-2 (NC_045512.2) shared 85.98% identity with Pangolin-SARS-CoV which suggests that Pangolin may be associated with SARS-CoV-2 evolution or subsequent outbreak [41,42].
Identification of hotspot mutations in SARS-Cov-2 complete genome from South American and African regions and analysis of main protease (3CL pro ) sequence
Recently, Pachetti et al. [43] has reported eight novel recurrent mutations of SARS-Cov-2 that have been identified in positions 1397, 2891, 14,408, 17,746, same locus indicates the high susceptibility of these genetic regions to change as the virus evolves. For its actions, single-stranded SARS-CoV-2 RNA viral genome encodes two protease polyproteins (i) papainlike cysteine-protease (PL pro ) and (ii) the chymotrypsinlike cysteine protease known as 3C-like protease (3CL pro ). 3CL pro , which is a main protease and therefore important in order to examine the incidence of any mutation in SARS-CoV-2 3CL pro . Multiple sequence alignment of the SARS-CoV-2 genome collected from patients in six different geographical locations exhibited 100% similarity and no discernible variations in sequences obtained from diverse geographical regions, for this enzyme.
SARS-CoV and SARS-CoV-2 similarity
SARS and SARS-CoV-2 complete genomes were collected from NCBI, GenBank database (NC_004718 and NC_045512). Protease nucleotide sequences were extracted from SARS (NC_004718) and were aligned with SARS-CoV-2 (NC_045512). Clustal Omega alignment of 918 SARS nucleotides showed around 95% similarity with SARS-CoV-2 (Additional file 2: Table S2). Higher amino acid sequence identity was also observed in SARS-CoV and SARS-CoV-2 main protease (3CL pro ) derived from Wuhan and US patients. SARS-CoV and SARS-CoV-2 3CL pro showed highly conserved region in both the catalytic sites, His41 and Cys145 [44] and substrate Previously confirmed mutations at positions nt3036, nt8782, nt11083, nt14408, nt23403, nt28144 and nt28881 were also present in South American and African populations. We normalize the mutation frequency percentage by estimating the frequency of genomes carrying mutation and comparing it with the overall number of collected genomes per geographical area. The graph shows the cumulative mutation frequency of all given mutations present in South American and African regions. Mutation localisation in viral genes are reported in the legend as well as the proteins (i.e. non-structural protein, nsp) presenting these mutations. b It is also evident that South American and African clusters show a differential pattern of novel mutations: mutation 1059 (black), 9477 (pink), 28,657 (green) and 28,878 (red) in South American, whereas mutation 1059 (black), 15,324 (orange), 28,878 (yellow) and 29,742 (magenta) are present with greater frequency in African patients (Fig. 4a), inferring that these proteases exhibit high similarities. Furthermore, 12 variant positions (Thr35Val, Ala46Ser, Ser65Asn, Leu86Val, Arg88Lys, Ser94Ala, His-134Phe, Lys180Asn, Leu202Val, Ala267Ser, Thr285Ala and Ile286Leu) were observed in SARS-CoV-2 3CL pro (Fig. 4b, c). The effects of mutations and potential resultant amino acids on SARS-CoV-2 3CL pro structure are expected to conserve the polarity and hydrophobicity, except when the resulting amino acid is Leucine at 286 position. However, it is important to mention that these 12 variants are not present in catalytic and substrate binding regions which are involved in critical proteolytic activity of the SARS-CoV-2 protease molecule.
Docking study of SARS-CoV-2 3CL pro inhibitors
The SARS-CoV-2 3CL pro receptor binding pocket was determined by superimposing SARS and SARS-CoV-2 3CL pro with their respective inhibitors (Fig. 4). Interestingly, Needleman-Wunsch alignment algorithm and BLOSUM-62 matrix analysis revealed 94.44% sequence identity between SARS (Fig. 5a, grey) and SARS-CoV-2 3CL pro (Fig. 5a, Cyan). Cys-His catalytic dyad (Cys145 and His41) comprises the active catalytic binding site in SARS-CoV-2 3CL pro (Fig. 5a' , b) and indicated the strong possibility that intended pharmacological inhibitors of SARS-CoV-2 3CL pro may also suppress the activity of SARS-CoV-2 3CL pro viral enzymes. Docking protocol for the Autodock 4.2 program was optimized by extracting and re-docking the alpha-ketoamide inhibitor named O6K in the binding pocket of SARS-CoV-2 3CL pro . The lowest binding energy − 6.45 kcal/mol and 18.72 µM inhibitory constant (Ki) was predicted for alpha-ketoamide inhibitor (shown in Table 1). Re-docking of O6K inhibitor occupied the similar docking pose in the SARS-CoV-2 3CL pro catalytic dyad active site as previously reported in the crystal structure (PDB ID: 6Y2G) (Fig. 5c, d).
Seven flavonoids and biflavonoid, three anti-malarial compounds, seven anti-viral drugs and three vitamin molecules were subjected to automated docking within the active site of SARS-CoV-2 3CL pro catalyticdyad. The superimposition of all docked flavones and biflavones (Fig. 6a), anti-malarial drugs (Fig. 6b), antiviral drugs (Fig. 6c) and vitamins (Fig. 6d) are shown in Fig. 6 and various binding parameter have been tabulated in detail in Table 2.
Amentaflavone, a biflavonoid showed the highest binding energy (− 8.49 kcal/mol) implicating a strong affinity with SARS-CoV-2 3CL pro . This corresponded with previously reported enzyme inhibitory assays with amentaflavone that showed the highest IC 50 value at low concentrations of the molecule, 8.3 ± 1.2 µM [46]. However, bilobetin demonstrated the lowest IC 50 value at a higher concentration of 72.3 ± 4.5 µM in SARS-CoV enzyme activity assays [46]. In contrast, our docking studies revealed that bilobetin, predicted almost comparable binding energy with that of amentaflavone (− 8.29 kcal/mol) suggesting that mutation in SARS-CoV-2 3CL pro could potentially disrupt hydrogen bonding or induce some conformational change that could result in alterations in the binding site thus affecting inhibitor interactions with the enzyme active site residues. Amentaflavone showed H-bond interactions with the catalytic dyad residues (Cys145 and His41) as well as noteworthy interactions with the SARS-CoV-2 3CL pro residues Thr26, Ser46, Ser144 and Glu166 whereas His164, and Gln189 amino acids contributed to the hydrophobic interactions for the SARS-CoV-2 3CL pro inhibitors (Fig. 7a). Three antimalarial drugs were then selected to study their inhibitory actions on SARS-CoV-2 3CL pro . We found, Artemisinin, a natural compound derived from Chinese herb Artemisia annua produces the highest docking score (− 6.40 kcal/mol) as compared to O6K, chloroquine (-4.95 kcal/mol) and hydroxychloroquine (− 5.77 kcal/mol) anti-malarial molecules. Importantly, Artemisinin has demonstrated broad anti-viral activity against human cytomegalovirus, herpes simplex virus type 1, Epstein-Barr virus, hepatitis B virus, hepatitis C virus, and bovine viral diarrhea virus [47]. Artemisinin was shown to exhibit hydrogen bonding with His41, Leu141, Asn142, Gly143, Ser144 and Glu166 SARS-CoV-2 3CL pro amino-acid residues (Fig. 7b).
Amongst the seven antiviral drugs, Ritonavir showed the highest binding energy (-7.45 kcal/mol) and lowest inhibitory constant Ki value (3.49 µM). Ritonavir produced hydrogen bond interactions with Thr26, His41 and Cys145 SARS-CoV-2 amino acids (Fig. 7c). A combination of two HIV-1 protease inhibitors, lopinavir and ritonavir, were given to critically ill SARS-CoV 2 infected patients [48]. However, the combination therapy of lopinavir and ritonavir was also stopped early in 13 patients (total recruitment 99 patients) due to associated gastrointestinal adverse events [48].
The severity of antiviral therapy adverse events has led researchers to explore the potential of macro-, microand phytonutrients that can potentially promote an immune response and suppress viral induced effects. Vitamins are known to modulate the host immune functions by providing anti-oxidants and anti-inflammatory activity [49,50]. Therefore, we selected vitamins, ascorbic acid (vitamin C), cholecalciferol (vitamin D) and alpha-tocopherol (vitamin E) to investigate their potential interactions with the enzyme SARS-CoV-2 3CL pro . Our docking results interestingly, showed that vitamin D has the lowest binding energy and Ki (− 7.75 kcal/mol and 2.08 µM respectively) as compared to vitamin C and vitamin E. Amino acid residues Thr24, Thr26, His41 and Cys145 of SARS-CoV-2 3CL pro showed hydrogen bond formation with vitamin D (Fig. 7d). Amino acid Thr is extensively involved in intracellular signalling changes through phosphorylation changes, and here we observed that cholecalciferol formed a strong hydrogen bond with Thr residues and could potentially block the phosphorylation of Thr residue in SARS-CoV-2 3CL pro enzyme. There is evidence that serious SARS-CoV-2 infected cases have reported severe vitamin D deficiency and thus therapeutic concentrations of this molecule could potentially be used clinically in SARS-CoV-2 cases [51,52].
Discussions
The novel coronavirus termed "nCovid-19" is now known as the third large-scale epidemic coronavirus introduced into the human population in the twenty-first century. At the time of writing, more than 3.67 million confirmed cases globally, with nearly 250,000 deaths had been reported by WHO. Clinically, nCovid-19 is similar to SARS regarding its presentation, however the sheer capacity and speed of which nCovid-19 has spread to global pandemic levels have left researchers asking what makes this outbreak so similar in presentation, yet so different in its virulence to previous coronaviruses. Genome sequence analysis has looked to investigate similarities in the phylogeny of SARS-CoV-2, which like SARS and MERS, have now placed it in the betacoronavirus genus [53]. The known severe and often fatal pathogenicity of betacoronaviruses has been highlighted in these previous epidemics and has reported higher transmission and pathogenicity than the milder and lesser known a-CoVs, which are often compared to the common cold [54].
Our study further compares the similarities between SARS-CoV and SARS-CoV-2 using Clustal Omega alignment to show that of 918 SARS nucleotides, there was a similarity of approximately 95%. Furthermore, we report high amino acid sequence identity in both SARS-CoV and SARS-CoV-2 main protease 3CL pro , which regulates coronavirus replication complexes [55]. Such highly conserved regions in both catalytic sites and the substrate binding regions of the enzymes has also been validated previously in studies by Huang et al. and Muramatsu et al. [44,45]. While this region provides an attractive target for anti-viral drug design, it also can begin to elucidate on viral origins and uncover its ease in transmission. Based on more recent virus genome sequencing results and evolutionary analysis, the origins and transmission of nCovid-19 have uncovered bats as the natural host of the virus origins [42]. As such, studies earlier this year queried the unknown intermediate host between bats and humans, and recent studies have pointed this to pangolins [41,42]. To determine the extent of the evolutionary relationship between bat-CoV, Pangolin-CoV and SARS-CoV-2, we corroborate that based on the nucleotide sequences of the whole-genome sequence, bat-SARS-CoV and SARS-CoV-2 are grouped together and share > 96% similarity, with Pangolin-SARS-CoV as the closest evolutionary ancestor [41,42]. Furthermore, we report that in isolates of human Wuhan SARS-CoV-2 there is an 85.98% similarity in identity to Pangolin-SARS-CoV, which suggests that Pangolin may be associated with the evolution of subsequent outbreaks of COVID-19.
Regarding nCovid-19 and its similarity in transmission to SARS-CoV, recent studies have also demonstrated that transmission occurs via the receptor angiotensinconverting enzyme 2 (ACE2) [42]. This may indicate why SARS-CoV-2 has often led to severe and in many cases fatal respiratory tract infections, like its two SAR-CoV predecessors. Since the SARS-CoV epidemic of 2002 was also known to use the ACE2 receptor to infect humans [56]. Bronchoalveolar lavage fluid taken from nCovid-19 patients have shown that ACE2 is widely distributed in the lower respiratory tracts of humans [42]. Furthermore, the virion S-glycoproteins expressed on the surface of coronaviruses adhere to ACE2 receptors on human cells [57]. This location provides a target for uncovering the mechanistic insights into the severity of the disease and how this region has assisted in the zoonosis of SARS-CoV-2 specifically. Additionally, mutations in the genomic structure of SARS-CoV-2 also might elucidate on the aggressiveness and pathogenicity of the viruses, which may in turn help to explain why some strains are evolutionarily much more virulent and contagious. Angeletti et al. have described mutations in the endosome-associated-protein-like domain of the nsp2 and nsp3 proteins, the former possibly accounting for the high virulence and contagion, while the latter suggesting a mechanism that differentiates nCovid-19 from SARS-CoV [58]. Our studies build on this knowledge and assist to begin to identify the sub-clinical causes for the virulence and unique pandemic pattern of this outbreak by identifying the evolving mutations from region to region. Additionally, previous studies by Pachetti et al. have reported novel recurrent mutations of the SARS-Cov-2, and our study corroborates these mutations in South America and Africa regions [43]. Drug discovery and vaccine development against SARS-CoV-2 infection require time and lengthy processes, however drug repurposing represents an alternative strategy in the current scenario. Some of these antivirals are currently being used clinically in SARS-CoV-2 treatment, including lopinavir [59], ritonavir [60], remdesivir [61], and oseltamivir [62]. However, in the clinical setting, lopinavir/ritonavir, a 3CL pro and RdRp inhibitors, showed no benefit in Covid-19 adult patients [48]. The double point mutation in RdRp gene identified in our study can potentially lead to a drug-resistance event. Moreover, other classes of drugs, such as chloroquine and hydroxychloroquine have shown antiviral properties by blocking viral entry into cells by inhibiting glycosylation of host receptors [63]. We observed no differences in the SARS-CoV-2 main proteinase, 3CL pro genome sequences, but important differences in SARS-CoV-2 3CL pro with SARs-CoV protein, underlining the extreme need for identification of inhibitors to target the viral life cycle. It is not known whether these mutations induce any alterations in the gene transcription or localisation of affected proteins which can be investigated in near future using biochemical and immunological approaches [64,65].
Conclusions
Various theories have been proposed regarding the origin of highly virulent SARS-CoV-2 particle. Our analysis shows that Bat-SARS-CoV shares > 90% similarity with the SARS-CoV-2, however it is possible that the bat coronavirus infected another "intermediate host", such as Pangolin, which subsequently transmitted the virus to humans. Pangolin isolates do share sequence identity with SARS-CoV-2 genomes and could be an intermediate host. We identified novel mutation hotspot regions from South American and African isolates of SARS-CoV-2 genome sequences. Interestingly, double point mutations in RdRp at position 14,805 and 14,808 and triple point mutations in nucleocapsid protein at position 28,881, 28,882 and 28,883 were identified in both South American and African genomic sequences, suggesting the vulnerability of these genetic loci to undergo change. In addition, a novel mutation pattern specifically oriented towards nucleocapsid phosphoprotein in both South American and African sequences was noted while novel ORF3a and RdRp specific variants were observed particularly from African genomic sequences. The potential effects of double and triple point mutations on translated proteins and the virulence of SARS-CoV-2 requires further investigations. SARS-CoV-2 main proteinase, 3CL pro genome was observed to be conserved across all collected genomic sequences. Despite significant similarities in the SARS-CoV 3CL pro structure with SARS-CoV protein, SARS-CoV-2 3CL pro revealed certain key differences, which highlight the extreme need for identification of novel mechanism-based drugs to target the virus processing. Repurposed drugs including natural flavonoids and bioflavonoids, antimalarial, antiviral and vitamins-based compounds have previously been shown to be beneficial in several viral infections and outbreaks. The novel data generated from this study enhances our knowledge of the fine molecular differences that differentiate SARS-CoV-2 virus SARS-CoV. It also highlights the emerging variations in the viral genome across different populations as the virus evolves to local genetic and environmental factors. These findings will likely play a key role in the development of mechanism-based and targeted therapeutic strategies to treat SARS-CoV-2 infection and reduce its virulence.
Additional file 1: Table S1. Acknowledgement table containing information about authors, originating, and submitting laboratories of the sequences deposited to GISAID database. Additional file 2: Table S2. SARS-CoV and SARS-CoV-2 sequence alignment of 3CLPro shares around 95% similarity.\===
Domain: Environmental Science Biology Chemistry Medicine. The above document has 2 sentences that start with 'SARS-CoV and SARS-CoV-2',
2 sentences that end with '(PDB ID: 6Y2G)',
2 sentences that end with 'SARS-CoV-2 3CL pro protein',
2 sentences that end with 'Pachetti et al',
5 sentences that end with 'SARS-CoV-2 3CL pro ',
2 sentences that end with 'in SARS-CoV-2 3CL pro (Fig',
2 paragraphs that start with 'Multiple sequence alignment',
2 paragraphs that start with 'Crystal structure of'. It has approximately 4563 words, 197 sentences, and 39 paragraph(s).
|
[{"added": "2020-07-09T15:02:11.648Z", "created": "2020-05-15T00:00:00.000", "id": "220417938", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://translational-medicine.biomedcentral.com/track/pdf/10.1186/s12967-020-02448-z", "pdf_hash": "fa9ba00962e31e9dc76faaa5117a3cf5b1364e93", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:31039", "s2fieldsofstudy": ["Medicine", "Chemistry", "Environmental Science"], "sha1": "fa9ba00962e31e9dc76faaa5117a3cf5b1364e93", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Identifying locations susceptible to micro-anatomical reentry using a spatial network representation of atrial fibre maps
Micro-anatomical reentry has been identified as a potential driver of atrial fibrillation (AF). In this paper, we introduce a novel computational method which aims to identify which atrial regions are most susceptible to micro-reentry. The approach, which considers the structural basis for micro-reentry only, is based on the premise that the accumulation of electrically insulating interstitial fibrosis can be modelled by simulating percolation-like phenomena on spatial networks. Our results suggest that at high coupling, where micro-reentry is rare, the micro-reentrant substrate is highly clustered in areas where the atrial walls are thin and have convex wall morphology, likely facilitating localised treatment via ablation. However, as transverse connections between fibres are removed, mimicking the accumulation of interstitial fibrosis, the substrate becomes less spatially clustered, and the bias to forming in thin, convex regions of the atria is reduced, possibly restricting the efficacy of localised ablation. Comparing our algorithm on image-based models with and without atrial fibre structure, we find that strong longitudinal fibre coupling can suppress the micro-reentrant substrate, whereas regions with disordered fibre orientations have an enhanced risk of micro-reentry. With further development, these methods may be useful for modelling the temporal development of the fibrotic substrate on an individualised basis.
Introduction
Atrial fibrillation (AF) is the most common cardiac arrhythmia with significant impacts on both morbidity and mortality [1]. Despite extensive research, there are significant disagreements within the cardiac electrophysiology community as to the mechanisms underlying AF [2,3].
Most studies into the mechanistic origin of AF focus on the maintenance of AF, typically arguing for either organised (mother waves or stable rotors) or disorganised mechanisms a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 (multiple reentrant wavelets). However, if both organised and disorganised mechanisms of AF coexist on a continuous spectrum of electromechanical organisation, as suggested recently [4], it is unlikely that any single treatment strategy will be successful across the full spectrum of AF mechanisms. It is for this reason that personalised, patient-specific approaches to AF treatment have become a key research focus in recent years [5,6].
Here we computationally investigate one mechanism of AF initiation and maintenance: the formation of micro-anatomical reentrant circuits, which are continuously activated electrical circuits anchored to the fibre structure of the atria [7]. This mechanism is controversial, with supporting clinical evidence still required, but it has some "potential to unify . . . previous discrepant observations" in the AF literature [8]. Importantly, the size of experimentally identified micro-reentrant circuits is often at, or below, the spatial resolution which can be resolved with conventional multi-electrode mapping [7,9]. Similarly, the interstitial fibrosis which insulates these circuits is often not easy to detect using conventional LGE-MRI [4], and extracting precise fibrosis densities is challenging given the variability in signal thresholding choices between scans and patients [10]. Hence, computational approaches allow for a degree of hypothesis testing which avoids some of these challenges.
In this proof of concept, we introduce a novel method to assess the feasibility of patient-specific predictions for the distribution of micro-anatomical reentry across the atria. In particular, the method is inspired by the idea that it may be possible to predict the emergence of microreentry before the atria have accumulated sufficient interstitial fibrosis, in contrast to most other approaches which take a static view of the AF substrate [5,11].
Starting from image-based models of the atria, we combine data regarding the atrial geometry and the underlying myocardial fibre structure to form a spatial network. By progressively removing connections in the atrial structure, we assess where in the network a path exists which is sufficiently long to harbour micro-reentry, an approach related to the study of percolation in network science [12]. In the current work, our primary goal is to understand the utility of the method and demonstrate its future potential. With this in mind, our research focuses on how the regions predicted by our method of being susceptible to micro-reentry depend on fibre structure and atrial geometry.
Aims
Experimental evidence from explanted human hearts has suggested that "AF may be driven by microanatomic reentrant AF drivers anchored to fibrotically insulated tracks within the complex atrial wall" [9]. To ensure that a given fibre tract can sustain micro-reentry, the reentrant pathway through a fibrotically insulated fibre tract must be at least one refractory wavelength long.
Our approach is illustrated schematically in Fig 1. Denoting the minimum refractory wavelength during fibrillation as τ, our aim is to assess where in the atrial structure fibrotically insulated reentrant pathways exist of length ℓ, such that ℓ � τ. In a healthy heart with minimal interstitial fibrosis, atrial myofibres are well coupled in both the longitudinal and transverse directions. Hence, the longest possible reentrant path is significantly shorter than one refractory wavelength, ℓ � τ. As the density of electrically insulating interstitial fibrosis increases, the longest reentrant pathway increases until a threshold density is reached at which ℓ � τ and the fibrotically insulated fibre tract can sustain continuous reentry during AF. Importantly however, this threshold density may vary across different regions of the atria; some regions may be susceptible to micro-reentry at a low fibrosis density, whereas other regions may require a significantly higher fibrosis density before micro-reentry can be induced. If the density of fibrosis is too high in a region, micro-reentry may be prevented due to the absence of any closed loops (compact fibrosis).
Our hypothesis is that the process of electrically insulating fibre tracts with interstitial fibrosis can be modelled using spatial networks. A spatial network is a graph consisting of nodes representing entities in space, and edges representing spatial connections between those entities [13]. In our case, a node represents one or more atrial cardiomyocytes in a specific atrial region, and an edge between two nodes indicates whether those groups of myocytes are electrically coupled, that is whether an activation wavefront can pass directly from the group of myocytes represented by one node, to the group of myocytes at the other node.
Starting from a network in which the local density of edges is high, our assertion is that the accumulation of insulating interstitial fibrosis in the atria is structurally equivalent to the progressive removal of edges in our spatial network. If the density of fibrosis is low, then the density of edges in the spatial network is high, and similarly, if the density of fibrosis is high, the In the healthy atrial myocardium, cardiomyocytes (pink cells) are well coupled along and across the principle fibre directions, such that the longest reentrant loop is too short to sustain micro-anatomical reentry (green arrows). (B) Over time, the atria accumulate interstitial fibrosis (yellow cells) forming short segments of electrically isolated fibre. (C) The length of these segments grows as more interstitial fibrosis accumulates until the isolated segments are sufficiently long to harbour micro-reentry (blue arrows). We hypothesise that the accumulation of interstitial fibrosis can be modelled as a spatial network. Right column: Nodes (yellow points) representing a group of cardiomyocytes are connected to their neighbours along fibres (red links) and across fibres (blue links). If the density of interstitial fibrosis is low, there are many transverse (blue) links. We model the increase in the density of interstitial fibrosis as the progressive removal of transverse links, and equate density of edges in the spatial network is low. Then, by identifying loops in the spatial network of length ℓ � τ, we can identify which atrial regions may act as a substrate for micro-anatomical reentry at a given fibrosis density.
Materials and methods
In this paper, we generate spatial networks from three atrial datasets (see below), each consisting of a voxel mesh with local fibre orientation vectors. For each dataset, we generate spatial networks according to two different approaches: one which preserves atrial fibre orientations in the spatial network (the fibre model), and a null model approach which ignores the fibre orientation data but retains other structural information such as wall thickness (the fibre-less null model). These two approaches are shown schematically in Fig 2. By utilising both approaches, we are able to identify how the underlying atrial fibre structure affects the observed microreentrant substrate.
Atrial datasets
The spatial network algorithm is tested on three atrial datasets, one human dataset with MRI derived geometry, but synthetic fibre orientations, and two sheep atrial datasets where the fibre structure has been inferred from high resolution serial surface imaging. Each dataset provides a 3d grid of vectors indicating the local fibre orientation at each point in the atria. The sheep datasets are high resolution and inferred directly from the anatomy of each dataset. Summary statistics for the three image-based atrial fibre maps are given in Table 1. 3.1.1 Synthetic atrial fibre map. The human dataset is detailed in [15] based on techniques developed in [14]. The atrial geometry used provides no information as to the atrial fibre structure. Therefore, a synthetic fibre map is constructed using a semi-automated rulesbased approach which is described in the supplementary material (SM) section 1.1.1.
3.1.2 Anatomically derived fibre maps. We study two anatomically derived sheep atrial datasets: one healthy sheep acquired in [16] and a sheep with pacing-induced heart failure [17]. These datasets are individual-specific, retaining significant local heterogeneity in the fibre maps, and are provided at high resolution, in contrast to the synthetic human fibre map.
Information regarding the data acquisition process is provided in SM section 1.1.2. In this proof of concept, our aim is not to assess the differences in micro-anatomical reentry between humans and sheep, or between healthy and heart failure (HF) sheep, but to understand how different sources of fibre orientation data affect our results.
Constructing spatial networks with fibre structure
The schematic shown in Fig 3 outlines the process by which each image-based model is converted into a spatial network in which the underlying fibre structure is retained.
3.2.1 Fibre tractography. Fibre tracts are generated across the atrial geometry by applying a modified version of the Evenly Spaced Streamlines (ESS) algorithm, see [18]: (1) Local fibre orientation data for a given atrial dataset is coarse grained to a specific voxel resolution, . The ESS algorithm was modified to maximise fibre length by generating longer fibres first. Importantly, unlike other common tractography methods, this modified ESS approach ensures that the density of fibres is approximately uniform across the atrial structure. Technical details for the tractography process are provided in SM section 1.2.
3.2.2 Generation of spatial network structure. The fibres generated from tractography are not connected to each other, Fig 3(C). To connect fibres in a single spatial network, nodes are placed at even intervals (steps of 1 voxel length) along fibres starting at a random seed point. These nodes connect to neighbouring nodes in the same fibre with probability p = 1. Any two nodes which do not lie on the same fibre are connected with probability pðx; r; cÞ ¼ 1 where x is the distance between the nodes, r = 7 is an arbitrary steepness parameter, and c is a characteristic distance. Connecting nodes according to a distance dependent attachment function is a standard technique when studying percolation on spatial networks [12]. For better performance, only nodes separated by x < 2 are considered for connection. By varying the spatial coupling parameter c, the number of transverse connections between fibres can be controlled from a state where fibres are strongly connected at c � 1, to a state where fibres become increasingly disconnected as c ! 0. We can think of the coupling c as representing the grade of fibrosis. If the coupling is large, there is no, or only low-grade, interstitial fibrosis. As coupling is reduced, this is equivalent to a greater burden of interstitial fibrosis. Ideally, our distribution of spatial coupling should be modulated using individual-specific fibrosis data, however, this information is not available with the current datasets. For each dataset, spatial networks are generated with five values of the characteristic distance, c. The specific values used and corresponding risk parameters (see section 3.5.1), are given in S2 Table in SM section 1.3.1 in S1 File. For convenience, specific parameter choices are labelled as having a low, medium or high global risk of micro-reentry.
Constructing spatial networks without fibre structure
For comparison purposes, we consider a null model where the spatial network excludes fibre orientation. All other geometric information is retained. The fibre-less spatial network is generated analogously to the fibre model, with the omission of the initial fibre tractography steps, illustrated schematically in Fig 4. Nodes are randomly distributed across each atrial mesh at a density of approximately 1 node per voxel and are added to the spatial network if the distance between the new node and any existing nodes is greater than or equal to d sep = 0.7. To construct the spatial network, nodes are connected using the same distance dependent probability function as the fibre model given in Eq (1). Unlike the fibre model, none of the nodes are already connected by edges which represent the underlying fibre structure of the atria (no pink edges); the probability of connecting two nodes is independent of their orientation.
For each atrial dataset, we generate spatial networks for three values of the characteristic coupling, corresponding to high, medium and low risk cases, see S3 Table in SM section 1.3.2 in S1 File. The characteristic coupling values for the fibre-less null model are significantly larger than for the fibre model since there are no permanent longitudinal edges.
Identifying the substrate for micro-anatomical reentry
For both the fibre and fibre-less models, we associate the substrate for micro-anatomical reentry with regions in the corresponding spatial network where closed loops exist of length ℓ � τ.
Here, τ represents a typical refractory wavelength. These are found by applying a discrete diffusion model (DDM) to the network, see SM section 1.4. (1), where c is a characteristic distance (dashed circle). There is no directional bias in the connection probability (blue edges only). Nodes which are closer together than the characteristic distance (green arrows) are more likely to be connected than those which are far apart (red arrows). (C) The micro-reentrant substrate is identified using the DDM. [URL]004 The DDM is related to several extremely simple physics models of micro-anatomical reentry [19][20][21][22][23]. These models are not electrophysiologically realistic models of AF. The strength of discrete models lies in the ease with which structural discontinuities can be modelled, in contrast to conventional models which are significantly complicated when no-flux boundary conditions are imposed at the local level [24]. This justifies the use of the DDM in the current context where we focus exclusively on the structural basis for micro-reentry. However, this approach is not suitable for studying the dynamics of micro-anatomical reentry. For a wider discussion of the role of model choice see [25].
Defining risk of micro-anatomical reentry.
To derive a measure for the risk of micro-reentry in each spatial network, we use the DDM and assume that all nodes in the network are equally susceptible to conduction block. If each structure has a similar number of nodes which, if blocked, may initiate a micro-anatomical reentry, then the probability of initiating any given circuit is approximately constant. Hence, for a spatial network with N potential substrates, the rate, λ, at which new micro-reentries are initiated in the DDM is directly related to the number of substrates N. This argument follows directly from similar arguments in [21]. A mathematically precise formulation of this argument is provided in SM section 1.5. The rate parameter, λ, is used as our measure of the global risk of micro-anatomical reentry across the atria as a whole. A uniform probability of conduction block is likely an unrealistic assumption. However, realistic conduction block behaviour is difficult to obtain in the DDM framework. These issues are discussed in more detail in the limitations section.
To compute the risk of micro-anatomical in a given atrial region, we define the local risk for a specific voxel, v, by calculating the number of circuits which form in that particular voxel, N v , when sampling a fixed number of identified circuits, N. The rescaled rate of reentry for the voxel v is then given byR where the denominator provides a normalisation. If a specific atrial region is referred to as "at risk", this signifies a non-zero value of the local risk,R v , in that region. Note that summing the local risk over all the voxels in the atrial geometry gives the global risk, l ¼ P vRv . Given that there are typically many more voxels in each atrial geometry than sampled circuits, we smooth the local risk using a 3d Gaussian convolution with a standard deviation of 5 voxels (* 1.5mm) when plotting onto the atrial surfaces.
Spatial clustering of reentrant circuits.
For each spatial network, we measure how spatially clustered the identified micro-reentrant substrate is. To do so, we sample 1000 voxels from the set of voxels in which micro-anatomical reentry is identified for each network and extract the coordinate of each sampled voxel. The number of independent spatial clusters is then derived by using the DBSCAN algorithm, a standard algorithm for spatial data clustering, see [26]. We set eps = 10 (� 3mm), which is the parameter controlling the distance between points for these to be considered in the same cluster. The number of points required for a cluster is set to 1. If only one cluster is identified by DBSCAN, this indicates that all 1000 sampled circuits in a given network fall within a single, small region of the atrial structure. Conversely, if 1000 clusters are identified, no two circuits are identified in the same location.
Atrial wall thickness & occupied voxel fraction.
Wall thickness is a common measurement used when analysing atrial geometry. One issue when calculating wall thickness is that it is not easily defined in the atrial bulk. On the surfaces, wall thickness can be defined as the distance through the atrial wall along the perpendicular surface vector (there are competing, but similar definitions). However, in the bulk many perpendicular surface vectors pass through the same voxel so that the thickness is not uniquely defined.
To resolve this issue, we introduce a novel measure analogous to wall thickness, the occupied voxel fraction (OVF), which quantifies the proximity of voxels to the atrial walls and captures differences in the local wall curvature and thickness gradients, see SM section 1.6. The measure is defined as the number of voxels within a radiusr ¼ 5 (approximately 1.5mm) which are inside the atrial structure, normalised by the total number of voxels within a sphere of radiusr. The OVF increases (decreases) if the atrial walls become thicker (thinner), and distinguishes between regions with convex, flat, or concave wall morphologies.
Longitudinal connection fraction.
For the spatial networks with fibre structure, we would like to associate the identified micro-reentrant substrate with the degree of local fibre alignment in a particular atrial region. Fibre alignment is not naturally defined in a spatial network. Therefore, we use the longitudinal connection fraction (LCF) as a proxy measure. This is defined as the average number of longitudinal connections (along the fibre direction), normalised by the total number of connections, either longitudinal or transverse. If the LCF in a voxel is large, then fibres in that region are well aligned. Conversely, if the LCF is small, the fibre structure is highly disordered in a local area. Note, however, that the LCF can increase both from an increase in the number of longitudinal connections, but also from a reduction in the number of transverse connections. Therefore, larger LCF does not necessarily imply more longitudinal connections. More detail is provided in SM section 1.7 where we validate that the LCF in each spatial network accurately reflects the expected degree of fibre misalignment measured in the underlying data.
It is interesting to note that we can compare the LCF to the degree of microscopic anisotropy, calculated using Eq (1), at the highest level of spatial coupling where micro-anatomical reentry is inducible in our spatial networks. This demonstrates that the onset of micro-reentry in our spatial networks occurs at approximately the same degree of structural anisotropy as found experimentally by Spach et al. [27], see SM section 1.8 for details. Figs 5 & 6 show the substrate identified as susceptible to micro-reentry in the human atria using the fibre and fibre-less models respectively, at low, medium and high global risk (progressively larger reentry rate, λ). Note that low, medium and high global risk refers to the overall risk of micro-reentry across the whole atrial tissue, and does not refer to whether specific regions identified are at low or high risk (which is captured by the local risk,R v ). The spatial distributions of local risk for the sheep atria are shown in SM section 2.1. How the differences in our datasets affect our results is discussed in SM section 2.2.
Spatial distribution of identified micro-reentrant substrate
The reader should be conscious that the local risk distributions shown do not represent the regions at risk for a single atrial tissue, but rather show the collation of the local risk regions identified across 1000 randomly generated spatial networks. If a region is highlighted in the left column, the low global risk case, this may be interpreted as an area which is susceptible to micro-reentry even if the fibrosis density is low (but non-zero). Regions identified in the middle and right columns are regions susceptible to micro-reentry at moderate and high fibrosis densities respectively. This illustrates a key result: different regions of the atria are susceptible to micro-reentry at different fibrosis densities. In particular, regions may have a characteristic fibrosis density range within which they are susceptible to micro-reentry; if the fibrosis density is too high or low these regions will not exhibit micro-reentry. At high global risk, the susceptible substrate is significantly more diffuse than at lower global risk. However, there remain multiple regions with very low or zero local risk, in particular the posterior right atrium (PRA), left atrial appendage (LAA) and inferior vena cava (IVC).
The dominant local risk substrate lies along the strip between two small holes in the superior right atrium. Secondary regions of high local risk are maintained in the PVs, with local risk migrating slightly further from the LSPV junction into the PLA. The opening of the the coronary sinus forms a new substrate which is not observed at higher coupling. The local risk observed previously at the IAS and PMs is largely absent. 4.1.2 Micro-reentrant substrate in spatial network without fibre structure. The local risk substrate identified by the fibre-less null model, see Fig 6, is qualitatively similar to the local risk regions identified in the fibre model. Key local risk regions including the superior right atrium and the PV sleeves are common to both models. Likewise, the PRA, LAA and IVC are not susceptible to micro-reentry in either model.
At low global risk, the dominant local risk substrate in the fibre-less null model is confined to two spatially isolated PM ridges, one of which exhibits local risk in the fibre model. Additional diffuse local risk is observed in the superior right atrium and in the right PV sleeves. At medium global risk, the low local risk substrate is retained, with additional local risk in all four PVs and diffuse risk across the superior right atrium. These local risk regions are consolidated at high global risk, with a small reduction in local risk observed along the spatially isolated PV ridges. The substrates observed for the fibre model along the coronary sinus (CS) opening and at the IAS are not observed in the fibre-less null model.
Clustering of micro-reentrant substrate
The observed micro-reentrant risk substrates for the fibre and fibre-less models suggest that micro-reentry is spatially confined for both models at low global risk, but that the substrate covers a wider area as the global risk of micro-reentry increases. Fig 7 shows that the substrate is more clustered for the fibre model than for the fibre-less null model. One possible interpretation of this result is that if all micro-reentrant circuits are confined to a small number of clusters, these will be easy to destroy or isolate via catheter ablation. However, if circuits are widely distributed, they will be more difficult, if not impossible, to destroy or isolate. significantly below the average value for each dataset. As global risk increases (increasing λ), the mean OVF increases. This indicates that the bias to thin convex regions of the atria is reduced as edges are progressively removed in each spatial network. Framed in terms of fibrosis densities, this result illustrates that regions with low OVF are susceptible to micro-reentry at a lower fibrosis density than regions with larger OVF. These results are supported by statistical analysis in SM section 2.3. In the superior right atrium, the micro-reentrant substrate at high global risk is concentrated along a strip between two small structural holes (region 1) for the fibre model, Fig 9(A). This region exhibits moderate risk in the fibre-less null model, Fig 9(B), with additional risk along the spatially isolated PM (region 2), the junction of the CT and a second PM ridge (region 3), and around the opening of the TV (region 4). Region 2 exhibits low local risk in the fibre model, despite being a low OVF region, but is the dominant risk region for the fibre-less null model when the rate of micro-reentry is low or medium, see Fig 6(E) left and centre columns. This is because the strong longitudinal coupling in region 2 suppresses micro-reentry in the fibre model. This effect is absent in the null model where fibre structure is omitted. Region 2 is most susceptible to micro-reentry in the fibre-less null model at a low and moderate fibrosis densities, with the local risk reducing at higher densities. Regions 3 and 4 show very low or zero local risk in the fibre model.
Comparing fibre and fibre-less micro-reentrant substrates
Comparing the identified risk regions to the OVF values shown in Fig 9(C), we note that regions 1-4 all exhibit low OVF values (blue on the colour scale). Of the four regions identified, we observe that the LCF values, Fig 9(D), in region 1 are low (blue), indicating fibre misalignment, whereas strong longitudinal coupling is observed in regions 2-4. This suggests that the local risk of micro-reentry is suppressed in the fibre model in regions with high LCF, but can be enhanced in regions with low LCF. Recall that the OVF is a geometric property of the atrial structure which is unaffected by the choice of coupling parameter, c, used to initialise the spatial networks. Hence, the distributions in Fig 10(A) & 10(D) align along the OVF axis. However, the LCF is not independent of the coupling parameter c used to initialise each network, and, consequently, is not independent of the global risk, λ. Therefore, the distributions in Fig 10(A) & 10(D) do not align along the LCF axis since an increase in global risk is associated with a decrease in transverse coupling, c. This is equivalent to an increase in the LCF. Fig 10(B), the voxel distribution is significantly skewed for both the OVF and the LCF with the weighted mean falling in the first OVF decile and the first LCF quartile. For the fibre-less null model at low global risk, Fig 10(C), the mean OVF value also falls in the first decile, but there is a small skew to LCF values larger than the median. This is an indirect effect and is explained by the strong longitudinal coupling along spatially isolated regions such as the PMs.
For both the fibre and fibre-less models the local risk density lies almost exclusively in the first OVF quartile, with negligible density at higher OVF values. In contrast, LCF values are clearly skewed in the fibre model, although some local risk density remains at larger LCF values. This implies that at low global risk, the OVF is the dominant factor in determining the local risk substrate, with fibre structure playing an important secondary role.
As the global risk of micro-reentry increases, we observe that the bias to low OVF values remains, but is slightly reduced, see LCF) has a protective effect, reducing the local risk of micro-reentry, relative to regions with lower LCF. However, if we compare the case of low global risk in Fig 10(A)-10(C), to the case of high global risk, Fig 10(D)-10(F), it appears the the increase in global risk is associated with a shift in the voxel distributions from low LCF to higher LCF. This would would seem to imply, incorrectly, that longitudinal connections enhance the local risk of micro-reentry.
To understand this, note that there are two ways in which the LCF may increase: Firstly, if we fix the number longitudinal connections in the atrial geometry and progressively remove transverse connections, this will result in the average LCF increasing since the relative abundance of longitudinal connections increases. Conversely, if we fix the number of transverse connections and increase the number of longitudinal connections, this will result in an increase in the LCF. The former is the mechanism relevant when comparing the low global risk case to the high global risk case, and is not associated with an increase in the number of longitudinal connections. However, the latter is the mechanism relevant in a given atrial geometry where transverse coupling is effectively fixed, but regions with higher LCF have more longitudinal connections than regions with lower LCF. It is this second mechanism that is associated with the protective effect of longitudinal connections.
Discussion
In this paper, we introduce a novel computational approach for analysing how the structural substrate for micro-anatomical reentry may develop as the atria accumulate diffuse interstitial fibrosis. The technique combines spatial networks with fibre tractography, and assumes that interstitial fibrosis accumulation can be modelled as the decoupling of a spatial network, a concept closely related to spatial percolation [12]. Although our work focuses on micro-anatomical reentry, we believe our approach is of general interest, offering new methods for modelling atrial fibre structures, as well as metrics for quantifying longitudinal coupling and structural curvature.
Many studies investigate the electro-anatomical basis for micro-reentry as a potential mechanism of AF, although robust clinical evidence to support the mechanism is still lacking. Broadly speaking, such studies fall into three categories. In the first, the substrate can be identified directly, given sufficiently high resolution imaging, by observing the presence of a microanatomical driver. Examples include [7,9], identifying several key factors in the formation of the reentrant substrate. Of particular note are the orientation of muscle fibres, the structure, thickness, and thickness gradients of the atria, and the accumulation of fibrosis, particularly interstitial fibrosis [28].
Attempting to quantify these factors, the authors in [9] applied optical mapping to explanted human atria to identify how specific values of wall thickness and fibrosis burden correlate to the location of the known micro-reentrant substrate [9]. Driver regions were found to correlate well with areas of 20-30% wall thickness and 20-30% fibrotic burden, notably the junction of the RIPV and PLA, and between the CT and fibrotically insulated PMs. Fibre misalignment was also implicated, specifically between the CT and the PMs which also exhibit abrupt changes in the local wall thickness. However, these studies took a largely static view of the micro-reentrant substrate, not considering how the substrate may change under electrical or structural remodelling. Note that these studies have also been criticised for inducing drivers under physiologically unrealistic doses of pinacidil for shortening the action potential duration.
In the second category, the substrate for micro-reentry has been established in the atria, but is hidden due to the lack of a clearly observable driver. One recent study has addressed this problem by noting that the visible micro-reentrant substrate varies strongly with variable atrial refractoriness [29]. Hence, the authors were able to demonstrate that the hidden micro-reentrant substrate could be unmasked and stabilised by shortening atrial refractoriness with adenosine. Such changes, induced pharmacologically, are analogous to some of the electromechanical changes that may be expected over time from atrial remodelling [4]. However, a robust framework for predicting the future micro-reentrant substrate is still lacking.
No study is yet to address the third category which is to predict how the substrate for micro-reentry will develop in the future on a patient-specific basis, a problem directly related to the increasing need for arrhythmic risk stratification [30].
In this proof of concept-which we acknowledge is far from clinical applicability-we have attempted to take a step towards addressing this and ask how the substrate for micro-reentry may develop over time. Specifically, we ask how different parts of the atria are susceptible to micro-reentry at different characteristic fibrosis ranges; some regions of the atria, particularly those where the atrial walls are thin and there is significant fibre misalignment, may be susceptible to micro-reentry at a low fibrotic density, whereas other regions may require higher fibrosis densities before micro-reentry can be induced. Such an approach is currently only possible in-silico, primarily due to ongoing challenges with accurate in-vivo atrial imaging [31].
Our results implicate the role of wall thickness and the misalignment of fibres, suggesting that thin atrial walls and reduced longitudinal fibre coupling both enhance the probability that a region is susceptible to micro-reentry at lower fibrosis densities, supporting the findings in [9]. However, our study suggests that the dependence on these factors evolves as the density of interstitial fibrosis grows in the atria. In particular, the spatial spread of the micro-reentrant substrate increases dramatically with small reductions in the spatial network coupling, and indicates that the bias to thin atrial regions with complex fibre morphology is reduced as the micro-reentrant substrate becomes more spatially diverse. Many of the specific regions highlighted as local risk substrates for micro-reentry in [9,29], such as PM ridges and the superior right atrium, naturally emerge as local risk substrates using our method, most likely due to their position in thin atrial regions.
One explanation for the importance of wall thickness, fibrosis density, and local wall curvature is that driver regions emerge from percolation-like dynamics where the critical fibrosis density is dependent on the thickness of the structure studied [32,33]. This in turn may result in micro-anatomical circuits anchoring to the atrial surfaces in paroxysmal AF but distributing across the atrial wall in persistent AF [34]. Aside from micro-reentry, modelling fibrosis using percolation-style distributions is known to perform particularly well when simulating patterns of AF maintenance [35], and has been used to explain complex fractionated electrograms and reentries in fibrotic border zones [36,37].
In the wider AF literature, a number of computational studies consider the role of atrial structure on AF dynamics. In most cases, these studies investigate how structural factors affect electrical wavefront dynamics by solving the mono-or bidomain equation coupled with a suitable atrial cell model [33,[38][39][40][41][42][43]. However, even the most advanced patient-specific modelling methodologies suffer from the continued struggle to extract precise fibre orientation data and high resolution fibrosis profiles (particularly diffuse interstitial fibrosis) in-vivo [6].
Until these issues are resolved, computational modelling must continue to be used as a platform for hypothesis testing, studying the dynamics of AF initiation and maintenance with a variety of methods and across scales. In particular, the clearest insights may be found from models which contrast the role of a specific electro-anatomical feature with null models in the absence of that feature. Examples include models with and without realistic tissue geometries [38], with and without patient-specific fibrosis [44], isotropic vs. anisotropic fibre structures [41], and continuous vs. discrete modelling choices [25]. We hope that the techniques introduced in this paper can add to the range of techniques employed for hypothesis testing in cardiac electrophysiology.
Micro-anatomical reentry in the wider context
The role of local drivers, and specifically micro-anatomical reentrant circuits, in initiating and maintaining AF is controversial. After initial findings showed promise in 2015 [7], some suggested that the mechanism may have "the potential to unify some of the previous discrepant observations" [8] on AF mechanisms. However, although research is ongoing, reproducible clinical evidence identifying micro-anatomical reentrant circuits in patients is still lacking. Despite this, it is important to reiterate that the size of micro-reentrant circuits are postulated to be at, or below, the spatial resolution which can be resolved with conventional multi-electrode mapping [7,9]. This may hamper attempts to acquire clinical evidence in support of the mechanism.
Opposing the local drivers hypothesis, many believe that AF is maintained by spatio-temporal chaos in the atria, whereby wavefronts continually collide into each other producing new fibrillatory wavefronts [45][46][47][48][49][50][51]; evidence for the mechanism is extensive and extends beyond AF-specific studies to more general studies on spatio-temporal chaos in excitable systems [52][53][54]. In our view, there is no reason that local drivers cannot co-exist with other non-local mechanisms, especially given recent evidence for a spectrum of mechanisms at different levels of organisation during fibrillation [4]. This is important because the interaction between local drivers and spatio-temporal chaos may have a significant effect on whether the ablation of local drivers is a feasible strategy for terminating AF.
Spatio-temporal chaos may have a non-trivial impact on local driver regions, likely resulting in an increased rate of circuit termination as the fibrotic substrate becomes more spread out, but also resulting in AF becoming more turbulent and difficult to terminate. If local drivers can be ablated, this may be avoided. However, local ablation in this context with numerous drivers across a diffuse substrate may become practically impossible given that (1) circuits are short-lived due to chaotic propagation, (2) circuits are spatially diverse, and (3) even if circuits can be destroyed, damage to the atrial myocardium may promote turbulent electrical propagation, driving more persistent AF. Despite this, local ablation strategies may still be worth pursuing if they lower the probability of initiating AF from sinus rhythm, where micro-reentrant circuits may act as a trigger.
Whether robust evidence for the micro-reentrant driver mechanism can be established is a question for the future. However, even without this evidence, the methods introduced here may be valuable to the wider cardiac electrophysiology community, adding to the computational toolkit for hypothesis testing. In particular, the methods introduced for constructing spatial networks using fibre orientation data may be of general interest to the community. Equally, structural metrics including the occupied voxel fraction and the longitudinal connection fraction may be useful for studying atrial geometries and fibre structures. Finally, although we acknowledge the limitations of the method (see below), the discrete diffusion model used to propagate signals across our spatial networks may be of general use, particularly in cases of high computational complexity where more phenomenologically accurate models constrain the number of simulations that can be run.
Limitations
The aims of this study are highly focused, discussing the structural basis for micro-anatomical reentry only. Many other factors may affect the probability that a micro-reentrant circuit forms in a given region of the atria, including other forms of fibrosis and ionic remodelling.
Our approach does not consider these factors and cannot exclude their importance to microanatomical reentry.
The key technical limitations of our work come under two main categories: (1) Imaging related limitations and (2) limitations arising from specific modelling choices in the construction of the spatial networks.
In our view, one of the most important limitations relates to the datasets used for analysis. In particular, the human fibre dataset has undergone extensive smoothing, lacking realistic local heterogeneity in the fibre structure; the observed heterogeneous regions are likely artefacts from the synthetic fibre generation method. Similarly, the atrial geometry is heavily smoothed, lacking fine structural detail in regions such as the RAA and LAA, but with numerous structural holes across the geometry. Any future work must ensure that predictions are not simply an artefact of the data acquisition and generation process.
Although using high resolution fibre maps like those in [16,17] may avoid these issues, acquiring such data is not feasible in-vivo, remaining a major ongoing research challenge [55,56].
Assuming the imaging data used is of an acceptable quality and accuracy, a number of limitations arise in the fibre map modelling process. Firstly, the fibre tractography methods applied to generate global fibre tracts from local fibre orientation data are only an approximation of reality. Without performing precise histology to determine the position of global fibre tracts in our model, not histology focusing solely on local fibre orientation, it is not possible for us to explicitly validate this approach. Since the imaging data used was acquired from previous studies, such a process is not possible in the current work but may be possible in future studies. As a simple check of the robustness of our results, the full simulation pipeline, including the regeneration of fibres from a different set of random seed points, was carried out three times for the healthy sheep atria, showing no evidence that this significantly affected our final results.
Once fibre tracts are generated, nodes are placed along the fibres, and are coupled to nearby nodes based on their separation. The attachment function to connect two nodes is uniform across the atrial structure and does not consider any possible differences in the local connectivity of cardiomyocytes. Likewise, the characteristic coupling, c, that is used in each model is constant across the atrial structure. This can be thought of as applying a uniform density of interstitial fibrosis across the atria. It is known that not all regions of the atria are equally susceptible to the accumulation of fibrosis [57], and likewise, different forms of fibrosis accumulate differently across the atria [58]. Our assumption of uniformity is a reasonable first approximation in the absence of patient-specific information regarding the distribution of interstitial fibrosis which may not be visible with LGE-MRI. We do not consider any macroscopic fibrosis in the model. Such information was not available with the datasets acquired, but should be considered in any future work. However, how to accurately map imagingderived fibrosis densities to precise spatial network edge densities is so far unclear and will likely pose a future technical challenge.
Once a spatial network has been constructed, the micro-reentrant substrate is identified by applying a simple discrete diffusion model on the network. The approach is loosely based on the techniques discussed in [20,34], although we stress that its purpose here is strictly to identify locations in which isolated fibres result in micro-anatomical reentry, not to simulate AF dynamics. Assessing the ability of these structures to initiate and maintain AF would warrant a more detailed computational analysis with a phenomenologically accurate model which may be challenging at the resolution of our spatial network, and may struggle to produce large scale statistics. Despite this, the use of phenomenologically accurate propagation models may be necessary within our spatial network framework if we are to fully gauge and weigh up the relative impact of structural factors on the probability of micro-reentry, relative to other considerations such as the formation of functional blocks. Future studies using either method would benefit from a detailed comparison to experimentally acquired atrial functional data.
The regions that are identified by the discrete diffusion model are a set of isolated fibre tracts where unidirectional conduction block at a single node is sufficient to induce a microreentrant circuit. The method to apply conduction block assumes uniform risk of block, avoiding the need to specify special electrical properties in key atrial locations, and ignoring the potential impact of functional blocks from source-sink mismatch. In practice, it is known that electrical properties of cardiomyocytes vary significantly across the atria. For instance, the proarrhythmic conditions found in the PVs such as conduction velocity slowing and shortened action potential durations are not considered. However, our results do indicate that the PVs can emerge as a key risk substrate without the inclusion of electrical proarrhythmic effects.
Finally, before our approach can have real clinical relevance, it would benefit from further validation. We have demonstrated that the spatial networks accurately preserve the underlying properties of each atrial fibre map, and that our techniques ensure an even density of nodes across the network. For clinical relevance, it is important to validate the approach directly against experimental data, ideally with datasets with a history of micro-anatomical reentry. If possible, raw data should be available such that fibre maps can be generated using a range of methods at different levels of smoothing, and that the influence of small structural holes can be tested. Data may be acquired at different levels of interstitial fibrotic density to assess the validity of our method's longitudinal predictions.
Conclusion
We have introduced a simple, proof of concept framework which attempts to study how the substrate for micro-anatomical reentry develops from the accumulation of interstitial fibrosis. The method, which is based on the application of percolation to spatial networks, suggests that the micro-reentrant substrate is critically dependent of local tissue geometry and areas of fibre misalignment. We suggest that the dependence on these factors is complex, continuously evolving with the absolute level of global micro-reentrant risk.
If robust clinical evidence can be found to support the micro-reentrant driver mechanism, and if our methods develop sufficiently, the approach introduced here may have potential for patient-specific risk stratification and personalised ablation strategies. However, our results also imply that the micro-reentrant substrate may become so spatially diverse, particularly at high fibrosis densities, that ablating these drivers may not be practically feasible.
Supporting information S1 File. Supplementary information document. This file contains additional details regarding the methods used in this paper, as well as expanded results. In particular, this document includes details of atrial geometry thickness measurements, and figures showing the microanatomical risk substrates for the sheep atrial datasets. (PDF)
==
Domain: Engineering Biology Medicine
|
[{"added": "2021-09-18T13:19:41.867Z", "created": "2021-09-15T00:00:00.000", "id": "237550371", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": null, "oa_url": null, "pdf_hash": "c44c1a4a5b951792ef4852d18a277d78ea759046", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:30831", "s2fieldsofstudy": ["Medicine", "Engineering"], "sha1": "2f70d42f69e727acd708a5a7764b0f7670970b95", "year": 2022}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Typical epidemiology of respiratory virus infections in a Brazilian slum
Abstract Host population size, density, immune status, age structure, and contact rates are critical elements of virus epidemiology. Slum populations stand out from other settings and may present differences in the epidemiology of acute viral infections. We collected nasopharyngeal specimens from 282 children aged ≤5 years with acute respiratory tract infection (ARI) during 2005 to 2006 in one of the largest Brazilian slums. We conducted real‐time reverse transcription‐polymerase chain reaction (RT‐PCR) for 16 respiratory viruses, nested RT‐PCR‐based typing of rhinoviruses (HRVs), and collected clinical symptoms. Viruses were common causes of respiratory disease; with ≥1 virus being detected in 65.2% of patients. We detected 15 different viruses during 1 year with a predominance of HRV (33.0%) and human respiratory syncytial virus (hRSV, 12.1%) infections, and a high rate of viral coinfections (28.3%). We observed seasonality of hRSV, HRV and human coronavirus infections, more severe symptoms in hRSV and influenza virus (FLU) infections and prolonged circulation of seven HRV clusters likely representing distinct serotypes according to genomic sequence distances. Potentially unusual findings included the absence of human metapneumovirus detections and lack of typical FLU seasonal patterns, which may be linked to the population size and density of the slum. Nonetheless, most epidemiological patterns were similar to other studies globally, suggesting surprising similarities of virus‐associated ARI across highly diverse settings and a complex impact of population characteristics on respiratory virus epidemiology.
| INTRODUCTION
Acute respiratory tract infections (ARI) are the main cause of morbidity and mortality among children aged <5 years in the developing world. 1 Respiratory viruses cause up to 80% of ARI. 2 Respiratory viruses are spread via three different transmission routes: contact, droplet, and aerosol transmission. 3 Host population size, density, immune status, age structure, and contact rates affect the transmission patterns of viruses causing acute predominantly self-limiting infections, such as respiratory viruses. 4 Residents of resource-limited communities such as slums may be particularly vulnerable to virus-associated ARI. Hypothetically, virus transmission may be facilitated in these dense populations, characterized by frequent interindividual contact, crowded housing, improper sanitation systems, poor education, and poor nutritional status, exemplified by inversely correlated influenza virus prevalence and family income in a study from Bangladesh. 5 The United Nations define slums as human settlement areas that combine the following attributes: lack of basic services as sanitation and water sources, substandard housing or illegal and inadequate building structures, overcrowding and high density, unhealthy living conditions and hazardous locations, insecure tenure; characterized by irregular or informal settlements, poverty, and social exclusion. 6 Close to 880 million people worldwide reside in urban slums, and this number is expected to double by 2025. 7 Nonetheless, little is known about how and whether disease patterns in urban slums differ from affluent settings. 8 Pivotal epidemiological studies conducted in slum cohorts from Bangladesh and Kenya highlighted the importance of respiratory viruses in these communities. 5,9,10 In Brazil, 11.4 million people, nearly 6% of the country's population live in slums ( [URL]/). Data on virus-associated ARI from slum communities, particularly from Brazil, are scarce. In a study based on clinical symptoms, ARI in children inhabiting a Brazilian slum were very frequent, representing 50% of child infections events through a 1-year period. 11 A study combining clinical and virological data reported ARI symptoms in 60% of children inhabiting another Brazilian slum and in 35% of these children, the virus was isolated by cell culture. 12 Moreover, surprisingly high virus or bacterial detection rates (up to 85%) were observed in a study conducted in northeastern Brazil in children under 5 years from low-income families presenting with ARI. 13 We previously found that virus-associated ARI showed similar epidemiological patterns between a rural African and an urban European setting, including an overall similar spectrum of viruses, age associations and seasonal fluctuation despite drastic differences of socioeconomic status (SES) and climatic patterns. 14 Hypothetically, lower SES in the African setting may have been equaled by higher population density in the European setting, both of which likely facilitate virus transmission.
Here, we investigated 16 respiratory viruses by real-time (reverse transcription)-polymerase chain reaction (RT-PCR)-based methods in one of the biggest slums of Latin America, namely Paraisópolis. This community is located within the urban area of São Paulo city, inhabited by 89.000 residents in a 1.5 km 2 area (approximately 59.300 inhabitants/km 2 ) and distributed in 21 thousand dwellings including shacks, masonry buildings, townhouses and older and solid buildings 15 ( Figure 1A).
| MATERIALS AND METHODS
Nasopharyngeal aspirates were collected longitudinally from human parechoviruses (hPeVs), adenoviruses (AdVs) and influenza A and B viruses (FLU) as described previously. 14,16 HRV typing was done using a nested RT-PCR assay targeting the VP4/VP2 domains. 17 The HRV evolutionary history was inferred using a Neighbor-Joining algorithm, a p-distance substitution model and a complete deletion option in MEGA6 (www.megasoftware.net/) on a data set comprising 434 nucleotides from the viral VP2/VP4 domains after deletion of 5′-untranslated region sequence portions. Recombination in the data set was discarded using RDP V4.95. Only HRV was typed due to frequent detection and rapid evolution allowing sufficient phylogenomic resolution.
Statistical analysis was performed by using SPSS 13.0 (IBM) with Χ 2 or Fisher's test when analyzing numbers less than 5 in any cell.
Confidence intervals were calculated using Open Epi (www.openepi. com) using the Wilson method.
| RESULTS
Many epidemiological features of respiratory viruses in our study were similar to those from other reports worldwide. The first similarity was a high overall virus detection rate of 65.2% (184 of 282 patients) together with the genetic diversity of the respiratory pathogens, including all viruses that were tested except hMPV (Table 1). Both attributes were previously reported in studies conducted in temperate and tropical regions of Brazil (~60.0%), 18,19 in a slum community from Kenya (71.0%), 20 and in a German cohort (56.6%). 14 The second similarity was the predominance of HRV (33.0%) and hRSV (12.1%) detections (Table 1), also observed in hospitalized patients in distinct cities of Brazil, in a slum community in Kenya [18][19][20] and in German children. 14 The third similarity was the high rate of viral coinfections (28.3%) ( Figure 1B) which was similar to rates observed in a study from Curitiba, southern Brazil (29.0%) 19 and in a Kenyan slum community (27.0%). 20 Our data also point to the predominance of HRV in cases of coinfections (30 of 52 coinfections, 58.0%), a feature similarly reported in a study conducted in Brazil (69.0% of all coinfections) 19 that is likely influenced by the overall high number of HRV infections. Albeit not statistically significant, HRV, hRSV, and FLU were commonly detected as monoinfections (67.7%, 73.5%, and 76.9% respectively; P > .05 for all three viruses), whereas HCoV, AdV, hPiV, EV, and hPeV were significantly more frequently detected as coinfections (53.8%, 61.1%, 66.7%, 80.8%, and 83.3%, respectively; P < .05 for all) ( Figure 1C).
These data were consistent with results of a study including data from eight tropical countries showing that HRV, hRSV, and FLU were more commonly detected as monoinfections, AdV as coinfections and HCoV and hPiV equally distributed between both coinfections and monoinfections. 21 Similarly, preliminary studies from Brazil showed that hRSV and FLU were most frequently detected as monoinfections and EV and AdV as coinfections. 13,19 The fourth similarity was the seasonal variation of hRSV and HCoV detections. Particularly, hRSVs were more frequently detected during autumn (Fisher's exact, P = .004), a pattern already observed in surveillance studies carried out in two distinct São Paulo city hospitals. 22,23 HCoV were more frequently detected during winter (Fisher's exact, P = .011; Figure 1D), as observed previously in a study conducted during 20 years in the United States. 24 The fifth similarity included statistically significant associations of FLU and hRSV infections with more severe symptoms such as fever (Fisher's exact, P = .007 and P = .02, respectively) and of hRSV infections with dyspnea (Fisher's exact, P = .038) ( Figure 1E)
| 1319
Together with the low frequency of codetections of FLU and hRSV, this feature is consistent with higher pathogenicity of both viruses. 18 In contrast to hRSV, HRV infection was significantly less frequently associated with fever and dyspnea (Fisher's exact, P = .001, Χ 2 P = .005) ( Figure 1E). The lower proportion of HRV infections and cases of fever compared to other community-acquired respiratory virus infections was similar to a previous study conducted in patients hospitalized with ARI in Curitiba, southern Brazil. 19 The sixth similarity was a predominance of hRSV and hPeV detections in patients aged ≤1 year (Fisher's exact, P = .001 and P = .01, respectively) ( Figure 1F). Similarly, Annan et al 14 reported that pneumoviruses including hRSV were more frequently detected at younger ages in cohorts from Ghana and Germany, predominantly in patients less than 1 year of age.
The seventh similarity comprised two contrasting HRV epidemiological patterns, namely replacement of some HRV strains and maintenance of other HRV strains over time. HRV comprises three defined species termed HRV A-C, and >100 different types likely representing multiple distinct serotypes. 25 We successfully typed a total of 77 HRV strains representing all three HRV species (GenBank accession numbers MH824434-MH824510), whereas 16 HRV strains could not be typed. Proportions of individual HRV species in our study (53.2% HRV-A, 6.5% HRV-B, 40.3% HRV-C) were in agreement with other reports worldwide. 17,26 The majority of distinct HRV strains in our study were detected only in one season, a common HRV epidemiological pattern. 17,[25][26][27] In contrast, seven clusters termed I-VII, each composed of three to five patient-derived HRV strains presenting a very low mutual pairwise sequence distance (≤2%), were detected over more than two seasons. The seven clusters belonged to HRV species A and C and differed from one another by 15% to 39% mutual nucleotide sequence distance. Previous studies on HRV typing using the genomic fragment used in our study found that defined HRV serotypes differed by at least 10% mutual nucleotide sequence distance. 17 Following this criterion, strains belonging to one cluster in our study thus likely represent the same HRV serotype, whereas the seven clusters all represent distinct HRV serotypes. Community protective immunity can dramatically limit the circulation of defined viral serotypes, including HRV and other viruses. 17,28 This was apparently the case for some, but not all HRV serotypes in our study, the latter potentially facilitated by the population structure of the slum.
Prolonged circulation of HRV clusters was detected specifically during winter and spring ( Figure 1G, clusters I, III, V, VII), summer and spring (cluster II), winter and summer (cluster IV), and autumn and summer (cluster VI). These data were reminiscent of prolonged circulation of closely related HRV lineages for 4 to 12 months or over three consecutive seasons reported previously from Sweden and Finland 26,27 and are thus not unique to our study setting. Notably, we cannot exclude that HRV strains may have been re-introduced repeatedly from other areas of São Paulo, a large metropolis accumulating about 21 million inhabitants, potentially facilitated by commuting of slum inhabitants to nearby areas for labor. 15 We observed only two potentially unusual patterns including the total absence of hMPV detections and absence of FLU seasonality.
The first may be explained by widely documented local variation in annual hMPV circulation, 29 More specifically, usage of identical methodology enables direct comparisons between this study and our previous investigation of a rural area in Ghana and an urban environment in Germany. 14 approaches, yet studies focusing on respiratory virus epidemiology in slum populations are scarce compared to studies from affluent settings. 32,33 This knowledge will be crucial to inform potential public health interventions to reduce disease burden in the population.
ACKNOWLEDGMENTS
This study was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; n o 02/08460-6) and by the European Union's Horizon 2020 program through the ZIKAlliance project (grant agreement no. 734548).\===
Domain: Environmental Science Biology Medicine. The above document has 2 sentences that end with 'from affluent settings',
2 sentences that end with 'other reports worldwide',
2 sentences that end with 'mutual nucleotide sequence distance'. It has approximately 1989 words, 77 sentences, and 19 paragraph(s).
|
[{"added": "2019-11-27T14:04:51.718Z", "created": "2019-11-26T00:00:00.000", "id": "208300221", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/jmv.25636", "pdf_hash": "cc09e877cae9906919fbff5abe9aac1acebef307", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:32278", "s2fieldsofstudy": ["Medicine", "Environmental Science"], "sha1": "3b3a564961c68b4fcd7d7de60c3739847f92c2cc", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Endemic and Imported Measles Virus–Associated Outbreaks among Adults, Beijing, China, 2013
In 2013, a resurgence of measles occurred in Beijing, China. The outbreaks occurred among adults and were associated with endemic genotype H1 and imported genotype D8 viruses. Migrant workers were disproportionately represented in the outbreaks; thus, vaccinating such workers against measles may be an effective strategy toward the elimination of this disease.
In 2013, a resurgence of measles occurred in Beijing, China. The outbreaks occurred among adults and were associated with endemic genotype H1 and imported genotype D8 viruses. Migrant workers were disproportionately represented in the outbreaks; thus, vaccinating such workers against measles may be an effective strategy toward the elimination of this disease.
A ll 6 World Health Organization (WHO) regions have set goals to eliminate measles (1,2). In China, a nationwide measles supplementary immunization activity was conducted in 2010, and the incidence of measles in mainland China subsequently reached its lowest reported level in 2012 (6,183 cases, 4.6 cases/million total population). However, in 2013, a nationwide resurgence of measles occurred primarily among young, unvaccinated children (3). In contrast to the nationwide resurgence, the measles resurgence in China's capital, Beijing, was primarily among adults >15 years of age (65.7% of cases) and occurred in large, clothing wholesale markets.
Beijing has >12.96 million permanent residents and an additional ≈7.73 million floating residents (i.e., internal migrants who move into the city, usually for employment) (4). The routine measles vaccination schedule in use in Beijing consists of 3 doses of measles-containing vaccine; the first dose is administered at 8 months, the second at 18 months, and the third at 6 years of age. Also, since 2006, an additional dose has been administered to college students who move to Beijing to study (5). In this study, we used genotype analysis to describe the measles outbreaks among adults in Beijing, and we suggest an immunization strategy to help prevent similar outbreaks in the future.
The Study
In early 2013, a resurgence of measles in Beijing was reported to the China Information System for Disease Control and Prevention, Chinese Center for Disease Control and Prevention (Beijing); by the end of the year, a total of 1,233 suspected cases had been reported and investigated. Serum and throat swab samples were collected from 97.3% and 96.8% of the suspected case-patients, respectively. The samples were tested for measles IgM by using the VIRION SERION ELISA Measles Virus IgM test (Virion/Serion, Wurzburg, Germany) or for measles virus genes by using a real-time reverse transcription PCR kit (Jiangsu Bioperfectus Technologies, Jiangsu, China). Of the 1,233 samples, 558 were positive, and 5 additional cases were confirmed by epidemiologic linkage. Thus, a total of 563 measles cases in Beijing were confirmed in 2013; this number represents a 6-fold increase from the number of cases in 2012.
The number of reported cases was highest during March-May. Most cases occurred among a floating population of adult migrant workers and permanent adult residents and their children (Figure); 67.3% of the cases in adults were in migrant workers. Reports of cases increased shortly after the national holidays associated with the spring festival, during which many persons travel to visit relatives.
Among the persons with confirmed measles, 22.8% (128) were <8 months of age, 11.5% (65) were 8 months-14 years of age, and 65.7% (370) were >15 years of age and defined as adult patients. The median age of the adult patients was 23.5 years (range 15.0-70.0 years). Vaccination history was unknown for 87.6% (324/370) of the adults. Among the 65 patients in the age group targeted for measles vaccination, 36.9% (24) had not been vaccinated because of contraindications.
We used real-time reverse transcription PCR-positive samples to try to isolate and genotype virus from 468 of the 563 positive samples. Genomic sequencing and phylogenetic analyses were based on N450 nucleotide sequences, as previously described (6)(7)(8) (Table). Measles viruses within a transmission chain had identical or nearly identical N450 sequences (9). Phylogenetic analysis showed that 216 genotype H1 viruses were associated with 30 different chains of transmission (online Technical Appendix Figure, (10) showed that genomic sequences of genotype D8 viruses from the outbreak shared 99.8%-100% nucleotide identity with genomic sequences of strains from measles patients in Russia, France, Canada, Thailand, Denmark, Germany, and other locations.
Genotype D8 measles viruses were associated with at least 2 outbreaks in different large, wholesale clothing markets. The outbreaks occurred during March-July 2013 and were almost completely confined to adults; only 1 child was infected, possibly because of high coverage of measles-containing vaccine among Beijing children. We were unable to identify the source of the infections. Phylogenetic analysis suggested that the genotype D8 virus might have been imported to Beijing from Europe, America, or another location (10)(11)(12)(13) and subsequently spread beyond Beijing by virus transmission from infected adults (data not shown). For at least 21 years, genotype H1 measles viruses have been the only endemic measles circulating in China (6-8); measles cases caused by all other genotypes have been associated with imported viruses (7).
Conclusions
Our findings show that transmission of measles virus among adults contributed to a resurgence of measles in Beijing during 2013. Nonendemic genotype D8 measles viruses were associated with at least 2 outbreaks in different wholesale clothing markets during March-July, 2013. Many persons from domestic and international areas visit wholesale markets every day; thus, such markets are highrisk settings for the transmission and importation of measles viruses (14).
Because migrant workers were disproportionately affected in the Beijing outbreaks and because their work settings have high measles transmission potential, we support an outreach strategy to prevent measles among this floating population. These workers usually live with their families and register with the local authorities for government services. Thus, we suggest that the offer of measles vaccine to workers as they register to live and work in the commodity markets might be a reasonable strategy to prevent future measles outbreaks. Serologic surveys can be used to refine such a strategy by assessing immunity within the target population.
The foundation strategy for eliminating measles globally is based on the timely vaccination of young children with 2 valid doses of measles-containing vaccine. However, laboratory-supported surveillance analysis and outbreak investigations are critical to the identification of gaps in immunity among older age groups, which may need to be filled, and to the identification of strategies to prevent similar outbreaks. The fact that more than a third of infected children in the vaccine-targeted age group were unvaccinated because of vaccination contraindications suggests that an evaluation is needed to ensure the use of valid contraindications only.
It is difficult to identify narrow, age-based risk groups to target for vaccination when a high proportion of adults are unvaccinated and may still be susceptible to measles. Unselective vaccination of millions of adults, based solely on their age, is unlikely to be feasible in China. Additional risk factors for measles need to be identified to develop more feasible immunization strategies. Floating populations represent internal migrants who move to an area temporarily, usually for employment (e.g., migrant workers).
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2015-11-19T21:22:29.539Z", "created": "2015-03-01T00:00:00.000", "id": "23484916", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.3201/eid2103.140646", "pdf_hash": "ed3f681555b9fd818739a0cdec0a8b0d808b2f08", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:32352", "s2fieldsofstudy": ["Medicine", "Environmental Science"], "sha1": "b84c2f36e46d41a51568adc46aedd69d3d470874", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Antimicrobial potential of some plant extracts against Candida species
The increase in the resistance to antimicrobial drugs in use has attracted the attention of the scientific community, and medicinal plants have been extensively studied as alternative agents for the prevention of infections. The Candida genus yeast can become an opportunistic pathogen causing disease in immunosuppressive hosts. The purpose of this study was to evaluate dichloromethane and methanol extracts from Mentha piperita, Rosmarinus officinalis, Arrabidaea chica, Tabebuia avellanedae, Punica granatum and Syzygium cumini against Candida species through the analysis of Minimum Inhibitory Concentration (MIC). Results presented activity of these extracts against Candida species, especially the methanol extract.
Introduction
Medicinal plants and corresponding preparations have been used for a wide range of purposes and for many centuries people have been trying to treat diseases as well as alleviate symptoms by using different plant extracts and formulations (Cowan, 1999). Plants such as Mentha piperita (Iscan et al., 2002), Rosmarinus officinalis (Nascimento et al., 2000), Arrabidaea chica (Taylor, 1998), Tabebuia avellanedae (Machado et al., 2003), Punica granatum (Nascimento et al., 2000;Duraipandiyan et al., 2006) and Syzygium cumini (Chandrasekaram and Venkatesalu, 2004) have been used due to their antimicrobial properties.
In the 2000-2006 period, approximately 50% of new chemical molecules extracted from natural products demonstrated their importance for the development of drugs filtered thereafter. The vegetal residue was reextracted with methanol (Labsynth PA). Solvents were evaporated under reduced pressure and dried using a rotary evaporator (Buchi ® R-200 model). Crude extracts were monitored by chromatography in Thin Layer in silicagel chromatoplaques (60 F254 Merck 1.05554). The dried plant extracts were dissolved in Tween20 (Labsynth) and sterile distilled water, filtered through a 0.22-µm membrane filter (TPP) and stored at 4 °C until further use. The extracts were diluted at the moment of use with the concentration ranging from 1 to 0.001 mg/mL.
Microorganisms: Screening for antimicrobial activities: Preparation of inoculum for susceptibility tests was carried out by microdilution as set forth by the CLSI's M27-A2 recommendation protocol (CLSI, 2002). The yeasts were grown overnight at 37 °C in Sabouraud Dextrose Agar (Merck) plates, and inocula for the assays were prepared by diluting scrape cell mass in 0.85% NaCl solution, adjusted to 0.5 Mc Farland scale and confirmed by spectrophotometric reading at 625 nm. Cell suspensions were finally diluted to 5.0 × 10 3 CFU/mL.50 µL of diluted extract were added in 50 µL of RPMI-1640 in microplates (96 wells) + 100 µL of microorganisms. After that, the samples were incubated at 37 °C for 24-48 hours, in duplicate. Fluconazol was used as control standard in concentrations ranging from 64-0.125 µg/mL. The microplates were incubated at 37 °C for 48 hours.
Results
All Candida species in the in vitro test presented sensitivity to the plant extracts in use, though not for all extracts (Table 1).mainly HIV-positive patients (Erköse and Erturan, 2007). A particular characteristic of Candida is its ability to invade oral mucosa tissues by the development of hyphae, which adhere to tissue surfaces and lead to inflammation (Ellepola and Samaranayake, 2001).
Although the colonisation by Candida albicans is common and causes severe injuries in immunocompromised patients, other Candida species have been isolated from such patients, healthy patients and children with Down's syndrome, such as C. glabrata, C. krusei, C. tropicalis, C. lusitaniae, C. parapsilosis, C. guilliermondii (Erköse and Erturan, 2007;Höfling et al., 2001;Ribeiro et al., 2006;Rodrigues et al., 2004) and C. dubliniensis. These yeasts were recognised as the source of mucosal infection in HIV-positive patients and regarded as a significant cause of infections in humans, such as abdominal infections and fungemia (Erköse and Erturan, 2007).
The purpose of this study was to evaluate the potential activity of extracts from six selected plants against ten Candida species. Preparation of extracts: Each selected plant (5 g) was extracted with 600 mL of dichloromethane (Labsynth PA) with a Dispersive Extratur (Quimis ® Q-252-28 model) and
Discussion
Because of the increasing development of drug resistance to human pathogens and the appearance of undesirable effects of certain antifungal agents, the search for new antimicrobial agents is of great concern today (Phongpaichit et al., 2005). A multidisciplinary approach to drug discovery, involving the generation of truly novel molecular diversity from natural product sources combined with total and combinatorial synthetic methodologies, and including the manipulation of biosynthetic pathways, provides the best solution to the current productivity crisis facing the scientific community engaged in drug discovery and development (Newman and Cragg, 2007).
Our results revealed a strong activity of Punica granatum, Syzygium cumini and Rosmarinus officinalis (dichloromethane and methanol extracts), Arrabidaea chica (dichloromethane extract), Mentha piperita and Tabebuia avellanedae (methanol extract), with MIC varying from 0.06 to 0.001 mg/mL; and no or less activity of Arrabidaea chica (methanol extract), Mentha piperita and Tabebuia avellanedae (dichloromethane extract). The methanol extract of A. chica showed activity against yeasts within 24 hours (data not shown). However, these yeasts showed resistance to this extract after 48 hours. Among the 10 yeasts used in this study, the most resistant were C. glabrata and C. utilis, and C. krusei and C. guilliermondii were the most sensitive strains to the tested extracts.
Data obtained from other studies demonstrated positive results for these plants as well. Duraipandiyan et al. (2006) observed inhibitory effects of P. granatum, while Nascimento et al. (2000) showed activity of extracts of P. granatum, S. cumini and R. officinalis against Candida albicans. Chandrasekaram and Venkatesalu (2004) reported effectiveness of methanol extract of S. cumini and Portillo et al. (2001) found positive results for the dichloromethane extract of T. avellanedae against yeast C. albicans, thus validating our results. On the other hand, Duarte et al. (2005) obtained moderate activity testing M. piperita oil against C. albicans while Duraipandiyan et al. (2006), Portillo et al. (2001) and Dulger and Gonuz (2004) observed resistance of the extracts of S. cumini, T. avellanedae (methanol extract) and R. officinalis (ethanolic extract), respectively. Although these results are not in accordance with ours, there may be many other contributing factors, such as the seasonal period when plants are collected.
The findings presented in this paper indicate that the extracts obtained from the selected plants had anti-Candida Selected plants: the species Mentha piperita L. (leaves, voucher no. UEC 1253), Arrabidaea chica (Bonpl.)B. Verl.(leaves, voucher no. UEC 1254), Rosmarinus officinalis L. (leaves, voucher no. UEC 1264), Syzygium cumini L.(seeds, voucher no. UEC 143724), Punica granatum L. (fruit, voucher no. UEC 143723) and Tabebuia avellanedae Lor.ex Griseb (bark, voucher no. UEC 1256) were collected from the experimental field of the Research Centre for Chemistry, Biology and Agriculture, State University of Campinas (CPQBA/UNICAMP), Brazil. Voucher specimens were deposited at the Herbarium of the State University of Campinas (UEC) and were identified by Professor Jorge Yoshio Tamashiro, Ph. D.
Table 1 .
In vitro antifungal activity of dichloromethane and methanol extracts (mg/mL). Punica granatum and Syzygium cumini extracts exerted strong antifungal activity and can be a source for the development of new therapeutic agents, as they inhibited the growth of Candida. Subsequently, bioguided fractionation shall be conducted on Puninca granatum to identify the active compounds against Candida genus species.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2017-08-15T10:17:44.338Z", "created": "2010-11-01T00:00:00.000", "id": "24075823", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/bjb/a/N3FgPxyWxcJg3xmWykfzTNh/?format=pdf&lang=en", "pdf_hash": "d8f768f1d666f529e22782c0694d905a8e7fc2ff", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:36555", "s2fieldsofstudy": ["Medicine", "Environmental Science", "Biology"], "sha1": "d8f768f1d666f529e22782c0694d905a8e7fc2ff", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The Keap1/Nrf2 Protein Axis Plays a Role in Osteoclast Differentiation by Regulating Intracellular Reactive Oxygen Species Signaling*
Background: Nuclear factor E2-related factor 2 (Nrf2) is a master regulator of cytoprotective enzymes. Results: Nrf2 overexpression-mediated cytoprotective enzymes' augmentation blocked RANKL signaling via intracellular ROS attenuation and thereby blocked bone destruction. Conclusion: Nrf2-dependent cytoprotective enzyme expressions play a role in the regulation of osteoclastogenesis by controlling intracellular ROS. Significance: The Keap1/Nrf2 axis could be a novel therapeutic target for the treatment of bone destructive disease.
Reactive oxygen species (ROS) act as intracellular signaling molecules in the regulation of receptor activator of nuclear factor-B ligand (RANKL)-dependent osteoclast differentiation, but they also have cytotoxic effects that include peroxidation of lipids and oxidative damage to proteins and DNA. Cellular protective mechanisms against oxidative stress include transcriptional control of cytoprotective enzymes by the transcription factor, nuclear factor E2-related factor 2 (Nrf2). This study investigated the relationship between Nrf2 and osteoclastogenesis. Stimulation of osteoclast precursors (mouse primary peritoneal macrophages and RAW 264.7 cells) with RANKL resulted in the up-regulation of kelch-like ECH-associated protein 1 (Keap1), a negative regulator of Nrf2. It also decreased the Nrf2/Keap1 ratio, and it down-regulated cytoprotective enzymes (heme oxygenase-1, ␥-glutamylcysteine synthetase, and glucose-6-phosphate dehydrogenase). Nrf2 overexpression up-regulated the expression of cytoprotective enzymes, decreased ROS levels, decreased the number of tartrate-resistant acid phosphatase-positive multinucleated cells, reduced marker genes for osteoclast differentiation, and attenuated bone destruction in both in vitro and in vivo models. Overexpression of Keap1 or RNAi knockdown of Nrf2 exerted the opposite actions. In addition, in vivo local Nrf2 overexpression attenuated lipopolysaccharidemediated RANKL-dependent cranial bone destruction in vivo. This is the first study to show that the Keap1/Nrf2 axis regulates RANKL-dependent osteoclastogenesis through modulation of intracellular ROS signaling via expression of cytoprotective enzymes. This raises the exciting possibility that the Keap1-Nrf2 axis may be a therapeutic target for the treatment of bone destructive disease.
It is well known that ROS act not only as intracellular signaling molecules but also exert cytotoxic effects such as peroxidation of lipids and phospholipids (14) and oxidative damage to proteins and DNA (15). Cells have several protective mechanisms against these oxidative stressors (16). One of the major cellular antioxidant responses is the induction of cytoprotective enzymes, including antioxidative and carcinogen-detoxification enzymes; this occurs through the operation of the cytoplasmic oxidative stress system, which may be activated by a variety of natural and synthetic chemopreventive agents (17). Transcriptional factor nuclear factor E2-related factor 2 (Nrf2) transcriptionally controls the gene expression of many cytoprotective enzymes, such as heme oxygenase-1 (HO-1) (18), NAD-(P)H:quinone reductase (NQO1) (19), ␥-glutamylcysteine synthetase (GCS) (20), and glucose-6-phosphate dehydrogenase (21). However, kelch-like ECH-associated protein 1 (Keap1) negatively regulates Nrf2-dependent transcription of cytoprotective enzymes by inhibiting nuclear translocation of Nrf2, with cytoplasmic ubiquitination and degradation of Nrf2 (22).
Although ROS have been reported to act as intracellular signaling molecules to regulate osteoclast differentiation, little is known about the cytoprotective mechanisms against oxidative stress during osteoclastogenesis, particularly the regulation of cytoprotective enzyme expression by the Keap1/Nrf2 axis. In this study, we hypothesized that Nrf2-dependent induction of cytoprotective mechanisms against oxidative stress would be attenuated during RANKL-mediated osteoclastogenesis. To investigate this hypothesis, an in vitro cell culture system and an in vivo bone destruction model system were utilized.
EXPERIMENTAL PROCEDURES
Animals-All experimental protocols were approved by the Internal Animal Care and Use Committee, Tohoku University, Japan.
Mouse Peritoneal Macrophages-Three C57/B6 mice (Japan SLC Inc., Hamamatsu, Japan) were sacrificed, and cold phosphatebuffered saline solution (PBS) was injected into the peritoneal cavities. Peritoneal lavage fluid was collected and centrifuged to harvest the cells. The cells recovered from the precipitated pellet were pre-cultured with 50 ng/ml of recombinant macrophage colonystimulating factor (Wako Pure Chemical Industries, Ltd., Osaka, Japan) for 4 days, and the attached cells were used as peritoneal macrophages (23).
Mouse Cells from RAW 264.7 Monocytic Cell Line-RAW 264.7 cells were obtained from the Riken BioResource Center (Tsukuba, Japan).
In Silico Gene Chip Analysis of Mouse Osteoclastogenesis-The publicly available microarray dataset for mouse osteoclastogenesis was downloaded from the Genome Network Platform. Cells from the mouse macrophage cell line, RAW 264.7, were stimulated with 100 ng/ml RANKL for 1 day, and the gene expressions at day 0 and day 1 were compared. Using this dataset, the expressions of Keap1, Nrf2 and cytoprotective enzymes were analyzed.
RNA Interference-Validated siRNAs specific for mouse Nrf2 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Nrf2-specific siRNAs or nonsilencing siRNAs were transfected into RAW 264.7 cells using the X-tremeGENE siRNA transfection reagent (Roche Applied Science). siRNA (100 nmol/liter final concentration) was diluted into dilution medium (without serum, antibiotic-free) in 6-well plates; X-tremeGENE siRNA reagent was added, and this was incubated for 20 min at room temperature. A suspension of RAW 264.7 cells (50 ϫ 10 4 cells/well) was then added, and the cells cultured for 24 h with medium containing the transfection reagent.
Overexpression of Keap1 and Nrf2-Expression plasmids encoding human KEAP1 (hrGFP-Keap1; Addgene plasmid 28025) (24) and human NRF2 (pcDNA3-EGFP-C4-Nrf2; Addgene plasmid 21549) (25) were obtained, respectively, from Dr. Qing Zhong (University of California, Berkeley, CA) and Dr. Yue Xiong (University of North Carolina, Chapel Hill) via Addgene. The expression plasmids were transfected into RAW 264.7 cells using the X-tremeGENE HP DNA transfection reagent (Roche Applied Science). After culture for 24 h with medium containing the transfection reagent, the cells were used to confirm overexpression of Keap1 or Nrf2, or for the osteoclastogenesis assay.
Intracellular ROS Detection-Intracellular ROS levels were detected using the Total ROS/Superoxide Detection Kit (Enzo Life Sciences Inc., Farmingdale, NY), according to the manufacturer's instructions. Cells were stimulated with soluble RANKL (sRANKL; Wako Pure Chemical Industries, Ltd.) for 6 h, washed, and collected in PBS supplemented with 2% fetal bovine serum (FBS). The cell suspension was incubated with oxidative stress detection reagent for 30 min on ice. After washing with PBS, intracellular ROS were detected using an Accuri C6 flow cytometer (BD Biosciences). The viable monocyte/ macrophage fraction was gated on an FSC/SSC plot, and ROS levels were monitored in the FL-1 channel.
Osteoclastogenesis Assay-Cells were plated onto 96-well plates at a density of 10 3 cells/well (RAW 264.7 cells) or 10 4 cells/well (mouse primary peritoneal macrophages), in the presence or absence of recombinant sRANKL (50 ng/ml final concentration). The mouse primary peritoneal macrophage culture was supplemented with recombinant macrophage colony-stimulating factor (20 ng/ml). After culture, cells were stained for tartrate-resistant acid phosphatase (TRAP) using an acid phosphatase kit (Sigma) according to the manufacturer's instructions. Dark-red multinucleated cells (Ն3 nuclei) were counted as TRAP-positive multinucleated cells.
Resorption Assay-Mouse primary peritoneal macrophages were seeded on synthesized calcium phosphate substrate (bone resorption assay plate; PG Research, Tokyo, Japan) and stimulated with RANKL. After 7 days of cultivation, cells were removed with bleach, and the calcium phosphate substrate was washed with distilled water and then dried. Photographs were taken, and the average of the resorbed area per field was calculated from 12 images of each sample, using ImageJ software (National Institutes of Health, Bethesda).
Preparation of Nuclear Protein Lysate-Nuclear protein lysate was prepared from RAW 264.7 cells using the DUALXtract cytoplasmic and nuclear protein extraction kit (DualSystems Biotech AG, Schlieren, Switzerland) according to the manufacturer's instructions. Cultured cells were washed with PBS and treated with cell lysis buffer. After centrifugation, the nuclear pellet was washed twice and lysed with nuclear lysis reagent. After centrifugation, the cleared supernatant was used as the nuclear protein extract. The protein concentrations of each of the nuclear lysates were measured with the Quick Start TM protein assay kit (Bio-Rad), and the concentrations were adjusted to be the same. After mixing with 4ϫ sample buffer containing -mercaptoethanol, the samples were heat-denatured at 70°C for 10 min.
Western Blot Analysis-The prepared nuclear lysates, containing equal amounts of protein, were electrophoresed on a TGX Precast gel (Bio-Rad), and the proteins were transferred to a polyvinylidene difluoride (PVDF) membrane using an iBlot blotting system (Invitrogen). The transferred membrane was treated with Qentix Western blot signal enhancer (Thermo Fisher Scientific Inc., Rockford, IL) according to the manufacturer's instructions. After washing with deionized water, the membrane was blocked with BlockAce (DS Pharma Biomedical Inc., Osaka, Japan) for 30 min, and then incubated for 1 h with rabbit IgG anti-Nrf2 antibody (Santa Cruz Biotechnology) in Can Get Signal Solution-1 (Toyobo Co. Ltd., Tokyo, Japan). After thorough washing with PBS containing 0.5% Tween 20 (PBS-T), the membrane was incubated for 1 h with HRP-conjugated protein A/G (Thermo Fisher Scientific Inc.) in Can Get Signal Solution-2 (Toyobo Co. Ltd) and washed with PBS-T. Chemiluminescence was produced using Luminata Forte (EMD Millipore Corp., Billerica, MA) and detected with an ImageQuant LAS-4000 digital imaging system (GE Healthcare). To confirm the equivalence of loaded nuclear protein, the membrane was re-probed with Restore Plus Western blot stripping buffer (Thermo Fisher Scientific Inc.) for 15 min, washed, blocked, and then blotted in anti-histone H3 antibody (Cell Signaling Technology Japan, Tokyo, Japan) followed by HRPconjugated protein A/G. Cytoplasmic protein samples were used for the detection of HO-1, NQO1, and GCS. The antibodies used in these experiments were anti HO-1 antibody (StressMarq Biosciences Inc., Victoria, British Columbia, Canada), anti-NQO1 antibody (Abcam plc, Cambridge, MA), and anti-GCS antibody (Thermo Fisher Scientific Inc.). To confirm the equivalence of loaded cytoplasmic protein, TGX Stain-Free precast gels (Bio-Rad) were used for cytoplasmic protein samples. Proteins in the electrophoresed gel were visualized with ultraviolet (UV) treatment before transfer to a PVDF membrane. Visualized protein on the membrane was imaged with an ImageQuant LAS-4000 under UV transillumination, and the value of the total protein band density was used for calibration.
In Vivo Bone Destruction Model-We utilized repeat injections of LPS for our in vivo bone destruction model (32). This bone destruction model induces RANKL-dependent osteoclastogenesis and extensive bone destruction. Thirty two 8-weekold C57B/6 male mice were used in the experiments. The mice were randomized into four equal groups as follows: mock vector injection group (control group); local Nrf2 overexpression group (Nrf2 overexpression group); LPS-induced bone resorption group (LPS group); and LPS-induced bone resorption and local Nrf2 overexpression group (LPS ϩ Nrf2 overexpression AUGUST 9, 2013 • VOLUME 288 • NUMBER 32
Keap1/Nrf2 Axis, ROS, and Osteoclast Differentiation
JOURNAL OF BIOLOGICAL CHEMISTRY 23011 group). Purified LPS from Escherichia coli O111:B4 (Sigma) was injected at a dose of 10 g per site, with transfection reagent, on days 1, 3, 5, 7, and 9. For in vivo transfection, an HVJ envelopevector kit (GenomONE, Ishihara-Sangyo Kaisha Ltd., Osaka, Japan) was used, according to the manufacturer's instructions (33). Administration of the HVJ envelope vector containing the human NRF2 expression plasmid (8 g) was performed at the same time and sites as LPS or PBS injection. Nrf2 vector solution (containing human NRF2 expression plasmid enveloped by HVJ) was injected under anesthesia with a 30-gauge needle at a point on the midline of the skull located between the eyes on days 1, 3, 5, 7, and 9. Mock vector (the original expression plasmid pcDNA3) was injected into the corresponding area as a control in the same way. On day 11, the mice were sacrificed by cervical dislocation. RNA from the excised cranial tissue around the injected site (three RNA samples per group) was extracted and reverse-transcribed as detailed above. Gene expression was examined by real time PCR. Other cranial tissue samples (five samples per group) were fixed overnight with 4% paraformaldehyde in PBS, and the specimen was subsequently scanned with an x-ray microtomography (microCT) system (ScanXmate-E090; Comscantecno Co., Ltd., Kanagawa, Japan). The scanned microCT images were reconstituted using ConeCTexpress software (Comscantecno Co., Ltd.). After reconstitution and obtaining the DICOM files, three-dimensional volume rendering was performed using Pluto software.
Statistical Analysis-All data are presented as the mean Ϯ S. D. All in vitro data are the means of three independent experiments. Multiple comparisons were performed with Tukey's test. p Ͻ 0.05 was considered to be statistically significant.
In Silico Gene Chip Analysis of Mouse Osteoclastogenesis-
Microarray analysis using publicly available dataset revealed that both Keap1 and Nrf2 were down-regulated following stim- (Fig. 1, A and B). At day 1, the same time point as that used for the microarray analysis, the expressions of both Keap1 and Nrf2 were down-regulated. Of note, Nrf2 was extensively down-regulated compared with Keap1, resulting in a reduced Nrf2/Keap1 ratio. At day 2, Keap1 expression was up-regulated, maintaining the decrease in the Nrf2/Keap1 ratio, although the reduction seemed attenuating at day 2. Fig. 1C shows Western blot analysis of intranuclear level of Nrf2. RANKL stimulation gave reduction of intranuclear level of Nrf2. These results suggest that Nrf2-dependent cytoprotective enzyme expression may be attenuated not only at day 1 but also at day 2, which would favor intracellular ROS signaling.
To determine whether this might be the case, the expression of cytoprotective enzymes was measured in RANKL-stimulated RAW 264.7 cells in mRNA level (Fig. 2, A and B) and protein level (Fig. 2C). All cytoprotective enzymes assessed (HO-1, NQO1, and GCS) were down-regulated by stimulation with RANKL, both at day 1 and day 2 (Fig. 2, A and B). As expected, protein level expressions of HO-1, NQO1, and GCS were all attenuated. These results suggest that stimulation with RANKL favors intracellular ROS signaling by attenuating the expression of cytoprotective enzymes.
Keap1 or Nrf2
Overexpression, or Nrf2 Knockdown, in RAW 264.7 Cells-To further clarify the role of the Keap1/Nrf2 axis in RANKL-mediated osteoclastogenesis, expression plasmids for Keap1 or Nrf2, or siRNA for Nrf2, were transfected into RAW 264.7 cells, and real time PCR analysis was performed at day 1 (Fig. 3, A and B). Transfection of the Nrf2 or Keap1 expression plasmids successfully induced high expressions of human Nrf2 mRNA or Keap1 mRNA, respectively, in RAW 264.7 cells. Transfection of siRNA for Nrf2 reduced Nrf2 mRNA expression. The effects of overexpression/knockdown were observed even at day 4 (data not shown). Western blot analysis for Nrf2 using nuclear protein extracts revealed an increase in nuclear Nrf2 in Nrf2-transfected cells (Fig. 3C). In contrast, both Keap1 overexpression and Nrf2 RNAi decreased nuclear Nrf2. These data thus verify that our transfection and RNAi techniques had successfully modified intranuclear level of Nrf2.
Cytoprotective Enzyme Expression in RAW 264.7 Cells Was Increased by Nrf2 Overexpression but Decreased by Keap1
Overexpression and Nrf2 Knockdown-Next, real time PCR was used to observe the effects of changes in intranuclear level of Nrf2 on the expression of cytoprotective enzymes (Fig. 4A). A decrease in Nrf2 nuclear translocation, induced by Keap1 overexpression or Nrf2 RNAi, was associated with a significant reduction in the mRNA expressions of HO-1, NQO1, and GCS. In contrast, the enhancement of Nrf2 nuclear translocation induced by Nrf2 overexpression significantly increased the mRNA expression of these cytoprotective enzymes. These results indicate that our transfection and RNAi techniques were able to regulate transcriptionally the Nrf2-dependent expression of cytoprotective enzymes. Western blot analysis revealed that protein level expressions of HO-1, NQO1, and GCS were augmented by Nrf2 overexpression (Fig. 4B). In contrast, protein level expressions of HO-1, NQO1, and GCS were attenuated by Keap1 overexpression and Nrf2 knockdown. These data thus verify that our transfection and RNAi tech- AUGUST (Fig. 5). Stimulation with RANKL substantially increased intracellular ROS levels in RAW 264.7 cells (Fig. 5A). Nrf2 RNAi, shown above to decrease cytoprotective enzyme expression, further increased intracellular ROS levels as compared with RANKL stimulation alone (Fig. 5B). Similarly, Cytoplasmic extracts containing equal amounts of protein were electrophoresed and transferred to PVDF membrane. The membrane was subjected to Western blot analysis. To confirm the equivalence of loaded cytoplasmic protein, TGX Stain-Free Precast Gels were used. Proteins in the electrophoresed gel were visualized with UV treatment before transfer to a PVDF membrane. Visualized protein on the membrane was imaged under UV transillumination, and the value of the total band density was used for calibration. Densitometry analysis was used to calculate the ratio relative to the control, and the mean values of three experiments are shown above each panel. NS, no significant difference between groups.
Keap1/Nrf2 Axis, ROS, and Osteoclast Differentiation
Keap1 overexpression also elevated intracellular ROS levels (cyan line in Fig. 5C). In contrast, Nrf2 overexpression, shown above to increase cytoprotective enzyme expression, reduced intracellular ROS levels as compared with RANKL stimulation alone (orange line in Fig. 5C). These results suggest that RANKL-mediated ROS production is regulated by the Keap1/ Nrf2 axis via the expression of cytoprotective enzymes.
Osteoclastogenesis Was Reduced by Nrf2 Overexpression and Increased by Keap1 Overexpression or Nrf2 Knockdown-Next, it was determined whether changes in intracellular ROS levels influenced RANKL-mediated osteoclastogenesis in the mouse macrophage cell line (RAW 264.7 cells) and mouse primary macrophage (Fig. 6). Similar results were found both in RAW 264.7 cells and in primary peritoneal macrophages. The enhancement of ROS levels by Nrf2 RNAi or Keap1 overexpression significantly increased the number of tartrate-resistant acid phosphatase-positive (TRAP ϩ ) multinucleated cells, compared with RANKL stimulation alone. In contrast, a reduction in ROS levels by Nrf2 overexpression significantly reduced the number of TRAP ϩ multinucleated cells, compared with RANKL stimulation alone. Not only the number of TRAP ϩ multinucleated cells but also the size of TRAP ϩ multinucleated FIGURE 6. Osteoclastogenesis assay. A, mock, Nrf2, and Keap1 expression plasmids were transfected into RAW 264.7 cells; the cells were then stimulated with RANKL for 3 days, and TRAP staining was performed. The data shown are representative of three independent experiments performed in triplicate. *, p Ͻ 0.05 versus mock transfection. Bar, 100 m. Arrowheads indicate TRAP ϩ multinucleated cells. B, nonsilencing or Nrf2 siRNA was transfected into RAW 264.7 cells; the cells were then stimulated with RANKL for 3 days, and TRAP staining was performed. The data shown are representative of three independent experiments performed in triplicate. *, p Ͻ 0.05 versus nonsilencing siRNA transfection. Bar, 100 m. Arrowheads indicate TRAP ϩ multinucleated cells. C, osteoclastogenesis assay using mouse primary peritoneal macrophages. The cells were transfected with the Nrf2 or Keap1 expression plasmid, or Nrf2 siRNA, 12 h before stimulation with RANKL. The cells were stimulated with RANKL for 7 days, and TRAP staining was performed. The culture medium was exchanged once at day 4. All culture media for peritoneal macrophage culture were supplemented with recombinant macrophage colony-stimulating factor (20 ng/ml). The data shown are representative of three independent experiments performed in triplicate. Bar, 100 m. *, p Ͻ 0.05 versus stimulation with RANKL alone. NS, no significant difference between groups. AUGUST 9, 2013 • VOLUME 288 • NUMBER 32 cells were affected by knockdown or transfection of Nrf2 and Keap1.
Keap1/Nrf2 Axis, ROS, and Osteoclast Differentiation
To explore further the influence of changes in intracellular ROS levels on RANKL-mediated osteoclastogenesis, the expression of osteoclast differentiation marker genes (ATP6v0d2, cathepsin K, DC-STAMP, MMP9, OSCAR, and TRAP) was examined using real time PCR (Fig. 7). Consistent with the observed changes in the numbers of TRAP ϩ multinucleated cells, the elevation of ROS levels by Nrf2 RNAi or Keap1 overexpression significantly induced the expression of osteoclast differentiation marker genes, whereas a reduction in ROS levels induced by Nrf2 overexpression was associated with a significant reduction in the expression of these genes. These data support the proposal that intracellular ROS levels, regulated by the Keap1/Nrf2 axis, play a role in osteoclastogenesis.
Osteoclast Resorption Activity Was Attenuated by Nrf2 Overexpression-Next, the effects of Nrf2 overexpression and knockdown on the resorption activity of osteoclasts were investigated (Fig. 8). Consistent with our observations of the number of TRAP ϩ multinucleated cells and the expression of osteoclast marker genes, Nrf2 overexpression caused a reduction in the resorbed area, whereas Nrf2 knockdown resulted in an increase in the resorbed area. These data indicate that exogenous regulation of Nrf2 in osteoclasts is a significant mechanism to control not only osteoclastogenesis but also osteoclast activity.
Local Nrf2 Overexpression Ameliorates Bone Destruction in Vivo-Finally, we examined whether local Nrf2 overexpression was able to reduce bone destruction in vivo. Local injections of lipopolysaccharide (LPS; repeated on five occasions) induced the gene expression of osteoclast markers (MMP9 and TRAP) and osteoclastogenic cytokines (RANKL and TNF-␣) (Fig. 9A). Consistent with this induction, cranial bone resorption was observed in the samples following injection of LPS alone (Fig. 9, B and C). Real time PCR analysis for human Nrf2 revealed that local Nrf2 gene transfer gave extensive induction of human Nrf2 mRNA expression in sites (C t value are as follows: not detectable (over 50 cycles) in control or LPS injection group; 24.1 in control ϩ Nrf2 overexpression group; and 20.8 in LPS injection ϩ Nrf2 overexpression group, respectively). Local Nrf2 overexpression with LPS inhibited the LPS-induced gene expression of these osteoclast markers and osteoclastogenic cytokines (Fig. 9A), as well as LPS-induced bone resorption (Fig. 9, B and C). No statistically significant difference in bone destruction was found between the control and LPS ϩ Nrf2 overexpression group, signifying almost complete amelioration of bone destruction was obtained by local Nrf2 overexpression. These results suggest that exogenous Nrf2 induction may be a potential therapeutic target for the treatment of bone destructive disease.
DISCUSSION
The main findings of this study are that the Keap1/Nrf2 axis plays a major role in the regulation of RANKL-mediated osteoclastogenesis, by controlling intracellular ROS levels via transcriptional regulation of cytoprotective enzyme expression. The results, together with the publicly available microarray data, indicate that there was a reduction in the Nrf2/Keap1 ratio during osteoclastogenesis, and this led to transcriptional down-regulation of cytoprotective enzymes, signifying the induction of intracellular ROS. Our studies to detect ROS by flow cytometry, as well as those of other research groups (13,34,35), show an enhancement of intracellular ROS levels during RANKL-mediated osteoclastogenesis. Furthermore, our experiments involving overexpression and knockdown of Keap1 and Nrf2 clearly demonstrate that an increase in Nrf2 attenuated osteoclastogenesis, whereas a reduction in Nrf2 induced osteoclastogenesis. In addition, intracellular ROS removal by ROS scavenger such as catechin also attenuated RANKL-mediated osteoclastogenesis (data not shown). Taken together, the results provide strong evidence that the Keap1/Nrf2 axis plays a role in osteoclastogenesis by modulating intracellular ROS levels via cytoprotective enzymes.
The role of ROS in intracellular signaling has been investigated previously, revealing signal transduction between RANK and ROS. Wang et al. (36) reported that Rac1 is responsible for regulating ROS generation during osteoclast differentiation, and Lee et al. (34) reported that TRAF6 plays a key linkage role in ROS production by RANKL. Furthermore, the latter group and others have reported that ROS derived from NADPH
Cells have protective mechanisms against oxidative stressors such as ROS, via induction of cytoprotective enzymes through transcriptive regulation by Nrf2 (16). These protective mecha-nisms could potentially interfere with ROS induction by RANKL. However, to date there have been no reports, to our knowledge, concerning the cytoprotective mechanisms against oxidative stress during osteoclastogenesis, especially the regulation of cytoprotective enzyme expression by the Keap1/Nrf2 axis. In this study, it was hypothesized that Nrf2-mediated induction of cytoprotective mechanisms against oxidative stress would be attenuated during RANKL-dependent oste- . Local Nrf2 overexpression ameliorates bone destruction in vivo. LPS (10 g/site) or PBS were injected five times into each mouse, every other day, with or without local human Nrf2 gene transfer. A, real time PCR analysis was performed using three RNA samples per group. Mean C t values of human Nrf2 in each group are given under "Results." *, p Ͻ 0.05 between groups. B, x-ray microtomographic images were taken using five samples per group, and representative photographs are shown. Arrowheads indicate resorbed lacunae or holes formed in the cranial bone. Bar, 1 mm. C, resorbed area in the cranial bone was calculated using ImageJ software. Mean values are shown for percent resorbed (calculated from the ratio of the number of pixels in the resorbed area in the cranial bone to the number of pixels in the cranial bone in the analyzed image). *, p Ͻ 0.05 versus control. NS, no significant difference versus control. †, p Ͻ 0.05 between groups. AUGUST 9, 2013 • VOLUME 288 • NUMBER 32 oclastogenesis. As predicted by this hypothesis, the Nrf2/Keap1 ratio decreased following stimulation by RANKL, and this was associated with a reduction in cytoprotective enzyme expression. The mechanism by which stimulation with RANKL reduces Nrf2 is not currently known. There is a report that Keap1 has highly reactive thiol groups in its structure and that oxidation of this domain leads to substantial conformational changes in Keap1, resulting in dissociation from Nrf2 and hence nuclear translocation of Nrf2 (39). In addition, Nrf2 appears to autoregulate its own expression through an antioxidant response element-like element located in the proximal region of its promoter (40). Taken together, this evidence implies that an increase in ROS levels induced by stimulation with RANKL may up-regulate Nrf2, although any potential increase in Nrf2 expression in osteoclast precursors following stimulation with RANKL was not detected in the present experiments. It has also been reported that Nrf2 regulates Keap1 by controlling its transcription (41). Change of stability of Nrf2 mRNA or decrease of translation by miRNA or other hidden mechanism regulates RANKL-dependent Nrf2 down-regulation. Bach1, an inhibitor of Nrf2 binding to the antioxidantresponse element, could participate this mechanism, because Bach1 knock-out mice show attenuated osteoclastogenesis (42). Further extensive investigations will be required to clarify the regulatory mechanisms linking Nrf2 to stimulation with RANKL. Induction of intranuclear Nrf2 induces up-regulation of cytoprotective enzyme, including HO-1. HO-1 catalyzes heme and produces biliverdin, iron, and carbon monoxide, which have anti-oxidative functions (43). We presumed that up-regulated cytoprotective enzyme decreases intracellular ROS, thereby attenuating ROS-mediated RANKL signaling.
Keap1/Nrf2 Axis, ROS, and Osteoclast Differentiation
In addition to the findings in cultured osteoclasts from the RAW 264.7 cell line or mouse primary peritoneal macrophages, it was also observed that local Nrf2 gene transfer caused in vivo inhibition of LPS-mediated RANKL-dependent osteoclastogenesis. It is likely that this phenomenon was due to direct inhibition of osteoclastogenesis by Nrf2 overexpression. However, other cells such as immune cells and osteoblast/stromal cells also participate in the regulation of osteoclastogenesis during in vivo bone destruction. Further investigations of the effects of Nrf2 overexpression on immune cells and osteoblast/ stromal cells will be required. Fig. 10 summarizes our proposed mechanism by which the Keap1/Nrf2 axis regulates osteoclastogenesis. Briefly, stimulation by RANKL reduces Nrf2 nuclear translocation, which leads to the attenuated expression of cytoprotective genes. Intracellular ROS, which are important signaling molecules downstream of RANK, may be increased by the attenuation of cytoprotective enzyme expression, thereby favoring osteoclastogenesis. Our results suggest that Nrf2 overexpression might be a potential therapeutic target for the treatment of bone destructive disease such as periodontitis and rheumatoid arthritis.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2017-09-09T03:43:54.383Z", "created": "2013-06-25T00:00:00.000", "id": "23686502", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "http://www.jbc.org/content/288/32/23009.full.pdf", "pdf_hash": "c0e30106a385b819e640f9c3e17d8b5df6f9da12", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:36633", "s2fieldsofstudy": ["Medicine", "Environmental Science", "Biology"], "sha1": "f0110de98cdf68288e3a9a46b3198f3d16923643", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Reactive Oxygen Species Released from Hypoxic Hepatocytes Regulates MMP-2 Expression in Hepatic Stellate Cells
Hypoxia is a common environmental stress factor and is associated with fibrogenesis. Matrix metalloproteinase-2 (MMP-2), produced by hepatic stellate cells (HSCs), plays an important role in liver fibrogenesis. However, inconsistent results have been reported on the impact of hypoxia on MMP-2 expression and activity in HSCs. We speculated that cell–cell interaction is involved in the regulation of MMP-2 expression and activity at low oxygen level in vivo. Therefore, in this report we investigated the mechanism by which hypoxic hepatocytes regulates MMP-2 expression in HSCs. Our results showed that the conditioned medium from hypoxia-treated rat hepatocytes strongly induced the expression of MMP-2 mRNA and protein in rat HSC-T6 cells. Reduced glutathione neutralized ROS released from hypoxic hepatocytes, leading to reduced MMP-2 expression in HSC-T6 cells. In addition, phospho-IκB-α protein level was increased in HSC-T6 cells treated with hypoxia conditioned medium, and NF-κB signaling inhibitor inhibited MMP-2 expression in HSC-T6 cells. Taken together, our data suggest that ROS is an important factor released by hypoxic hepatocytes to regulate MMP-2 expression in HSCs, and NF-κB signaling is crucially involved in ROS-induced MMP-2 expression in HSCs. Our findings suggest that strategies aimed at antagonizing the generation of ROS in hypoxic hepatocytes and inhibiting NF-κB signaling in HSCs may represent novel therapeutic options for liver fibrosis.
Introduction
It is estimated that over 100 million people suffer from liver fibrosis in the world. As the main complication of chronic liver damage, liver fibrosis is a wound healing process characterized by the accumulation of extracellular matrix (ECM) proteins in the liver. Although liver fibrosis is caused by a variety of etiologic agents, including chronic viral hepatitis, alcohol toxicity, autoimmune disease, and hereditary metabolic disorders, it is now generally accepted that a central pathologic mechanism underlying liver fibrosis is the generation and proliferation of smooth muscle α-actin (α-SMA)-positive myofibroblasts of periportal and perisinusoidal origin that arise as a consequence of the activation of hepatic stellate cells (HSCs) [1][2][3]4]. HSCs play a critical role in the excessive production and secretion of ECM, resulting in the generation of fibrous tissue and scar formation [5]. The activation of HSCs is considered to be the key factor responsible for liver fibrosis [1]. In addition to cytokines and other soluble factors released by Kupffer cells, inflammatory cells and damaged hepatocytes, changes of ECM composition have been suggested to be implicated in the activation of HSCs [6].
Enzymes known to degrade type IV collagen include matrix metalloproteinase-2 (MMP-2) and matrix metalloproteinase-9 (MMP-9), both of which are members of zinc-dependent matrix metalloproteinase family. MMP-2 is involved in ECM remodeling during tissue and organ development, wound repair, tumor invasion and metastasis. It has been shown that activated HSCs proliferate quickly and produce a large amount of ECM and MMP-2 [7,8]. Therefore, MMP-2 and HSCs activation are interdependent. During the progression of liver fibrosis, deposition of type Ⅳ collagen increases significantly, which might be due to the inability of MMP-2 to degrade type Ⅳ collagen. If the expression and activity of MMP-2 were interfered at different stages, the progression of liver fibrosis may be changed. Thus, further characterization of HSCs is crucial for the understanding of the pathogenesis of hepatic fibrosis.
Many factors have been shown to affect MMP-2 expression and activity in HSCs, such as reactive oxygen species (ROS) generated during oxidative stress, transforming growth factor β1 (TGF-β1), p53, AP-1, membrane-type matrix metalloproteinase (MT1-MMP), tissue inhibitor of matrix metalloproteinase-2 (TIMP-2), plasmin, thrombin, typeⅠcollagen, Zn 2+ , reversion-inducing-cysteine-rich protein with Kazal motifs (RECK), and the interaction between hepatocytes and hepatic stellate cells [9][10][11]. Oxygen is necessary for cellular metabolic process and acts as a modulator of gene expression [12]. One of the most prominently pathological significance of hypoxic mammals is an increase in ECM synthesis. The steady-state concentration of the collagens depends on both the rate of their synthesis and their degradation. Hypoxia is involved in the pathogenic process of liver fibrosis and may activate HSCs [13][14][15]. In addition, our previous studies have shown that hypoxia stimulated MMP-2 synthesis in HSCs in vitro, and the expression of hypoxia inducible factor-1α was increased in hepatocytes in the rat liver fibrosis tissues [16][17][18]. However, it is unknown whether and how hepatocytes modulate MMP-2 expression in HSCs under hypoxic conditions. Therefore, in this study we investigated the mechanism by which hypoxic hepatocytes regulates MMP-2 expression in HSCs. Our results suggest that ROS is an important factor released by hypoxic hepatocyte to regulate MMP-2 expression in HSCs and NF-κB signaling is crucially involved in these processes.
Cell Culture and Preparation of Conditioned Media
Rat hepatocyte BRL-3A cells and hepatic stellate cell HSC-T6 cells were obtained from Type Culture Collection of Chinese Academy of Sciences and cultured in Dulbecco's modified Eagle medium (DMEM, Invitrogen, USA) containing 10% fetal calf serum (FBS, Gibco, USA) and 100 units/mL penicillin/streptomycin at 37 °C in a humidified atmosphere with 5% CO 2 . Cells were plated in 6-well plates at a density of 2 × 10 5 cells/well. At 80% confluence, BRL-3A cells were treated with hypoxia. The cells were replaced with serum-free DMEM and then incubated in the chambers flushed with air mixture containing 1% O 2 , 5% CO 2 and 94% N 2 for 30 min. The ultimate oxygen tension was 5%. The chambers were sealed and placed at 37 °C and incubated for 12 h. Controls included parallel cultures in which cells were exposed to ambient air (e.g., normaxia/21% oxygen tension). The supernatant was collected from the cultured cells at 12 h and passed through a 0.22 μm filter. The filtrate was defined as hypoxia conditioned medium and normoxia conditioned medium. The conditioned medium was used immediately for following experiments.
To induce HSC-T6 cell differentiation and MMP-2 expression by hypoxia conditioned medium, HSC-T6 cells were grown to 80% confluence and then washed twice with PBS. Then HSC-T6 cells were cultured in hypoxia conditioned medium and collected after 6 h, 12 h and 24 h, respectively. As controls, HSC-T6 cells were cultured in DMEM medium or normoxia conditioned medium. All the cells were cultured in a humidified atmosphere of 5% CO 2 and 95% air at 37 °C.
Real-Time RT-PCR
Total RNA was extracted from HSC-T6 cells with Trizol reagent (Invitrogen, USA) following the manufacturer's instructions. Total RNA (1 μg) was used for cDNA synthesis using PrimeScript RT reagent kit (TaKaRa, Dalian, China). Aliquot of diluted first-strand cDNA was amplified with a Real-Time PCR Detection System (ABI7300, USA) using SYBR PrimeScript RT-PCR Kit according to the manufacturer's instructions. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control. The following primers were used: MMP-2, 5-AGGGCACCTCCTACAACAGC-3 and 5-CAGTGGACATAGCGGTCTCG-3 (126 bp) [19]; GAPDH, 5-GAAGGGCTCATGACCACAGT-3 and 5-GGATGCAGGGATGATGTTCT-3 (117 bp) [20]. RT-PCR was performed three times in triplicate. The relative mRNA level of MMP-2 was compared to that of GAPDH and was calculated by the 2 −ΔΔCt method. Each Ct value used for these calculations was the mean of the triplicate for each reaction.
Gelatin Zymography for MMP-2
MMP-2 activity was determined by gelatin zymography. Supernatants were collected from HSC-T6 cells and centrifuged. Protein content of the supernatants was determined by the Bradford assay. Total protein for each sample (15 μg) was fractionated on 10% SDS-PAGE gel containing 0.1% gelatin under non-reducing conditions. Gelatin zymography was performed using a MMP Zymography assay kit (Applygen Technologies Inc.) according to the manufacturer's instructions. Gelatinolytic bands were observed as clear zones against the blue background and the intensity of the bands was estimated using the ScnImage Software.
Quantitative Assay for Reactive Oxygen Species
BRL-3A cells were cultured and treated with hypoxia as previously described in 96-well plates. The final oxygen tensions in the chambers were 5% and 10%, respectively. Controls included parallel cultures in which cells were exposed to normoxia (21% oxygen tension). 12 h later, the supernatants were collected and put in another 96-well cell culture plates. The ROS in the supernatants was measured by ROS assay kit (Genmed, China). Briefly, the samples were incubated with 10 μL 3,3',5,5' tetramethylbenzidine (TMB Substrate) at 37 °C for 1 h. This would result in blue color development proportional to the amount of ROS and the level of ROS in each sample was measured with the spectrophotometer at a wavelength of 650 nm.
Treatment of BRL-3A Cells with Reduced Glutathione
BRL-3A cells were cultured in 6-well plates. At 80% confluency, the medium was changed to serum-free DMEM medium containing reduced glutathione (0, 0.5, 2.5, 10 mmol/L). After 30 min, the plates were incubated in a chamber flushed with air mixture containing 1% O 2 , 5% CO 2 and 94% N 2 for 30 min. The chambers were sealed and placed at 37 °C and incubated for 12 h. The level of ROS in each sample was measured according to the method above. Meanwhile, the supernatants of BRL-3A were collected from the cells and passed through a 0.22 μm filter. The filtrate was designated as ROS-reduced-hepatocyte-conditioned medium and stored at −20 °C for later use. The ROS-reduced-hepatocyte-conditioned medium was used to culture HSC-T6 cell as previously described for 6 h, 12 h, and 24 h. Then the supernatants were collected, and total protein and RNA were extracted from HSC-T6 cells.
Statistical Analysis
Experiments were repeated at least three times. The results were presented as means ± SEM. Comparison of multiple parameters was performed with one-way ANOVA test (SPSS 11.5), followed by Tukey-Kramer's post hoc test. p < 0.05 was considered as statistically significant.
Hepatocyte Conditioned Medium Upregulates MMP-2 Expression in HSCs
Our prior study has shown that hypoxia could regulate the expression of MMP-2 mRNA in HSC-T6 cell [16,17,21]. To investigate whether hypoxic hepatocytes affect the expression of MMP-2 mRNA in HSC-T6 cells, we measured MMP-2 mRNA level in rat HSC-T6 cells cultured in hepatocyte-conditioned medium for 6, 12 and 24 h, respectively. As shown in Figure 1, real-time RT-PCR analysis showed that MMP-2 mRNA expression was increased in HSC-T6 cells treated with hepatocyte-conditioned medium. Time factor analysis by one-way ANOVA showed that MMP-2 mRNA level in HSC-T6 cells treated with the conditioned medium was significantly higher than that in the controls (p < 0.001). These results indicate that hepatocyte conditioned medium upregulates MMP-2 expression in HSCs. To provide further evidence for this, we performed Western blot analysis to examine MMP2 expression at the protein level in HSC-T6 cells cultured with hepatocyte-conditioned medium at 6, 12 and 24 h, respectively. The results showed that MMP-2 protein was induced in HSC-T6 cells treated with the hepatocyte-conditioned medium. The protein level of MMP-2 at 12 h was higher than that at 6 h and 24 h (Figure 2). Taken together, these data suggest that MMP-2 expression in HSC-T6 cells exhibited a slow response to hepatocyte-conditioned medium.
To find out whether anoxic hepatocytes affect the activity of MMP-2 in HSC-T6 cells, gelatin zymography was conducted. The results showed that MMP-2 activity was inhibited by hepatocyte-conditioned medium. The effect within 24 h was more intensive. Time factor analysis by one-way ANOVA test showed that MMP-2 activity in HSC-T6 cells in hepatocyte-conditioned medium and in the controls had significant differences (p = 0.024). Group factor analysis (p < 0.001) was shown in Figure 3, indicating that hepatocyte-conditioned medium inhibits MMP-2 activity in rat HSCs.
Reduced Glutathione Antagonizes the Generation of ROS in the Supernatants of Hepatocytes
Because ROS could regulate MMP-2 expression, we determined the level of ROS in the supernatant of hepatocytes treated with hypoxia. BRL-3A cells were treated with different levels of oxygen (5% O 2 , 10% O 2 , and 21% O 2 ) for 12 h and supernatants were collected for quantitative colorimetric assay. The results showed that ROS level gradually increased with the decrease of oxygen tension ( Figure 4A). To investigate whether reduced glutathione (GSH) could eliminate the release of ROS to the supernatant of hepatocytes treated with hypoxia, we treated BRL-3A cells with GSH (0, 0.5, 2.5, 10 mmol/L), and examined ROS level in the conditioned medium. The results showed that the level of ROS in hypoxia conditioned medium decreased gradually with the increase of reduced glutathione ( Figure 4B).
ROS Generated by Hypoxic Hepatocyte Contributes to the Downregulation of MMP-2 Expression and Activity in HSCs
To confirm that ROS generated by hypoxic hepatocyte is one of the main factors that contribute to increased expression of MMP-2 in HSCs, we examined MMP-2 expression at both mRNA and protein levels in HSC-T6 cells cultured in ROS-neutralized-hepatocyte conditioned medium. Real-time RT-PCR analysis showed that with the increase of GSH concentration, MMP-2 mRNA expression in HSC-T6 cells was significantly inhibited ( Figure 5A, p < 0.001). In addition, Western blot analysis showed that MMP-2 protein expression in HSC-T6 cells was reduced corresponding to increased GSH concentration ( Figure 5B, p < 0.001). The result of gelatin zymography ( Figure 5C) was consistent with Real-time RT-PCR and Western blot (p < 0.001). Figure 5. ROS-neutralized hepatocyte conditioned medium downregulates MMP-2 expression and activity in HSCs. BRL-3A cells were treated with reduced glutathione (0, 0.5, 2.5, 10 mmol/L) and exposed to 5% oxygen for 12 h. Then the supernatant was collected and HSC-T6 cells were cultured with ROS-neutralized-hepatocyte-conditioned medium for 6, 12 and 24 h. As a control, HSC-T6 cells were cultured in serum-free DMEM medium. (A) relative MMP-2 mRNA level in HSC-T6 cells were determined by Real-time RT-PCR. Data were expressed as means ± SEM from 3 independent experiments. * p < 0.05 vs. the former group; (B) MMP-2 protein level in HSC-T6 cells was detected by Western blot. Shown were representative blots from three independent experiments with similar results; (C) MMP-2 activity measured by gelatin zymography.
Hepatocyte Conditioned Medium Upregulates MMP-2 Expression in HSCs via NF-κB Signaling
To investigate whether ROS in hepatocyte conditioned medium modulates MMP2 expression via NF-κB signaling, we measured the level of phospho-IκB-α protein in HSC-T6 cells cultured with serum-free DMEM, normoxia conditioned medium, hypoxia conditioned medium, respectively, for 12 h. Western blot analysis showed that phospho-IκB-α protein level in HSC-T6 cells treated with the hypoxia conditioned medium was increased compared to that in cells cultured in normoxia conditioned medium ( Figure 6A, p < 0.05). BAY 11-7082 is a commonly used NF-κB inhibitor. When we treated HSC-T6 cells with BAY 11-7082, we found that MMP2 expression in HSC-T6 cells was significantly inhibited compared to cells treated with vehicle control ( Figure 6B, p < 0.05). As expected, BAY 11-7082 inhibited the activation of NF-κB in HSC-T6 cells ( Figure 6C). Collectively, these data suggest that NF-κB signaling mediates ROS-induced MMP-2 expression in HSCs.
Discussion
Hypoxia is crucially involved in acute and chronic liver injury. As a repair process in response to a variety of chronic injury stimuli, the progression of liver fibrosis is accompanied by hypoxia. Following a fibrogenic stimulus, HSCs are activated which then synthesize and deposit a large amount of ECM in the liver [22][23][24]. The activation of HSCs is a key event of fibrogenesis [4,7]. We have found that hypoxia induced the expression of MMP-2 at both mRNA and protein levels, but inhibited its activity in the in vitro assay [16][17][18]. In vivo experiment suggested that at the beginning of liver fibrosis, MMP-2 expression and activity were decreased, and fewer HSCs were activated. However, during the development of fibrosis, MMP-2 activity was increased, which degrades collagen Ⅳ [17,[25][26][27]. Given these inconsistent results between in vivo and in vitro studies, we believe that cell-cell interaction is involved in the regulation of MMP-2 expression and activity at low oxygen level in vivo. Therefore, in this report we investigated the mechanism by which hypoxic hepatocytes regulates MMP-2 expression in HSCs.
First, we examined whether hypoxic hepatocytes affect MMP-2 expression in HSC-T6 cells. Real-time RT-PCR analysis showed that hypoxic hepatocytes may secrete some factor(s) to affect MMP-2 expression at mRNA level in HSC-T6 cells. Consistent with RT-PCR analysis, Western blot analysis further showed that hepatocyte-conditioned medium induced the expression of MMP-2 in HSC-6T cells at protein level. Taken together, these data indicate that hypoxic hepatocytes release some factor(s) to upregulate the expression of MMP-2 in HSC-T6 cells.
Next, a serial of experiments was conducted to identify the factor(s) released from hypoxic hepatocytes responsible for the upregulation of MMP-2 expression in HSCs. It is known that ROS serves as a signal molecule that regulates many important cellular events, such as transcription factor activation, gene expression, cell differentiation and proliferation. ROS is also involved in the regulation of MMP-2 expression and activation, mainly through Ras and mitogen activated protein kinase (MAPK) signaling cascades [21,28]. Therefore, we detected ROS level in the supernatant of BRL-3A cells treated with different levels of oxygen and found that with the decrease of oxygen tension, ROS was gradually increased. Reduced glutathione (GSH) is a non-protein antioxidant and can protect cells from ROS damage. After treatment with GSH, we detected ROS in the supernatants of hypoxia conditioned medium to determine whether ROS can be scavenged by reduced glutathione. The results evidently demonstrated that as the concentration of GSH increased, the level of ROS in the supernatants of hypoxia conditioned medium was gradually decreased, proving that ROS generated in hepatocytes was scavenged efficiently by reduced glutathione.
To provide further evidence that ROS generated in hepatocytes contributes to the upregulation of MMP-2 expression in HSC-T6. Last, we characterized the potential mechanism by which ROS regulates MMP-2 expression in HSCs. By probing phospho-IκB-α protein level as an indication of the activation of NF-κB signaling, we found that hypoxia conditioned medium induced increased phospho-IκB-α protein level in HSC-T6 cells, corresponding to increased MMP-2 protein level. Furthermore, inhibition of NF-κB signaling by specific inhibitor BAY 11-7082 led to reduced MMP2 protein level in HSC-T6 cells. Based on these data we speculate that NF-κB signaling is an important, although not exclusive, signaling pathway that mediates ROS-induced MMP-2 expression in HSCs.
Gelatin zymography analysis ( Figure 5C) showed that hypoxic hepatocytes released ROS to persistently inhibit the activity of MMP-2 in HSC-T6 cell within 24 h. This resutl was consistent with that of Real-time RT-PCR and Western blot. However, in contrast to the result showed in Figure 3, we found that MMP-2 activity was inhibited by hepatocyte-conditioned medium. These results suggest that ROS may be the main factor released by anoxic hepatocyte to regulate MMP-2 mRNA and protein expression, but may not be the main factor that regulates MMP-2 activity.
MMP-2 is secreted as a latent form of zymogen and can be activated when the pro-domain is cleaved and bound to TIMP-2 and MT1-MMP. Many factors are involved in regulating MMP-2 activity. As the main inhibitor, tissue inhibitor of metalloproteinase (TIMP) is the most important regulator of MMP-2 activity. It is possible that the factor released by anoxic hepatocyte regulates MMP-2 activity by means of disturbing the balance between MMP-2 and its endogenous inhibitors TIMPs [29]. As a novel membrane-anchored matrix metalloproteinase (MMP) inhibitor, reversion-inducing cysteine-rich protein with Kazal motifs (RECK) also plays a role in the regulation of MMP activity. RECK inhibits MMP-2, MMP-9, and membrane type-1 MMP (MMP-14) secretion and activity [30]. MMP-2 is Zn-dependent enzyme. Therefore, Zn 2+ is necessary for MMP-2 activity [31]. In addition, MMP-2 is also directly activated by oxidizing the sulphydril bond between a cysteine residue of the prodomain and the Zn 2+ catalytic center, resulting in partial enzyme activation followed by an intramolecular cleavage of the propeptide [32]. In our future study we will identify the factors generated by anoxic hepatocytes that inhibit MMP2 activity.
MMP-2 has been shown to promote the formation and development of liver fibrosis, mainly through the regulation of the activation, proliferation, and migration of HSCs. In vitro studies found that MMP-2 could degrade type IV collagen-rich matrix around the HSCs, which could activate HSCs. The activated HSCs proliferate actively, produce large amounts of extracellular matrix, and are main source of MMP-2 in the liver [33]. In addition, oxidative stress induced by some pathogenic factors promoted the proliferation and migration of HSCs through MMP-2 [34]. MMP-2 also promoted the proliferation of HSCs through release of growth factors stored in the matrix [35]. Furthermore, MMP-2 has been shown to induce the release of VEGF [36], which is involved in hepatic sinusoidal capillarization and promotes the development of liver fibrosis [7,[37][38][39]. Further study is important to investigate the mechanism underling the regulation of the expression and activity of MMP-2 in liver fibrosis.
Conclusions
In summary, in this study we provide several lines of evidence that ROS is an important factor released by hypoxic hepatocyte to regulate MMP-2 expression in HSCs, and NF-κB signaling is crucially involved in ROS-induced MMP-2 expression in HSCs. Our findings suggest that strategies aimed at antagonizing the generation of ROS in hypoxic hepatocytes and inhibiting NF-κB signaling in HSCs may represent novel therapeutic options for liver fibrosis.
==
Domain: Environmental Science Biology Medicine
|
[{"added": "2015-09-18T23:22:04.000Z", "created": "2011-04-07T00:00:00.000", "id": "16178447", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/1422-0067/12/4/2434/pdf", "pdf_hash": "2bf587c31a3e2cebeba7d55fc5aad3eb567a6173", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:32394", "s2fieldsofstudy": ["Medicine", "Environmental Science", "Biology"], "sha1": "2bf587c31a3e2cebeba7d55fc5aad3eb567a6173", "year": 2011}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Correção da magnitude da mortalidade por câncer do colo do útero no Brasil , 1996 – 2005 Correction for reported cervical cancer mortality data in Brazil , 1996 – 2005
MÉTODOS: Os dados sobre os 9.607.177 óbitos foram obtidos do Sistema de Informação sobre Mortalidade, para o período de 1996 a 2005. Para a correção do sub-registro, foram utilizados os fatores de expansão gerados pelo Projeto Carga Global de Doença no Brasil – 1998. Para correção das categorias de diagnósticos desconhecidos, incompletos ou mal defi nidos de óbitos, foi aplicada redistribuição proporcional. Os dados ausentes de idade foram corrigidos por imputação. As correções foram aplicadas por Unidade Federativa e os resultados apresentados para o Brasil, região e áreas geográfi cas (capital, demais municípios das regiões metropolitanas e interior) por meio do percentual de variabilidade da magnitude das taxas, antes e após a correção dos óbitos. O comportamento das correções foi analisado por modelo de regressão linear multivariada com termos de interação entre região do País e área geográfi ca.
INTRODUCTION
Cervical cancer is characteristically a preventable disease that can be detected in early, non-invasive forms. Yet cervical cancer is still an ongoing serious challenge for public health authorities, especially in developing countries accounting for 83% of all cases and 86% of all deaths from cervical cancer. 10cording to the Brazilian Ministry of Health, a age-adjusted cervical cancer mortality rates increased from 4.97 to 5.29 per 100,000 women-years for the period 1979 to 2005, a 6.4% increment in 26 years. However, these rates are low compared to other Latin American countries such as Venezuela (15.2),Ecuador (18.6), Bolivia (22.2),Nicaragua (26.1), and Haiti (53.5). 3 Original Articles a Ministério da Saúde. Departamento de informática do Sistema Único de Saúde. Sistema de Informação de Mortalidade. Brasília; 2006[cited 2009 Dec 30]. Available from: [URL] The Brazilian National Mortality Database (SIM) faces many challenges, mainly related to underreporting of deaths and deaths from ill-defi ned causes. 12They vary across Brazilian regions but are mostly serious in the North and Northeast regions. It can be thus assumed that reported rates have been underestimated, making it necessary to assess the actual impact of cervical cancer in Brazil.
In addition to SIM-related general issues, there is a particular obstacle specifi c to cervical cancer. A significant proportion of deaths from cervical cancer is reported as neoplasm of "uterus, not otherwise specifi ed (NOS)," which hinders the assessment of the actual impact of this disease. In an attempt to address this issue and generate more realistic statistics, researchers 7,9,19 have included in their analysis total cases of uterine cancer not otherwise specifi ed, in addition to cervical cancer cases. In fact, Wünsch & Moncau (2002 ) 19 claim that this inclusion has allowed to reset trends based on socioeconomic profi le of populations and Brazilian macroregions.
The literature describes several approaches that aim to correct issues related to mortality databases and death certifi cates reporting as primary cancer site "uterus, NOS." 1,2,4,14,16,17 The objective of the present study was to develop a methodology for correction of reported cervical cancer deaths in Brazil by combining two approaches: correction of SIM-related issues and those specifi cally related to death certifi cates reporting as primary cancer site "uterus, NOS."
METHODS
All 9,607,177 deaths reported by SIM for the period between 1996 and 2005 were included in the correction of reported cervical cancer deaths. Of them, 39,618 deaths were from cervical cancer. Data prior to 1996 were excluded because factors generated by the Global Burden of Disease Study in Brazil for the year 1998 were used for correction of underreporting of deaths.b The application of these factors would not be suitable for data prior to 1996 because issues related to underreporting and quality of death certifi cates were more signifi cant in the early years after SIM implementation. 12derreporting of deaths was corrected using factors indirectly generated by the Global Burden of Disease Study in Brazil for the year 1998.b Underreporting of deaths in those under one 6 and those one year old and older 15 was thus separately corrected by gender and macroregion. These factors were recalculated c since in that study the corrections were made by macroregion as a whole, disregarding capital cities, other cities in metropolitan areas and interior cities. These aspects were taken into consideration in the present study, but only data from interior cities was corrected assuming that reporting of deaths was close to 100% in capital cities and other cities in metropolitan areas nationwide. For the states of Rio de Janeiro, São Paulo, Mato Grosso do Sul and Rio Grande do Sul reported deaths in interior cities were not corrected assuming that reporting of deaths was also close to 100%. Based on the reference study correction factors were recalculated for so-called "natural" causes of death assuming that reporting of deaths from external causes was close to 100%.
Deaths with unknown (missing), incomplete or ill-defi ned data in SIM (variables such as gender; macroregion; primary cause of death "uterus, NOS;" incomplete diagnoses; and ill-defined signs and symptoms) were corrected applying proportional distribution of deaths coded in each chapter of the International Statistical Classifi cation of Diseases and Related Health Problems -10th revision (ICD-10) among deaths from specifi ed causes, by age groups and excluding external causes. 17For example, all deaths from cancer of the uterus, NOS were redistributed by age, macroregion, geographic area and year among deaths from cervical and uterine body cancer following its original proportion. To estimate missing data on age, single imputation was applied and missing data were replaced by median age, taking into account macroregion, gender and cause of death (three digits).
Since deaths from cancer are most often better reported than other causes of death, 12 the representation of cancers among ill-defi ned causes would be smaller, making a proportional reallocation of all deaths from ill-defi ned causes incorrect. Thus, in an attempt to avoid overestimating deaths from cervical cancer, and because there was no other consensus for reallocation of deaths from ill-defi ned causes, 50% of the estimated weight was attributed to neoplasms by gender, age, geographic area and type of cancer.
Overall, the correction process comprised eight steps: There were included all deaths identified as from cervical cancer after correction, coded ICD-10 C53.
Non-corrected cervical cancer mortality rates were estimated based on deaths obtained from SIM whereas corrected rates were estimated from deaths identifi ed as from cervical cancer after the fi nal process of correction and based on their estimated populations. Rates were also estimated for each stage of correction.
Cervical cancer mortality rates were fi rst estimated for all ages (Table 1) and for the age group 20-79 years old (Tables 2 to 5), aggregated into fi ve-year age groups. The rates were then adjusted for age using a direct method based on world population. 8After adjusting for all age groups, expected deaths of those under 20 years and of those 80 years and more were excluded to obtain adjusted rates for the age group 20-79 years old.
Annual mortality rates from cervical cancer and average rates for the entire period (1996-2005) In the fi nal model, the terms (3) and ( 5) were not included, and their exclusion did not affect the estimates or the precision of confi dence intervals.
The study was approved by the Research Ethics Committee of the Instituto de Medicina Social (Protocol No. FR186658, on 04/10/2008).
RESULTS
In Brazil, for the period between 1996 and 2005, the average crude and adjusted annual rates of deaths from cervical cancer based on world population were 4.6 and 5.1 deaths per 100,000 women-years, respectively. After corrections of deaths, mortality rates from cervical cancer showed an increment of 103.4% in Brazil nationwide, ranging from 35% in capital cities of the South to 339% in interior cities of the Northeast. There were wide variations across regions, within a region, across capital cities, other cities of metropolitan areas, and interior cities (Table 1).
Table 2 shows average annual mortality rates from cervical cancer, total percent variation and percent variation by step of correction. The steps of correction with higher increments in cervical cancer mortality rates were: reallocation of deaths from cancer of the uterus, NOS (55.6%), correction of underreporting (22.0%), and reallocation of deaths with ill-defined causes (21.2%). These three steps of correction accounted for a 98.8% increment in the mortality rate nationwide (17.39 deaths per 100,000 women aged 20-79). Other steps of correction, i.e., reallocation of deaths with unknown macroregion, reallocation of deaths with incomplete diagnosis (missing codes) and reallocation of deaths with unknown gender, accounted for a 4.6% increment in the annual average rate nationwide and for all the study period.
The reallocation of deaths from cancer of the uterus, NOS produced the greatest effect on mortality rates from cervical cancer in all regions and nationwide, even greater than the percent increment seen after the correction of underreporting and reallocation of deaths from ill-defi ned causes. However, the same was not seen in interior cities of the North and Northeast, where the correction of underreporting had a greater effect than reallocation of deaths NOS. These regions also showed the greatest percent increments after the reallocation of deaths with ill-defi ned causes, 57% in the Northeast and 10% in the North.
The corrections of mortality rates form cervical cancer by age group are shown in Table 3. The greatest correction percents were seen in older age groups. The curve of correction of deaths showed an increasing trend with age for Brazil nationwide and all macroregions.
The correction of mortality rates ranged from 70% to 80% in the age group 20-49, up to 120% in the age group 70-79.
The corrections of mortality rates from cervical cancer by year of death are presented in Table 4. They showed a steady pattern over the time series studied, especially nationwide. This steady pattern is evident after excluding the corrections made for the year 2005, which were higher than those made for previous years, and in all regions. The South and Northeast had the lowest and the highest variation, respectively, compared to the percent increment in annual mortality rates from cervical cancer. Table 5 presents the results of the multivariate regression analysis. A comparison of the two models (with and without interaction terms) illustrates the effects of the interaction between corrections in the interior and the reference region (South), especially between the North and Northeast. After the inclusion of interaction terms in the regression model, most corrections for interior cities in these two regions are associated with an increment of the estimated mortality rate from cervical cancer.
DISCUSSION
As expected the corrections of deaths from cervical cancer were more signifi cant in regions with lower socioeconomic status, and within these regions, they were more pronounced for death rates in interior than capital cities or other cities in metropolitan areas. These results are consistent with the literature since higher death rates from cervical cancer are expected in less developed regions. 3,10,18In addition, more corrections were expected in these Brazilian regions because they have the lowest reporting rates and the poorest quality of death certifi cates.f The percent variation of rates at each step of correction also showed a close relationship with prevalent socioeconomic conditions in the regions and their geographical areas (capital, interior and other cities of metropolitan areas).
After correction, the mortality rate from cervical cancer in Brazil had a percent increment above 100%, from 5.1 to 10.4 deaths per 100,000 women-years, adjusted Another study in Latin America estimated a mortality rate of 11.6 cervical cancer deaths per 100,000 womenyears in Brazil in 2000. 3Other Brazilian studies for the correction of cervical cancer rates focused only on the reallocation of deaths from cancer of the uterus, NOS in some cities. Antunes & Wünsch (2006) reallocated cancer of uterus, NOS and obtained an almost 50% increment of rates in São Paulo, southern Brazil. In another study 13 in Recife, northeastern state of Pernambuco, the authors reviewed death certifi cates against medical records and concluded that half of the cases previously reported in the SIM as cancer of the uterus, NOS were due to cervical cancer, increasing deaths from cervical cancer by 20%.
The results of the present study show the relative weight of deaths from uterine cancer, NOS in increasing corrected rates from cervical cancer: over 50% of total increments in the rates nationwide can be attributable to reallocation of the primary cause of death. It can be thus inferred that the Brazilian health system has low diagnostic capacity and that part of female population does not have access to or does not undergo cervical screening. These women are seen at advanced stages of disease when it is more diffi cult to make an accurate diagnosis. The effectiveness of a population-based screening program depends not only on the performance of Pap smears, but especially on the capacity of health services to ensure treatment and follow-up of all patients with abnormal results. 18However, our results show inadequate diagnostic ability in addition to inadequate cervical cancer prevention.
The high proportion of death certifi cates with uterine cancer, NOS as primary cause of death could be explained by the diffi culty in making the diagnosis of cervical cancer at advanced stages of disease. 11owever, the quality of data points to a need for training health providers in death certifi cate completion. 2,4,14,16,17t is also possible that health providers do not see the value of appropriate completion of death certifi cates. In a study of deaths from uterine cancer, NOS the authors were able to reclassify most diagnoses by primary site using a simple approach of data recovery and retrieval. 13herefore it should be further explored the reasons that prevent adequate completion of death certifi cates.
Another example of diagnosis inaccurately completed in death certifi cates is "cardiac arrest," classifi ed as a well-defi ned diagnosis in several ICD revisions in the chapter of diseases of the circulatory system.h It was eventually considered an ill-defi ned cause based on studies showing that, in almost all deaths, the primary cause reported by health providers as "cardiac arrest" was not a cardiac condition. Death could be due to well-defi ned causes, but providers would report only "cardiac arrest" since it is a primary cause easily to be reported, or even because is a well-established cause of death.h It would be valuable to investigate using a combined approach of death information retrieval and providers' survey the reasons that make them report the primary cause of death as cancer of the uterus, NOS where it is possible to classify these diagnoses.
With respect to corrections of mortality rates from cervical cancer by age group, there were increasing corrections with age, which is consistent with the Brazilian 1,13 and international literature. 2 In contrast, corrections by year of death showed a steady pattern over the time series analyzed, while a declining pattern would be expected due to improvements in reporting of deaths and quality of completion of death certifi cates in the SIM.f This fi nding should be interpreted with caution given the intrinsic limitations of correction of underreporting based on expansion factors generated by the Global Burden of Disease Study in Brazil for the year 1998. Improvements in the reporting of deaths may have been masked in this study. However, there are still defi ciencies in the SIM that need to be addressed so that information from all Brazilian regions can be available with similar rates of reporting and data quality. To be continued Regarding the correction of underreporting, although there were corrected only deaths occurring in interior cities of each macroregion, assuming that reporting of deaths was close to 100% in capital cities and other cities in metropolitan areas, we are aware that major cities were included as "interior" ones. For instance, the cities of Uberaba, Uberlandia and Juiz de Fora in the southeastern state of Minas Gerais show adequate reporting i and could have been excluded from the corrections of underreporting. However, the interior of each macroregion was corrected in a single block. Thus, the use of a more accurate method of correction of underreporting 5 could improve the results of this study, providing a more realistic overview of the profi le of mortality from cervical cancer in Brazil by macroregions and their geographic areas.
Another limitation of the present study is regarding the approach for correction of deaths from ill-defi ned causes, which followed the proportional reallocation of deaths from well-defi ned causes by age groups. 17his approach can be inadequate, especially when it is intended to reallocate cancer deaths since they are usually adequately reported. 12But, since there is no consensus in the literature on this subject, we believe that by using 50% of the weight of neoplasms for reallocation of deaths from ill-defi ned causes we have prevented overestimation of cervical cancer mortality.
Despite its limitations, consistent corrections in terms of geographical areas and their agreement with data from literature 3,10,18 confi rm that the method of correction used in this study provided a more realistic estimate of mortality rates from cervical cancer in Brazil.
Corrections of specifi c death estimates are significant especially for cervical cancer which is the most common cancer among low-income women. The direct use of data without taking into consideration quality and other characteristics of mortality information systems can lead to underestimations of actual mortality rates.
Corrected estimates should be used to identify Brazilian regions requiring priority actions for prevention and control of cervical cancer.
In conclusion, given the magnitude of corrected mortality rates from cervical cancer in Brazil (10.4 deaths per 100,000 women-years in the period 1996-2005 among women of all ages), it is evident that cervical cancer mortality is actually higher than that offi cially reported.
The correction of mortality rates is an essential strategy for planning actions for disease control. Other studies including validation and reliability testing of methods of correction of data on the primary cause of death are needed for consistently improving the SIM.
The variability of rates before and after correction of deaths was assessed by percent variation [(corrected rate / non-corrected rate) × 100]. Multiple regression analysis was performed to assess factors associated with variations of mortality rate corrections, where the dependent variable was the ratio between corrected and non-corrected annual cervical mortality rates (corrected rate / non-corrected rate). The independent variables included in the model were: calendar year, macroregion, and geographical area (capital, other cities in metropolitan areas and interior). The Southern region and geographical area of capital cities were considered reference categories.
and geographic division (capital, interior and other metropolitan areas). The current geographical division according to the Brazilian Institute of Geography and Statistics (IBGE) d was used to aggregate geographic areas into capital cities, other cities in metropolitan areas and interior cities. The category corresponding to "other cities in metropolitan areas" was created from a list of cities in nine reference metropolitan areas used in offi cial IBGE publications, excluding capital cities. The nine metropolitan areas were distributed as follows: one in the North (metropolitan area of Belem); three in the Northeast (metropolitan area of Recife, Fortaleza, and Salvador); three in the Southeast (metropolitan area of Belo Horizonte, Rio de Janeiro and São Paulo); and two in the South (metropolitan area of Curitiba and Porto Alegre). Following this classifi cation, the Central-west region does not have a metropolitan area. Population-based and mortality data were collected from the Brazilian Ministry of Health Database (DATASUS) website.eFor the assessment of potential interactions between Brazilian macroregions and their related geographic areas seven region-area interaction terms were fi rst included in the multivariate regression model: (1) North region and the interior area; (2) Northeast region and the interior area; (3) Southeast region and the interior area; (4) Central-west region and the interior area; (5) North region and other cities of metropolitan areas;(6)Northeast region and other cities of metropolitan areas; and (7) Southeast region and other cities of metropolitan areas.
Table 1 .
Average mortality rates from cervical cancer (per 100,000) in women of all ages and variation after correction by macroregion. Brazil, 1996-2005.
a Adjusted based on world population b Other cities of nine metropolitan areas: Belém, Recife, Fortaleza, Salvador, Belo Horizonte, Rio de Janeiro, São Paulo, Curitiba, and Porto Alegre.
Table 2 .
Average mortality rates from cervical cancer (per 100,000) in women aged 20 to 79 years and variation after correction by macroregion. Brazil, 1996-2006.
a Adjusted based on world population.bOther cities of nine metropolitan areas: Belém, Recife, Fortaleza, Salvador, Belo Horizonte, Rio de Janeiro, São Paulo, Curitiba, and Porto Alegre.fMinistério da Saúde. Secretaria de Vigilância à Saúde. Saúde Brasil 2005 -uma análise da situação de saúde. Brasília; 2005.based on world population. This result confi rms the estimated risk of death from cervical cancer by the International Agency for Research on Cancer (IARC) in Brazil in 2002, 10.2 deaths per 100,000 women-years.g
Table 3 .
Average mortality rates from cervical cancer (per 100,000 women) and variation after correction by age in women aged 20-79 years. Brazil, 1996-2006.
Table 4
a Adjusted based on world population.
Table 4 .
Mortality rates from cervical cancer (per 100,000) in women aged 20-79 years and total percent variation after correction by annual series. Brazil, 1996-2006.
Table 5 .
Results of multivariate linear regression models of corrected mortality rates from cervical cancer in women aged 20-79 years by geographical areas. Brazil, 1996-2005. Adjusted by age and other variables of the table, Model I (R2 = 0.566) and II (R2 = 0.747) with and without interaction terms, respectively.
==
Domain: Political Science Biology Medicine
|
[{"added": "2017-06-19T08:25:39.006Z", "created": "2010-08-01T00:00:00.000", "id": "23986055", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/rsp/a/SNFYYFNPfV7D7Sp79hXrqnF/?format=pdf&lang=pt", "pdf_hash": "519b70ff2b8b8416f0bf4cd4f0546a4728d2235a", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:35738", "s2fieldsofstudy": ["Medicine", "Political Science"], "sha1": "519b70ff2b8b8416f0bf4cd4f0546a4728d2235a", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Dual Sensory Impairment in Older Adults Increases the Risk of Mortality: A Population-Based Study
Although concurrent vision and hearing loss are common in older adults, population-based data on their relationship with mortality is limited. This cohort study investigated the association between objectively measured dual sensory impairment (DSI) with mortality risk over 10 years. 2812 Blue Mountains Eye Study participants aged 55 years and older at baseline were included for analyses. Visual impairment was defined as visual acuity less than 20/40 (better eye), and hearing impairment as average pure-tone air conduction threshold greater than 25 dB HL (500–4000 Hz, better ear). Ten-year all-cause mortality was confirmed using the Australian National Death Index. After ten years, 64% and 11% of participants with DSI and no sensory loss, respectively, had died. After multivariable adjustment, participants with DSI (presenting visual impairment and hearing impairment) compared to those with no sensory impairment at baseline, had 62% increased risk of all-cause mortality, hazard ratio, HR, 1.62 (95% confidence intervals, CI, 1.16–2.26). This association was more marked in those with both moderate-severe hearing loss (>40 dB HL) and presenting visual impairment, HR 1.84 (95% CI 1.19–2.86). Participants with either presenting visual impairment only or hearing impairment only, did not have an increased risk of mortality, HR 1.05 (95% CI 0.61–1.80) and HR 1.24 (95% CI 0.99–1.54), respectively. Concurrent best-corrected visual impairment and moderate-severe hearing loss was more strongly associated with mortality 10 years later, HR 2.19 (95% CI 1.20–4.03). Objectively measured DSI was an independent predictor of total mortality in older adults. DSI was associated with a risk of death greater than that of either vision loss only or hearing loss alone.
Introduction
Impaired vision and hearing are common among older adults [1,2] and can occur separately or in combination [3]. Visual loss impacts negatively on functional independence, mental health and cognition, and reduces quality of life, as well as increasing the need for support services [4][5][6][7]. Age-related hearing loss is more frequent and is associated with an increased risk of depression, and impairs quality of life and the ability to conduct activities of daily living, as well as leading to an increased reliance on community and informal supports [2,[8][9][10][11][12][13].
There are also prospective data to suggest that vision loss was associated with a greater mortality risk in older adults [14][15][16]. In a US study of 2520 adults aged between 65-84 years old, worse baseline acuity was associated with a 5% higher mortality rate. Recently, Karpa et al. [14] showed in the Blue Mountains Eye Study (BMES) of adults aged 49 years and over that the presence of presenting visual impairment increased the risk of all-cause mortality by 29%, after adjusting for potential confounders. Agerelated hearing loss has also been shown to be associated with a greater risk of all-cause mortality in older adults [17]. In the Blue Mountains Hearing Study of adults aged 55 years and over, indirect path analyses showed that any level of hearing loss (.25 dB HL) was associated with increased all-cause mortality via three mediating variables: disability in walking, cognitive impairment and self-rated health [17].
There is a lack of population-based studies that have examined the association between the presence of objectively measured hearing and vision impairment, i.e. dual sensory impairment (DSI) and mortality risk. In the US National Health and Nutrition Examination Survey of 5444 adults aged 55-74 years, clinically confirmed DSI, however, was not found to be associated with an increased risk of mortality [18]. In an Italian study of 1140 adults aged 70-75 years, a significant association between DSI and 6year total mortality was observed in men but not in women [19]. In this study, hearing loss was not measured objectively using pure-tone audiometry (i.e. 'gold-standard'), but using the free-field whispered voice test. Finally, results from the US National Health Interview Survey of adults aged 18 years and older, showed that self-reported DSI was associated with a higher risk of all-cause mortality among men and women as compared to those reporting either visual impairment alone or hearing impairment alone [20,21]. Many previous studies that have assessed the association between DSI and mortality had a relatively small number of participants with DSI at baseline and/or did not adequately adjust for mortality risk markers (e.g., cardiovascular diseases, diabetes, cognitive impairment and self-rated health) [18,19,22].
In the present study, we aimed to address previous gaps in knowledge, by examining the association between clinically confirmed DSI with 10-year mortality risk in a large cohort of adults aged 55 years and older, after adjusting for potential confounders such as self-rated health, walking disability and cognitive impairment, in addition to traditional mortality risk markers.
Study Population
The BMES is a population-based cohort study of common eye diseases and other health outcomes in a suburban Australian population located west of Sydney. Study methods and procedures have been described elsewhere [23]. Following a door-to-door census of the region, baseline examinations of 3654 residents aged .49 years were conducted during 1992-4 (BMES-1, 82.4% participation rate). Of the baseline participants, 2335 (75.1% of survivors) returned for 5-year follow-up examinations during 1997-9 (BMES-2), and 1952 participants (53.4% of the original cohort, or 76.6% of survivors) returned for 10-year follow-up examinations during 2002-4 (BMES-3). The study was approved by the Human Research Ethics Committee of the University of Sydney and was conducted adhering to the tenets of the Declaration of Helsinki. Signed informed consent was obtained from all the participants at each examination.
Audiological Assessment
Pure-tone audiometry at both visits was performed by audiologists in sound-treated booths, using TDH-39 earphones and Madsen OB822 audiometers (Madsen Electronics, Denmark). Sound-proof rooms were set-up according to International Standards Organization protocol 8253-2. Bilateral hearing impairment was determined as the pure-tone average of audiometric hearing thresholds at 500, 1000, 2000, and 4000 Hz (PTA 0.5-4 kHz ) in the better ear, defining any hearing loss as PTA 0.5-4 kHz .25 dB HL; mild hearing loss as PTA 0.5-4 kHz .25-40 dB HL; and moderate to severe hearing loss as PTA 0.5-4 kHz .40 dB HL.
Assessment of Visual Impairment
Monocular distance logMAR (logarithm of the minimum angle of resolution) visual acuity was measured with forced-choice procedures using the retroilluminated chart with automatic calibration to 85 cd/m 2 (Vectorvision CSV-100 TM; Vectorvision Inc, Dayton, Ohio) according to the Early Treatment Diabetic Retinopathy Study protocol [23]. This was conducted with habitual correction (presenting visual acuity, with current eyeglasses, if worn) and after subjective refraction (best-corrected visual acuity). For each eye, visual acuity was recorded as the number of letters read correctly from 0 to 70. For the present study, any visual impairment was defined as presenting or bestcorrected visual acuity of the better eye less than 39 letters (,20/ 40). DSI was defined as concurrent visual (either presenting or best-corrected) and hearing impairment, as determined using the above definitions.
Information on Mortality
To identify and confirm persons who died after BMES-2, demographic information including surname, first and second names, sex and date of birth of the examined participants were cross-matched with Australian National Death Index (NDI) data for deaths, to December 2007. A probabilistic record linkage package was used, adopting a multiple pass procedure in which both data sets were grouped based on different characteristics (e.g., date of birth, name, sex) each time. Matches were divided into exact and non-exact. All non-exact matched records were examined manually and accepted if there was only one non-exact matched characteristic that was not critical. Information provided by family members during follow-up was also included if the participant was reported to have died on or before December 2007. The International Classification of Diseases, Ninth Revision (ICD-9) [24] and International Statistical Classification of Diseases, 10 th Revision (ICD-10) [25] cause of death codes were also obtained. The following cause-specific mortality codes were used in our analyses: coronary heart disease (ICD-9:410.0-9, 411.0-8, 412, 414.0-9, and ICD-10: I21.0-9, I22.0-9, I23.0-8, I24.0-9, I25.0-9); stroke (ICD-9:430.0-438.9 and ICD-10: I60.0-I69.9); or cancer (ICD-9:140-208). The primary cause of death (i.e. death from any cause, stroke, coronary heart disease or cancer) was used in statistical modeling. The validity of Australian NDI data has been reported to have high sensitivity and specificity (93.7% and 100.0%, respectively) for all-cause mortality [26]. The census cutoff point for mortality was December 2007 (10-year follow up).
Assessment of Covariates
A face-to-face interview with trained interviewers was conducted, and comprehensive data including information about medical history, hearing, demographic factors, socio-economic characteristics, lifestyle and health risk behaviour such as exercise, and smoking, were obtained from all participants. The medical history included cardiovascular or other systemic disease and associated risk factors, and medications used. A past history of angina, diabetes, myocardial infarction, and stroke was determined by responses to a question: ''Has a doctor advised you that you have any of the following conditions?'' Cognitive decline was assessed using the mini mental state examination (MMSE) questionnaire [27]. Participants with scores ,24 were considered cognitively impaired. Self-rated health was assessed by asking: ''For somebody your age, would you say your health is excellent, very good, good, fair, or poor?'' Low self-rated health was defined as fair or poor.
Classification of hypertension was based on the 2003 World Health Organization/International Society of Hypertension guidelines [28]. Participants were classified as having hypertension stage 1 if systolic blood pressure was 140 to 159 mm Hg or if diastolic blood pressure was 90 to 99 mm Hg. Participants were classified as having hypertension stage 2 if they were previously diagnosed with hypertension and were using antihypertensive medications, if systolic blood pressure was 160 mm Hg or greater, or if diastolic blood pressure was 100 mm Hg or greater at examination. Body mass index was calculated as weight in kilograms divided by height in meters squared. Disability in walking at baseline was assessed as present if the participant was observed by a trained examiner to have walking difficulties or used walking aids or a wheelchair.
Statistical Analysis
SAS statistical software (SAS Institute, Cary NC) version 9.1 was used for analyses. The association between single sensory impairment and DSI with mortality was examined using Cox regression models to estimate hazard ratios (HR) and 95% confidence intervals (CI). Multivariable regression models were first adjusted for age (entered as a continuous variable) and sex, and then further adjusted for confounders that were found to be significantly associated with mortality i.e. body mass index, systolic blood pressure, current smoking status, poor self-rated health, walking disability, presence of hypertension and/or diabetes, history of cancer, angina, stroke and/or acute myocardial infarction and cognitive impairment. We estimated the proportion surviving using the Kaplan Meier method. Kaplan-Meier survival curves are generated from the fitted Cox model using mean covariate values of age and sex. Table 1 shows the baseline characteristics of study participants with and without DSI. Those with DSI compared to those without DSI were more likely to be older, male, hypertensive and cognitively impaired, and also to have poorer self-rated health, walking disability, diabetes, stroke, angina, acute myocardial infarction, higher systolic BP and lower BMI (Table 1). We assessed the 10-year mortality risk in participants with presenting visual impairment (bilateral) and hearing impairment ( Table 2). Among participants with presenting visual impairment (better eye) and bilateral hearing impairment at baseline, 64% had died 10 years later compared to 11% without any sensory loss, and 26% and 30% with only a vision or hearing loss, respectively. After multivariable adjustment, participants with DSI (i.e. presenting visual impairment and any level of hearing impairment) had a 62% increased risk of dying 10 years later compared to those without any sensory impairment, HR 1.62 (95% CI 1. 16-2.26). This association was more marked in participants with both presenting visual impairment and moderate to severe hearing loss, multivariable-adjusted HR 1.84 (95% CI 1.19-2.86). Having a single sensory loss (i.e. either presenting visual impairment or any level of hearing loss) did not significantly increase the risk of mortality over 10 years. Figure 1 shows the 10-year survival of participants by the presence or absence of sensory loss. Persons with DSI had lower survival than those with a single sensory impairment (either presenting vision or hearing loss alone) or without any sensory impairment.
Results
Non-significant associations were observed between concurrent presenting visual impairment and hearing impairment at baseline with 10-year cause-specific mortality, after multivariable adjustment: coronary heart disease mortality -HR 1.47 (95% CI 0.83-2.58); stroke mortality -HR 1.05 (0.50-2.22); and cancer mortality -HR 1.79 (95% CI 0.99-3.23). We also assessed the risk of mortality in participants with bestcorrected visual impairment (bilateral) and hearing impairment ( Table 3). Among participants with best corrected visual impairment and bilateral hearing impairment at baseline, 76% had died 10 years later compared to 11% without any sensory loss, and 40% and 33% with only a vision or hearing loss, respectively. After multivariable adjustment, having both best corrected visual impairment and moderate to severe hearing loss increased the mortality risk by two-fold, when compared to those without any sensory loss. Having any or mild hearing loss (but without best corrected visual impairment) was associated with an increased risk of dying of 29% and 32%, respectively.
Discussion
To our best knowledge, this is the first population-based study to demonstrate that older adults with objectively measured DSI are at an increased risk of death from all causes compared to those without any sensory loss or a single sensory impairment. Specifically, participants with both presenting visual impairment (better eye) and bilateral hearing impairment at baseline had a 62% increased risk of dying 10 years later, independent of age, sex, self-rated health and the presence of known mortality markers. This association with mortality was more marked among older adults with concurrent moderate to severe hearing loss and any presenting or best-corrected vision loss.
Older adults in the BMES with clinically confirmed DSI compared to their counterparts without DSI had a 62% increased risk of total mortality. This finding is relatively similar to the HRs reported in a U. S. study: among men (HR 1. 23 [20]. Also, in the Canadian National Population Health Survey [29], selfreported hearing impairment but not vision impairment were significantly associated with a greater risk of total mortality 12 years later, among 12,375 participants aged 18 and older. In our study, baseline DSI conferred a greater risk of dying 10 years later compared to having only presenting visual impairment or hearing impairment alone. This result suggests a potential interactive effect of DSI on survival, that is, the negative effects of vision are multiplied by the effects from hearing loss [12]. Our finding also concur with previous reports suggesting that the presence of more than one sensory impairment increases morbidity relative to single sensory impairments [21,[30][31][32]. We documented a gradient effect from the severity of DSI on mortality risk. Specifically, participants with concurrent moderate to severe hearing loss (.40 dB HL) and any visual impairment (,20/40) had a higher risk of dying 10 years later, than those with mild hearing loss (,25-40 dB HL) and any vision loss (particularly those with best-corrected visual impairment). This finding concurs with the US study by Lee et al. [21] which showed that moderate to severe concurrent hearing and vision impairment in women significantly increased their risk of mortality. This is not surprising, as with increasing level of measured hearing impairment the likelihood of communication difficulties also increases [33], which in turn can cause increasing social isolation [34] and also a higher likelihood of experiencing functional disability [13], factors that could negatively impact on life expectancy [35,36].
It is hypothesized that the increased risk of mortality observed in persons with DSI is mediated by factors known to increase the risk of hearing and visual impairment in older adults (e.g., cardiovascular disease, hypertension and diabetes) [20,37,38]. However, in the BMES, a significant reduction in survival was observed in persons with DSI, even after controlling for cardiovascular risk factors (e.g., body mass index, smoking, hypertension) and events (e.g., angina, acute myocardial infarction), which supports findings from the study by Lam et al. [20] Moreover, we did not observe significant associations between DSI with 10-year coronary heart disease or stroke mortality. Therefore, these data together suggest that exposure to vascular risk factors are not likely to be the link between DSI and mortality risk, although, residual confounding from vascular risk factors cannot be fully discounted.
Alternatively, it is speculated that presence of DSI could be a marker for frailty (e.g., handgrip strength, peak expiratory flow), illness or possibly accelerated aging processes [21,39]. Given that frailty is a strong predictor of mortality [40], the association between decreased vision and hearing and increased risk of dying is not unexpected. Additionally, a decline in psychosocial functioning could underlie the relationship between DSI and mortality risk. Visual impairment was shown to be associated with difficulty in performing ADL [41], social isolation [42] and depression [43]. Similarly, impaired hearing has also shown to negatively impact on participation in social activities [44], functional independence [13] and mental health [8]. Other potential mechanisms could be that both hearing and vision impairments are associated with a greater risk of accident and injury [45][46][47][48], which in turn could underlie the greater risk of mortality among older adults with DSI. However, we are unable to confirm this hypothesis as there were only around ,1% of BMES participants who had died from injury (e.g. falls related) or accident, hence; we did not have sufficient study power to analyze these associations with DSI.
As sensory problems are common experiences within older age groups, they are often overlooked or dismissed [30]. Our study could have potential public health implications, as it suggests that identifying and targeting DSI in older adults could be a potentially useful strategy for preventing a decline in their life expectancy.
Specifically, regular assessment of presence of DSI in older persons could lead to earlier detection and facilitation of rehabilitation and therapy that could reduce the negative impacts of DSI. Correction of visual and hearing impairment could improve survival [42,49], hence, strategies including the provision of corrective lenses, hearing aids, assistive devices (e.g., magnifiers, listening devices) and rehabilitative services such as visual, auditory and communication training [21,50] should be encouraged and/or implemented by clinicians in order to promote both longevity and an improved quality of life in their older sensory impaired patient.
The strengths of this study include the use of a representative cohort with a relatively high participation rate, the use of standardized audiometric and vision testing, with measures of sensory function at more than one point in time [51], and ascertainment of mortality and its causes using validated Australian National Death Index data. We previously showed that the door-to-door census which was used as a sampling frame for the BMES is less likely to be subject to selection bias [52]. This evidence together with the high response rate means that our door-to-door census can be regarded as being close to a gold standard as can realistically be achieved [53]. Despite the high internal validity of the BMES, we need to caution about the generalizability of our study findings. The study population is not a random sample of the wider Australian population. However, the findings from this study could be reasonably generalized to the Australian population over the age of 55, as it was previously observed that the demographic characteristics of this sample population are not markedly different from the Australian population apart from a lower proportion of immigrants [52]. Second, while we had robust data on a range of confounders, other unmeasured or unknown factors (e.g., lifestyle or societal factors) could have influenced our study findings. Finally, age was entered into the multivariable model assuming a linear relationship with mortality risk, hence, we cannot disregard the possibility that some of the effects of aging are being picked up by hearing impairment, vision impairment and/or other independent variables (e.g. walking disability).
In summary, we found that the presence of DSI independently predicted an increased risk of mortality in older adults. These findings emphasize to clinicians the importance of recognizing that older adults with concurrent vision and hearing loss are at an increased risk of mortality compared to their non-impaired counterparts or those with only a single sensory impairment. Public health strategies to encourage earlier identification of older adults with DSI and their appropriate referral to rehabilitative services and support could improve life expectancy in this vulnerable population.
==
Domain: Psychology Biology Medicine
|
[{"added": "2018-03-21T18:01:26.031Z", "created": "2013-03-04T00:00:00.000", "id": "4008905", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0055054&type=printable", "pdf_hash": "5112f0897b64b252d1e9965edf52a551e584efb7", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:35539", "s2fieldsofstudy": ["Medicine", "Psychology"], "sha1": "8eada898b2efddf1eb9e5331387bdba20f22baa3", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Pharmacological Alternatives for the Treatment of Neurodegenerative Disorders: Wasp and Bee Venoms and Their Components as New Neuroactive Tools
Neurodegenerative diseases are relentlessly progressive, severely impacting affected patients, families and society as a whole. Increased life expectancy has made these diseases more common worldwide. Unfortunately, available drugs have insufficient therapeutic effects on many subtypes of these intractable diseases, and adverse effects hamper continued treatment. Wasp and bee venoms and their components are potential means of managing or reducing these effects and provide new alternatives for the control of neurodegenerative diseases. These venoms and their components are well-known and irrefutable sources of neuroprotectors or neuromodulators. In this respect, the present study reviews our current understanding of the mechanisms of action and future prospects regarding the use of new drugs derived from wasp and bee venom in the treatment of major neurodegenerative disorders, including Alzheimer’s Disease, Parkinson’s Disease, Epilepsy, Multiple Sclerosis and Amyotrophic Lateral Sclerosis.
Introduction
Insect venoms have been used by traditional Chinese and Korean medicine as well as ancient Egyptian and Greek civilizations since 1000-3000 BC to control a number of diseases, including neurological disorders [1][2][3]. Moreover, religious texts such as the Vedas, the Bible and the Koran report the use of bee products to treat diseases [3,4].
The diversity of biologically active molecules from animal venoms is well-known and has long garnered the interest of toxinologists. However, progress is more evident in recent years due to advances in the fields of proteomics, transcriptomics and genomics [5]. The area of venom-based drugs in particular has benefited from these advances along with high throughput screening techniques, which have accelerated the discovery of useful venom-derived drugs.
Bee and wasp venoms are known to be rich in neuroactive molecules that may be valuable in the development of new drugs or act as pharmacological tools to study the normal and pathological functioning of the nervous system [6,7]. As such, this review focuses on the main results obtained for the use of wasp and bee venoms in the treatment of the most prevalent neurodegenerative disorders. It is important to note that several of these compounds could become important new sources for the development of more effective medication with fewer adverse effects. The bioprospection of these compounds is vital since the drugs currently used to treat major neurological disorders (i.e., Epilepsy, Parkinson's Disease (PD) and Alzheimer's Disease (AD)) provide only symptomatic relief, and the incidence of serious adverse effects remains high [8][9][10][11].
The nervous system is an important target for these toxins, which can modulate synapses as well as generate and propagate action potentials by selectively acting on different ion channels and receptors [12]. Interestingly, evolution has fine-tuned venoms for optimal activity, providing us with a vast array of potential therapeutic drugs, which can be used to design pharmacological agents for the treatment of several diseases, including central nervous system (CNS) disorders [12,13] (Figure 1). Among these neurological disorders, neurodegenerative conditions significantly impact not only individuals, but also caregivers and society. The most prevalent neurodegenerative disease is AD, followed by PD and Epilepsy. Neurodegenerative diseases are a heterogeneous group with relentless progression, where aging is a major risk factor in the development [16]. Despite their heterogeneity, all of these diseases are characterized by cognitive impairment, motor alterations and personality changes. Unfortunately, the specific etiology of neuronal death and protein deposition in these diseases remains unknown [16,17].
Alzheimer's Disease and Other Dementias
Dementia is one of the most frequent causes of cognitive impairment in older adults, with forecasts indicating a worldwide increase from 25 million in 2000 to 115.4 million by 2050. Alzheimer's alone is responsible for over half of these cases [18][19][20].
Alzheimer's is symptomatically characterized by memory deficits, cognitive impairments and personality changes [17]. In general, the first clinical signs are impaired short-term memory accompanied by attention and verbal fluency difficulties. Other cognitive functions also deteriorate with the evolution of the disease, including the ability to make calculations, visual-spatial skills and the ability to use everyday objects and tools [17,20].
Estimates indicate the disease will affect more than 80 million people by 2040 and increased life expectancy will see the number of people with AD grow by 300% in developing countries. Since the disease is progressive, patients require prolonged special care after diagnosis, with annual costs estimated at nearly EUR 20,000 per person, exceeding that of patients with cancer [17].
Major contributors to neurodegeneration in brains affected by AD are the deposition of senile plaques, composed primarily of Aβ peptide, and neurofibrillary tangles formed largely by tau protein, which accumulate in neuropils from the cerebral cortex and hippocampus. Moreover, mitochondrial alterations such as fission-fusion abnormalities, defects in electron transport chain proteins, cytoskeletal abnormalities, calcium metabolism, intrinsic apoptosis pathways and caspase activation, as well as free radical generation are also involved in AD pathology [21,22].
More than 100 years after identifying the hallmark lesions in AD, there is still no minimally effective disease modifying therapy available [22]. From 2002 to 2012, of 221 agents submitted to trials for disease-modifying potential, none was different from the placebo in terms of positively affecting primary outcomes [23]. Alzheimer's treatment is symptomatic and relies on the administration of cholinesterase inhibitors (AChEI) (only tacrine, donepezil, rivastigmine and galantamine are currently approved for AD treatment) and NMDA receptor antagonists (only memantine is approved) [17]. Intervention with AChEI decreases acetylcholine metabolism and enhances neurotransmission, which is associated with memory and cognition reduction in AD [24]. NMDA antagonists act by compensating abnormal tonic activation by glutamate and are more efficient in moderate to severe stages of the disease [25]. Given that these drugs merely provide symptomatic relief, there is an urgent need to develop neuroprotective treatments for AD.
Parkinson's Disease
Parkinson's Disease is a universal, incurable, multifactorial and neurodegenerative disorder characterized by gradual degeneration and loss of dopaminergic neurons in the substantia nigra (SN). This leads to nigrostriatal pathway denervation, with the presence of Lewy body cytoplasmatic inclusions, predominantly resulting in motor symptomatology. In addition, non-motor symptoms are often identified in PD patients and may precede motor signs [26]. The disorder affects 1% of the population during the fifth or sixth decade of life and is primarily related to aging, with no definitive biomarker available for PD diagnosis [27,28].
Although PD etiology is not yet fully understood, it is possible that a large set of environmental and genetic factors in association with intrinsic neuronal vulnerability in the SN could be involved in the neuronal death typically observed in PD, primarily by inducing oxidative stress and mitochondrial dysfunction [29]. These factors include pesticide exposure, glutamate excitotoxicity, protein misfolding and aggregation, an imbalance in calcium homeostasis and neuroinflammation by microglial activation [30,31]. However, no drug has been clinically proven to modify disease progression, either by protecting surviving dopaminergic cells from degeneration or by restoring lost cells.
In this context, pharmacological treatment for PD remains focused on motor symptoms, mostly by restoring striatal dopamine levels through the administration of dopamine agonists. L-DOPA, a dopamine precursor, is the gold standard for this approach and is often associated with an inhibitor of peripheral degradation (carbidopa and benserazide). Despite its efficiency, long-term L-DOPA treatment is linked to side effects such as motor fluctuations (shorter duration of action) and dyskinesias (abnormal involuntary movements), both of which can significantly reduce quality of life in patients [32,33].
Epilepsy
Epilepsy is an enduring predisposition of the brain to generate epileptic seizures along with the neurobiological, cognitive, psychological and social consequences that the condition causes [34]. More recently it has been defined according to events such as the occurrence of at least two unprovoked (or reflex) seizures in a 24 h period, one unprovoked (or reflex) seizure with the likelihood of further similar seizures, or diagnosis of an epileptic syndrome [35].
Estimates suggest that approximately 65 million people of all ages may be affected by epilepsy [36] and that the majority face treatment problems due to pharmacoresistance to antiepileptic drug (AED) therapy [37,38]. AEDs are classified into three generations, according to their introduction into the market. The first generation of these drugs was sold in the USA and Europe from 1857 to 1958, followed by the second generation between 1960 and 1975. Drugs introduced in the 1960s are potent enzymatic inducers of cytochrome P450 that lead to clinically significant adverse drug interactions and hypersensitive reactions [39]. The 1980s saw the introduction of 15 additional AEDs (third generation), providing more appropriate drug alternatives for patients. However, it is important to underscore that each drug has its advantages and limitations, making treatment a difficult process [40]. Furthermore, these drugs are still inefficient in drug resistant epilepsy, challenging our understanding of the underlying mechanisms of this phenomenon and how to overcome or prevent them. Recent progress in understanding the molecular and cellular events that cause this disease have allowed better management of strategies for the discovery and development of more effective AEDs [41].
Multiple Sclerosis
Multiple sclerosis (MS) is a chronic inflammatory, demyelinating and neurodegenerative disorder of the CNS that begins in young adulthood and may be the result of the interaction between genetic and environmental factors, together with certain pathological hallmarks of an autoimmune disease [42][43][44]. According to the National Multiple Sclerosis Society, the disease affects around 2.1 million people worldwide [45]. MS has a significant socioeconomic impact that is comparable to other neurological conditions. This is because mean disease duration is approximately 38 years, thus affecting individuals at a time when they are entering, developing, or consolidating their professional careers [42].
The pathogenesis of MS is complex and only partially understood, hampering diagnosis and thus the choice of appropriate treatment. Nevertheless, a group of experts recently revised the MS phenotypic classification that includes the five MS subtypes: Relapsing-remitting MS (RRMS), clinically isolated syndrome (CIS), radiologically isolated syndrome (RIS), primary-progressive MS (PPMS) and secondary-progressive MS (SPMS) [46]. Considering the complexity of MS pathophysiology and diagnosis, only a brief description will be given of the main phenotypes included since MS classification began (RRMS, PPMS, and SPMS).
Relapsing-remitting multiple sclerosis (RRMS) represents about 80% of all cases, lasts for about 15 years and is characterized by acute exacerbations from which patients completely or partially recover, with periods of relative clinical stability in between [43,44]. When neurological function declines, the disease progresses to the following stage and is known as primary-progressive multiple sclerosis (PPMS). This type affects 10% of patients, who often present with progressive cerebellar syndrome and myelopathy, or other progressive symptoms [44,47]. Secondary-progressive multiple sclerosis (SPMS) is characterized by a progressive loss of motor function after an initial relapse, occurring about 20 years after the initial event [48]. Furthermore, RRMS is best characterized by an intense focal inflammatory component, whereas PPMS and SPMS exhibit more neurodegenerative features with concomitant chronic inflammation and axon loss [49].
Similar to other neurodegenerative disorders, the limitations of current therapies for MS include lack of superior treatment efficacy, serious adverse effects and long-term safety [43]. Significant advances in the treatment of RRMS are observed when the main goal is to target inflammation and modify the course of the disease; however, the same cannot be said about progressive forms of MS [47,50]. In addition, halting or reversing disease progression is only possible by using remyelinating and neuroprotecting agents, which does not occur in current treatments.
Amyotrophic Lateral Sclerosis
Amyotrophic lateral sclerosis (ALS) is a devastating, progressive and incurable adult-onset neurodegenerative disease characterized by the loss of upper and lower motor neurons in the primary motor cortex, brainstem and spinal cord. The disease affects motor functioning, resulting in paralysis and eventual death, typically from respiratory failure [51][52][53][54][55][56]. Average survival is 3 years after the first symptoms emerge and 5%-10% of patients survive beyond 10 years [57].
The worldwide average incidence rate for ASL is 2.1/100,000 person-years and a point prevalence of 5.4/100,000 persons, strongly linked to increased age [57]. Although little is known about the etiology of ALS, some studies indicate that 10% of cases are familial ALS and 85%-90% are classified as sporadic [53,58,59].
There is increasing evidence that patients with familial and sporadic forms of ALS exhibit signs of multi-modal dysfunction, even in early stages. Previous population-based studies estimated that around 35% of patients exhibit these impairments, including behavioral changes and executive and cognitive function deficits. Furthermore, about 15% of those affected with ALS may also suffer from frontotemporal dementia (ALS-FTD). This leads to reduced quality of life, caregiver stress, clinical effects from ventilator use and gastrostomies, negatively influencing survival time [51,55,[60][61][62].
The mechanisms responsible for disease onset and progression remain unknown, hindering the development of targeted therapies for ALS [59]. Given the multifaceted nature of the disease, most of the current approaches employed in clinical trials focus on the emerging concept of stem cell-based therapeutics [59,63]. Riluzole is the only Food and Drug Administration (FDA) approved treatment for ALS and prolongs survival by only a few months [59,64].
Venoms and Toxins from Wasps and Bees to Combat Neurodegenerative Disorders
The biological capacity to develop a secretion with highly specialized functions and a venomous apparatus is limited to certain groups, including cnidarians, some mollusk families, arthropods, certain reptiles and fish [12]. All insects that can sting are members of the order Hymenoptera, which includes ants, wasps and bees. The most extensively characterized venoms are bee venoms, mainly from the Apis genus, as well as some social and solitary wasp genera [4,65].
Bee Venom
Apitherapy is the medicinal therapeutic use of honeybee products, consisting of honey, propolis, royal jelly, pollen, beeswax and, in particular, bee venom (BV). Depending on the disease being treated, BV therapy can be used by applying a cream, liniment, or ointment, via injection, acupuncture or even directly through a live bee sting [4]. However, the most commonly used method is bee venom acupuncture (BVA), which involves the injection of diluted bee venom into acupuncture points. It can be employed as an alternative medicine in patients with PD, pain and other inflammatory diseases, such as rheumatoid arthritis and osteoarthritis [66][67][68].
Bee venom therapy is based on the fact that these crude extracts exhibit a wide variety of pharmacologically active molecules. This pool of chemical compounds is formed by biogenic amine, enzymes (phospholipase A2), basic peptides and proteins (melittin and apamin) and a mixture of water-soluble and nitrogen-containing substances [5].
One of the main biological activities identified in the venom of Apis mellifera, the most widely studied honeybee, is the inhibition of inflammatory and nociceptive responses [68]. Studies have shown that inhibition can occur in multiple aspects, making apitherapy the most common application for the treatment of inflammatory diseases such as arthritis, bursitis, tendinitis, rheumatoid arthritis and Lyme Disease [68,69].
In regard to anti-neuroinflammatory activity, crude honeybee venom and its components are important tools for the treatment of diseases accompanied by microglial activation [75,76]. Microglia are a population of macrophage cells in the brain that play an important role in immune defense and CNS tissue repair and are vital in controlling normal homeostatic functions in the brain [77].
Under pathogenic conditions, microglia are rapidly overactivated in response to neuronal injury and migrate to the affected sites of the CNS, significantly contributing to neuronal death in specific brain regions [78]. Resting microglia are generally benign to the brain; however, once activated through injury or during removal of unwanted cellular debris, they produce inflammatory cytokines, glutamate, quinolinic acid, superoxide radicals (O2 − ) and nitric oxide (NO), undermining cerebral homeostasis.
In this context, the suppression of microglial activation and the neuroprotective effect of BV were observed in several in vitro and in vivo studies, as well as in clinical trials. Studies in humans have shown that BV may be beneficial in the treatment of diseases that trigger cell death by microglial activation, particularly PD [79,80]. Parkinson's patients treated with BV acupuncture obtained promising results in idiopathic Parkinson's Disease Rating Scale Tests [79], demonstrating the remarkable ability of BV acupuncture (BVA) to interfere with PD progression. In vivo models for BVA and PD have also been tested. Bilateral acupoint stimulation of lower hind limbs prevented the loss of dopaminergic (DA) neurons in the striatum and SN for MPTP-induced PD (1-methyl-4-phenyl-1,2,4,5-tetrahydropyridine) and increased striatal dopamine levels [81][82][83]. MPTP mimics PD in rodents, involving the progressive loss of neurons in SN and causing behavioral alterations typical of PD, making it the most widely used model to study the disease. Chung and colleagues (2012) corroborated the results previously recorded for dopaminergic neuroprotection and observed a reduction in the infiltration of CD4T cells and microglial deactivation in an MPTP-induced PD mouse model [85]. In addition, BVA suppressed neuroinflammatory responses by MAC-1 and iNOS, microglial activation and loss of neurons in SN in the same mouse model [82]. It is important to note that the protective effect of BV on DA neurons of the SN is not restricted to acupoint stimulation, since it is also observed when using intraperitoneal injections [83].
Recently, an extensive and important study indicated that BV was capable of normalizing neuroinflammatory and apoptotic markers and restoring brain neurochemistry after simulated PD injury in mice [85], revealing the significant potential of BV application for PD therapy. Moreover, BV exhibited no signs of toxicity on general physiological functions when administered subcutaneously within a higher therapeutic range (100-200 fold) [90].
Bee venom has also been investigated in the treatment of MS and ALS (Table 1). In 2007 and 2008, two reviews summarized relevant findings regarding the therapeutic potential of venoms and other non-conventional approaches in MS treatment [80,96]. An interesting cross-sectional study involving 154 patients with MS investigated how often they used complementary and alternative medicine (CAM), including apitherapy [97]. The authors concluded that about 61% used CAM, and more than 90% of these used it as an adjunct to allopathic treatments. Furthermore, 65.8% of the interviewees reported an improvement. Given its importance and the growing interest in BV therapy, the American Apitherapy Society began to track patients who receive this treatment regularly, enrolling over 6000 members who take BV for MS or rheumatoid arthritis [98].
An FDA-approved investigational new drug trial involving nine patients with progressive MS evaluated the safety of BV [70]. Intradermal injections of gradually increasing doses were administered for 17 weeks until treatment reached 2.0 mg/week. A questionnaire, functional neurological tests and changes in measurement of somatosensory-evoked potentials were used to assess responses to therapy. None of the subjects displayed severe allergic reactions, although four reported worsening neurological symptoms and had to discontinue treatment. Two other patients showed objective improvement and three exhibited subjective symptom improvement. This was a preliminary study performed on a small number of patients and, despite the few positive results obtained, it was difficult to establish definitive conclusions regarding the efficacy of apitherapy.
In the same year, a high quality clinical trial for apitherapy in MS [99] evaluated the effectiveness of BV in 26 relapsing-remitting or secondary progressive MS patients [100]. This crossover study tested two groups; one received bee sting therapy for 24 weeks and placebo for another 24 weeks, while the other was given the same treatments in reverse order. Live bees were used to administer BV three times a week, with an increasing number of stings in each session to a maximum of 20 bee stings. Although it was well tolerated with no serious adverse events, the therapy failed to reduce fatigue, disease activity or disability, or improve quality of life. By contrast, phase II of the study assessed the efficacy of BV in patients with either RRMS or chronic progressive MS and found that BV intradermal injections decreased functional debilitation [101]. Treatment was administered until positive clinical effects reached a plateau, with an initial dose of one bee sting. In general, more than 68% of patients experienced some beneficial effects from BV therapy, including better balance, coordination, bladder and bowel control, as well as improved extremity strength, fatigue, endurance, spasticity and numbness, providing important evidence for the use of BV in MS. The authors attributed most of the positive findings to patients suffering from chronic-progressive MS when compared to relapsing-remitting MS, largely due to inherent variability among these MS patients, hindering result assessment.
A more recent study showed significant positive effects attributed to BV treatment in an experimental allergic encephalomyelitis animal model for MS induced by guinea pig spinal cord homogenate [86].
The results indicate that BV significantly decreases clinical symptoms and immunization effects in Lewis rats, as well as penetration of inflammatory cells and serum TNF-α and nitrate levels.
Considering all the findings reported on BV therapy for MS and according to Namaka and collaborators (2008), the different results reported to date may be due to the therapeutic protocols used, type of animal model and/or type of challenged cell line, in addition to potential time and dose-dependent properties [96].
Bee venom has also been studied for the treatment of ALS. A study using a symptomatic animal model for ALS with mutant hSOD1 G93A transgenic mice showed an improvement in motor activity in the rotarod test and prolonged life span for mice treated with BV acupoint stimulation [87]. The results obtained were substantiated by reduced levels of cytokines, typically released by activated microglia and astrocytes, leading to the neuroprotective effect observed. Moreover, by contributing to the reduction of motor neuron degeneration, BV prevented mitochondrial disruption and activated cell survival signal transduction pathways.
Research using the same animal model found that transgenic mice that received BV exhibited reduced expression of α-synuclein modifications, ubiquitinated α-synuclein and recovered spinal cord proteasomal activity [102]. It is important to underscore that animals received only two subcutaneous injections of 0.1 μg/g of BV at an acupoint, which was sufficient to induce positive effects.
Interestingly, another recently published study compared the effects of BV treatment using different administration routes for the same symptomatic model of ALS [88]. It was noted that BV treatment through an acupoint was more effective than intraperitoneal (i.p.) BV administration and acupoint stimulation alone. The results demonstrated an improvement in walking function, lower levels of neuroinflammatory proteins (TLR4, CD14 and TNF-α) in the spinal cord and reduced nuclear abnormality in the quadriceps femoris muscle.
In a study evaluating the ability of BV to act on the impaired ubiquitin-proteasome system [103], NSC34 motor neuronal cells expressing the mutant gene hSOD1 G85R were used and stimulated with 2.5 μg/mL of BV for 24 h. Once again the results showed restored proteasome activity and a reduction in the amount of misfolded SOD1. However, BV did not activate the autophagic pathway in these cells, a process frequently impaired in ALS that results in the aberrant accumulation of misfolded and/or aggregated proteins within spinal cord cells. This BV effect is remarkable because it reduces protein aggregation by targeting the ubiquitin system as opposed to activating the autophagy pathway.
Thus, when taken together, these findings reinforce the therapeutic potential of BV treatment, demonstrating an antineuroinflammatory effect, reduced neuronal loss caused by misfolded protein aggregates and glutamate neurotoxicity, restoration of the ubiquitin-proteasome system and motor improvement. These results could have important clinical implications for BV use as a coadjuvant treatment in both ALS and other neurodegenerative disorders.
Wasp Venom
With respect to wasps, important studies reveal the pharmacological potential of these venoms, present primarily in the Polybia genus ( Table 2). In 2005, Cunha and colleagues described the effects on rats of an intracerebroventricular (i.c.v.) injection of crude and denatured venom of the social wasp Polybia ignobilis [104]. Interestingly, crude venom provoked severe generalized tonic-clonic seizures, respiratory depression and death. On the other hand, denatured venom had an antiepileptic effect on acute seizures induced by i.c.v. injection of bicuculline, picrotoxin and kainic acid, but not on pentylenetetrazole(PTZ)-induced seizures. In addition, the denatured venom inhibited [ 3 H]-glutamate binding in membranes from the rat cerebral cortex at lower concentrations than those used for [ 3 H]-GABA binding [105]. These results indicate that specific components in the venom of P. ignobilis may interact with GABA and glutamate receptors, representing a significant source of neuroactive molecules (Figure 1). Similarly, i.c.v. administration of the denatured venom of P. occidentalis inhibited epileptic seizures caused by the same chemical convulsants previously described and was ineffective against PTZ-induced seizures [105]. A subsequent study with low molecular weight compounds (LMWC) from P. paulista wasps demonstrated their ability to block PTZ-induced seizures [106]. This effect is likely due to the presence of different compounds that act on GABA receptors.
Finally, research on crude venom from the social wasp Agelaia vicina revealed its ability to competitively inhibit high-and low-affinity GABA and glutamate uptake [109]. This is an important result since diseases such as Stroke, Epilepsy and PD involve abnormalities in GABA and glutamate uptake systems [110,111].
Compounds Isolated from Wasp and Bee Venom for the Treatment of Neurodegenerative Diseases
In addition to crude venom, several venom components have been widely used in Oriental medicine to relieve pain and treat inflammatory diseases such as rheumatoid arthritis and tendinitis [68,69,112]. Other potential venom-related treatments include the inhibition of neuroinflammatory responses, useful in the treatment of PD, AD and MS. This section of the review highlights the most recent and innovative therapeutic and biological applications of bee venom compounds: Melittin and Apamin (Table 1); and wasp venom compounds: Pompilidotoxins, Mastoparans, Kinins and Polyamine toxins ( Table 2).
Melittin
Melittin is the main component found in BV, accounting for 40% to 60% of dry venom, and is the best characterized peptide in BV. This linear peptide has 26 amino acid residues, alkaline characteristics, a predominantly hydrophobic N-terminal region and a hydrophilic C-terminal, resulting in amphiphilic properties [113] (Figure 2A). It appears to be primarily responsible for intense local pain, inflammation, itching and irritation in higher doses. On the other hand, in very small doses Melittin can cause a wide range of central and systemic effects, including anti-inflammatory effects, increased capillary permeability and lower blood pressure, among others [114].
The effect of Melittin on the CNS has been documented since 1973, when studies showed its marked effect on inhibiting general behavior, exploratory activity and "emotionality", in addition to disrupting spontaneous and evoked bioelectric activity in the brain. Moreover, high doses of this peptide can induce a depressant effect evaluated by electroencephalography in anesthetized cats. This effect was associated with reduced systemic blood pressure [115,116].
In 2011, Yang and collaborators studied the therapeutic effect of Melittin in a transgenic mouse model for ALS. In this model, Melittin-treated animals exhibited a decline in the number of activated microglia and expression of proinflammatory factor TNF-α, inhibiting the increased neuroinflammation responsible for neuronal death in this disease. Moreover, Melittin regulates the production of misfolded proteins by activating chaperones and alleviating α-synuclein post-translational modification, an important mechanism for PD and ALS pathologies. Melittin also restored proteasome activity in the brainstem and spinal cord. Interestingly, treatment with this alkaline peptide in a symptomatic ALS animal model improved motor function and reduced neuronal death [117].
Additionally, in vitro assays revealed the potential in Melittin as an agent for the prevention of neurodegenerative diseases, considering its ability to inhibit the apoptotic factor and cell death in neuroblastoma SH-SY5Y cells [118]. Melittin also demonstrated a potent suppressing effect on proinflammatory responses for BV2 microglia by reducing proinflammatory mediators and production of NO, PGE2 and cytokines [89]. Thus, it is suggested that this compound may have significant therapeutic potential for the treatment of neurodegenerative diseases accompanied by microglial activation, such as PD (Figure 1). (D) Beta-pompilidotoxin [122]; (E) Philanthotoxin [123]; (F) Bradykinin [124]; (G) Transportan [125].
Recently, Dantas and colleagues (2014) investigated the pharmacological effects of Melittin on the nervous system of mice [126]. The animals were submitted to behavioral tests, including the catalepsy test, open field and apomorphine rotation tests. The results showed that mice treated with Melittin displayed no cataleptic effects or changes in motor activity, although there was a reduction in the effects induced by the apomorphine test. As such, the authors found that Melittin exhibited antipsychotic properties and may be an alternative for the treatment of psychotic diseases, reducing the classic side effects caused by conventional neuroleptic drugs.
Apamin
Neurotoxin Apamin is the smallest peptide, accounting for less than 2% of BV, with 18 amino acids residues, a high cysteine content and alkalinity ( Figure 2B). Moreover, it is well known for its pharmacological property of irreversibly blocking Ca + activated K + channels (SK channels) and is considered the most widely used blocker for this type of channel [113,127].
Small-conductance Ca 2+ -activated K + (SK) channels control the firing frequency of neurons, especially at AMPA and NMDA glutamatergic synapses, and are responsible for hyperpolarization following action potentials [128]. These channels can be positively or negatively modulated. Positive modulation involves binding the compound, which then facilitates channel activity, thus impairing memory and learning. The opposite is true for negative modulation, where memory and learning improve and calcium channel sensibility declines [129]. Apamin acts through the second mechanism described. In neurons, this SK channel blockage decreases hyperpolarizing effects, modulating synaptic plasticity and memory encoding [130,131]. In addition, when compared to other arthropod neurotoxins, Apamin has an unusual ability to cross the blood brain barrier (BBB) and acts mainly in the CNS, where SK channels are extensively expressed [130].
Alvarez-Fischer et al. (2013) studied the protective effect of this peptide on dopaminergic neurons in a chronic mouse model of MPTP-induced PD [83]. The animals received i.p. injections in two different dosages of Apamin (low: 0.5 µg/kg; high: 1.0 µg/kg) in order to assess brain lesions and behavioral effects in mice. Results showed that Apamin protected nigral DA neurons and increased striatal DA levels in the nerve terminals. In the behavioral test, data were paradoxical, indicating that mice treated with Apamin spent significantly less time on the spindle in comparison to saline-treated animals with MPTP brain lesions, despite the authors' suggestion that Apamin may improve neuroprotection of dopaminergic neurons [83]. In this context, cell cultures that mimic the selective demise of mesenphalic dopaminergic neurons showed a lower degeneration rate after Apamin treatment [132]. Furthermore, Apamin has also been evaluated for the treatment of PD using the motor score from the Unified Parkinson's Disease Rating Scale. In this study, Apamin exhibited primarily neurorestorative activity in PD, as well as symptomatic and neuroprotective activity [133].
Several behavioral and electrophysiological studies have suggested Apamin in the treatment of AD, indicating that the blockage of SK channels by this compound may enhance neuronal excitability, synaptic plasticity, and long-term potentiation in the CA1 hippocampal region (Figure 1) [134]. Likewise, Apamin is a valuable tool in the investigation of physiological mechanisms involved in higher brain functions, such as cognitive processes or mood control, and there is already a patented method for early diagnosis of AD using Apamin [135][136][137][138][139]. However, it is important to underscore that SK blockage may accelerate neurodegenerative processes, making additional research in this field imperative.
Pompilidotoxins
Pompilidotoxins are a group of neuroactive molecules that were first described by Konno et al. [140,141]. They consist of two neurotoxins known as α-and β-pompilidotoxin (PMTX), derived from solitary wasps Anoplius samariensis and Batozonelus maculifrons, respectively. These molecules are peptides composed of 13 amino acid residues, differing solely in the presence of an amino acid at position 12, corresponding to lysine in α-PMTX and Arginine in β-PMTX ( Figure 2C,D, respectively). This minimal structural difference appears to be responsible for the significant potency of β-PMTX, approximately five times higher than α-PMTX, when tested in the lobster neuromuscular junction. Moreover, both peptides act on mammalian central neurons, primarily by blocking Na + current inactivation [142].
It has been demonstrated that α-PMTX interrupts synchronous firing of rat cortical neurons, facilitates synaptic transmission in hippocampal slices and decelerates the inactivation of tetrodotoxin-sensitive voltage-gated sodium channels (VGSCs) from rat trigeminal neurons [143,144]. In turn, β-PMTX modulated spontaneous rhythmic activity in spinal networks [145] and acted on hippocampal CA1 neurons by interfering with postsynaptic potential, increasing excitatory potential and interrupting rapid inhibitory potential [146]. Given that the main action of Pompilidotoxins is to slow the inactivation of VGSCs, these peptides may provide a better understanding of the molecular determinants associated with alterations in these channels involved in neuropathological conditions. The alteration of sodium channels has been described as a contributor to the events involved in several neurological disorders, especially persistent sodium currents that can participate in the physiopathology of some types of epilepsy and MS [147,148]. It is important to note that finely orchestrated activation and inactivation is essential for the correct maintenance of neuronal excitability and the slightest change in this equilibrium can result in serious consequences for the individual.
AvTx-7
Research by Pizzo et al. (2004) showed that neuroactive peptide Avtx7, isolated from the venom of social wasp Agelaia vicina, acted on the blockage of tetraethylammonium and 4-aminopyridine (4-AP)-sensitive K + channels ( Figure 1) [149]. As such, this novel neurotoxin may be a valuable tool in better understanding how K + channels work on neurological diseases, such as dementia and MS. These results were obtained using cortical brain synaptosomes and by assessing glutamate release as a response to different potassium blockers. K + channels are critically involved in the nervous system, consequently, alterations in their function can lead to important perturbations in membrane excitability and neuronal function. For instance, the dysfunction of a subfamily or subtype of K + channels might induce AD or PD [150]. Thus, K + channel blockade, for instance by 4-AP, has been linked to an action potential extension with a consequent increase in duration, which is relevant for the treatment of MS. Since 1990, the use of 4-AP in patients with MS has been described to reduce fatigue and improve visual field defects [151]. However, despite its therapeutic effects, drawbacks include low selectivity, causing severe adverse effects and difficulty determining individual therapeutic dose. In this respect, research targets more selective pharmaceuticals to treat MS by using these blockers, though with fewer side effects [152].
In regard to potassium blockers, an important line of research proposes their use as a meaningful non-dopaminergic alternative for the treatment of neurodegenerative diseases, such as advanced-stage PD. The use of these blockers is favorable in three mechanisms: Increased neurotransmitter release (i.e., glutamate), modulation of neuronal network oscillation and greater cortical excitation. In relation to 4-AP, advanced clinical trials have shown satisfactory results, leading to FDA approval in 2013 for the treatment of movement dysfunction in patients with MS [153]. In this field, the discovery and identification of AvTx7 provides new pharmacological options, since its mechanism seems to be related to 4-AP.
Mastoparan
Mastoparan is a class of multifunctional peptides found in solitary and social wasp venom, with its primary activity described in mast cell degranulation, giving the peptide its name [154]. Thus, these peptides exhibit a number of remarkable pharmacological activities, such as antimicrobial, antitumor, insulinotropic and neurological effects [114,[155][156][157][158][159].
The first Mastoparan was identified and chemically characterized by Hirai et al. in 1979, when this molecule was isolated from the social wasp Vespula lewisii. Mastoparans are short cationic peptides with 10 to 14 amino acid residues, two to four lysine residues and C-terminal amidation, characteristics that are essential for proper peptide action [160,161]. These peptides can interact and penetrate biological membranes via the positively charged side-chains of their amphipathic α-helical structures [161]. In light of this property, Mastoparans were recently classified as cell-penetrating peptides (CPP) [162].
Crossing the BBB is a significant challenge in neuropharmacology. The BBB is responsible for regulating brain homeostasis through selective permeability that protects the CNS. However, these characteristics also affect drug delivery and bioavailability to the CNS. Advances in the fields of pharmacokinetics, molecular biology, nanotechnology and toxinology have resulted in strategies to facilitate the crossing of drugs through the BBB, thus, increasing drug concentration in the brain [163]. Cell permeable peptides (CPP), particularly Mastoparans, serve as vehicles for the delivery of different molecules and particles into the brain and neurons and have been studied in combination with compounds that act on the CNS [164].
With the aim of enabling neuroactive compounds to permeate the BBB, researchers have created new chimeric peptides (Transportan), connecting Mastoparans and the neuropeptide Galanin in two different ways. The first compound, named Transportan, is formed by 12 residues of Galanin and a full length Mastoparan connected by a lysine, resulting in a chimera with 27 residues [164] ( Figure 2G). The second compound, called Transportan 10, consists of seven terminal residues of Galanin and a full Mastoparan connected by a lysine residue [165].
Galanin, discovered in 1983, is a neuropeptide that in humans contains 30 amino acid residues and 29 in other species, for revision see [166]. Its name originates from the fusion of Glycin and Alanin, the N-terminal and C-terminal amino acids, respectively. Widely distributed in the peripheral and central nervous systems, Galanin has been associated with the pathophysiology of neurodegenerative diseases such as AD and Epilepsy [166]. Several studies report that the overexpression of Galanin detected in AD can preserve cholinergic striatal neuron function, which in turn may slow AD symptoms [167]. The chimeric construction of Transportan and Transportan 10 has been used as a drug delivery system for Galanin in the CNS and as treatment for neurodegenerative diseases, acting as a neuroprotective agent ( Figure 1).
Another important function of Mastoparans is that they act as an antidote to one of the most powerful neurotoxins in the world, Botulinum toxin A (BoTx-A). If inhaled, only one gram of crystallized BoTx-A dispersed in the air can kill a million people [168]. Intoxication is so rapid and severe that some countries developed biological weapons containing BoTx for use in World War II. Intoxicated patients are treated with serum therapy. However, this does not reverse the toxic effects already induced in the organism [169]. As such, in an effort to treat this intoxication, a group of researchers employed Mastoparan 7 as a CPP in a chimeric construction denominated Drug Delivery Vehicle-Mas 7 (DDV-Mas 7). Consisting of a non-toxic heavy chain fragment of BoTx-A and Mastoparan 7, this chimeric peptide induced neurotransmitter release in a culture of mice spinal cord neurons, reversing the effect of the BoTx-A and allowing Ach liberation, followed by muscular contraction [160].
Mastoparans also modulate G-protein activity without receptor interaction, currently considered a preeminent tool for the study and understanding of this complex intracellular signaling system [170][171][172][173]. Several neurological disorders, including Mood Disorders, Epilepsy, AD, and PD are related to G protein-coupled receptors [174][175][176]. Thus, over the last decade, natural, modified or chimeric Mastoparans have been used as a potential treatment for a number of neurological conditions.
Wasp Kinin
Another class of peptide frequently encountered in wasp venom is Kinin, composed of Bradykinin (BK) and its analogues, largely responsible for the pain caused after a wasp sting and the paralyzing action used for prey capture [177][178][179]. Naturally present in different animals, BK was first described in 1949 by Rocha and Silva as consisting of nine amino acid residues ( Figure 2F), with its primary activity described in mammal platelets [180]. This small peptide plays an important role in controlling blood pressure, renal and cardiac function, and inflammation [181]. It is important to note that Kinin was the first neurotoxin component isolated from wasp venom. In addition, Kinin acts on the insect CNS, where it irreversibly blocks the synaptic transmission of nicotinic acetylcholine receptors [179][180][181][182]. Furthermore, Kinin components, produced via the kallikrein-kinin system, have been found in abundance throughout both the rat and human CNS attracting interest in neuroprotective research [183] (Figure 1). Two major Kinin receptor families have been identified: B2 and B1 receptors. Their expression is low under normal conditions, but is up-regulated following injury, infection and inflammation [184].
Although several studies report that BK likely triggers a specific cascade of inflammatory events in the CNS, it has also been shown to possess anti-inflammatory (neuroprotective) properties, suppressing the release of inflammatory cytokines (TNF-α and IL-1β) from microglia in in vitro assays [183]. According to these authors, BK modulated microglial function by negative feedback for cytokine production, increasing prostaglandin synthesis and causing greater microglial cAMP production [183].
BK can also be beneficial after ischemic stroke, particularly if administered in the latter stages as opposed to the initial phases, where its harmful effects include inflammatory response and neurogenic inflammation [185]. It is noteworthy that molecular and functional evidence has suggested that interaction with B1 receptors may provide a new therapeutic approach in MS, primarily by reducing the infiltration of immune cells (lymphocytes T) into the brain [184]. Additionally, treatment with BK applied two days after transient forebrain ischemia in rats in post-conditioning studies provided 97% neuroprotection for the particularly vulnerable CA1 hippocampal neurons, as well as a decrease in Caspase3 expression and iNOS-positive cells, and also a suppression in the release of cytosolic cytochrome c and MnSOD [107,108]. This indicates that the neuroprotective mechanism initiated by BK may also inhibit the mitochondria-mediated apoptotic pathway [108]. The neuroprotective role of BK has also been reinforced by evidence of its action in the retina, protecting against neuronal loss induced by glutamatergic toxicity. This BK-induced protection caused a downstream reaction in NO generation and an upstream reaction in radical oxygen generation [186].
As observed, BK agonists may provide a new platform for drugs designed to treat neurodegenerative disorders that involve microglial activation, such as PD and acute brain damage. In this respect, wasp venom contains a multitude of Kinins with different activity potency profiles. A good example is Thr 6 -Bradykinin, a compound isolated from several wasp venom samples. The single substitution of serine for threonine in this compound results in enhanced action when compared to BK. According to , this peptide displays remarkable anti-nociceptive effects when injected directly into the rat CNS; it is approximately three times more potent and remains active longer than BK [187]. These results can be explained by a more stable conformation in its secondary structure and/or the modification may protect against hydrolysis through neuronal kininases, preserving the effect of the peptide on B2 receptors [187,188].
Polyamine Toxins as Therapeutic Sources
Polyamine toxins are a group of low molecular weight (<1 kDa), non-oligomeric compounds isolated primarily from the venom of wasps, followed by spider venoms [189,190] (Figure 2E). The first polyamine toxin described, Philanthotoxin-433 (PhTX-433), was isolated from the venom of the wasp Philanthus triangulum [191]. These small natural molecules exhibit a number of biological activities and have been used as tools in the study of ionotropic glutamate (iGLU; AMPA) and nicotinic acetylcholine (nACh) receptors since the 1980s [190,192,193]. Interest is centered on its action as a non-selective and potent antagonist of glutamate receptors in the invertebrate and vertebrate nervous system ( Figure 1) [192][193][194]. Moreover, it is believed that the abnormal activation of iGLU receptors is involved in neurological and psychiatric diseases such as AD, PD, Stroke, Depression, Epilepsy, Neuropathic Pain and Schizophrenia [195,196].
With respect to iGLU, current polyamine toxins (PhTXs) and their derivatives have the ability to differentiate which AMPA receptors are in fact permeable to Ca 2+ ion, acting as a non-selective open-channel blocker [190,197]. As a result, PhTXs can control the excessive opening of overactivated ion channels (due to pathological conditions) and block the exaggerated influx of calcium, culminating in neuroprotection [193,198]. Interestingly, this mechanism of action is similar to that of Memantine, a drug used in the symptomatic treatment of moderate to severe AD [199]. Thus, the existence of a drug that has obtained good clinical results and its similarity with polyamine toxins illustrates the potentially promising role of these molecules and highlights the need for further research.
Recently, a computational model approach was devised to better understand how polyamine toxins interact with ion channels coupled with glutamate receptors [200]. This study found that these molecules could bind to the narrowest central region of the ion channel and block local ion flow. Membrane potential is important in toxin-receptor interaction, and as such, polyamine toxins are generally highly voltage-dependent blockers of iGLU [200]. In this regard, Nørager et al. recently developed fluorescent templates using polyamine toxin analogues to visualize these ligands in iGLU of living tissue [201].
Conclusions
Due to the rising prevalence of neurodegenerative diseases among the elderly, there is a pressing need for better treatment to alleviate the social and financial burden of these disorders. There are multiple targets for treating neurodegenerative diseases, considered complex syndromes that are difficult to control in a stable and lasting manner. Effective treatment of these diseases may require that the different pathogenic events associated with neurodegenerative diseases, such as the clearance of disaggregated proteins targeted in conjunction with neuroprotective and immunomodulatory strategies. In this respect, therapy using bee and wasp venoms is considered a psychoneurological approach for autoimmune and neurodegenerative diseases. Since these venoms contain a number of compounds, mainly peptides, advances in modern identification and sequencing techniques have facilitated and subsidized the elucidation of their full composition, thus providing an arsenal of new possibilities to combat a series of neurodegenerative diseases, using different neuroactive mechanisms of action.
==
Domain: Psychology Biology Medicine
|
[{"added": "2015-09-18T23:22:04.000Z", "created": "2015-08-01T00:00:00.000", "id": "15834889", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/2072-6651/7/8/3179/pdf", "pdf_hash": "a2a6161ac4468d068692a56ca5eb3741ec4ef778", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:31491", "s2fieldsofstudy": ["Medicine", "Psychology", "Biology"], "sha1": "a2a6161ac4468d068692a56ca5eb3741ec4ef778", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Temporal integration is a robust feature of perceptual decisions
Making informed decisions in noisy environments requires integrating sensory information over time. However, recent work has suggested that it may be difficult to determine whether an animal’s decision-making strategy relies on evidence integration or not. In particular, strategies based on extrema-detection or random snapshots of the evidence stream may be difficult or even impossible to distinguish from classic evidence integration. Moreover, such non-integration strategies might be surprisingly common in experiments that aimed to study decisions based on integration. To determine whether temporal integration is central to perceptual decision making, we developed a new model-based approach for comparing temporal integration against alternative “non-integration” strategies for tasks in which the sensory signal is composed of discrete stimulus samples. We applied these methods to behavioral data from monkeys, rats, and humans performing a variety of sensory decision-making tasks. In all species and tasks, we found converging evidence in favor of temporal integration. First, in all observers across studies, the integration model better accounted for standard behavioral statistics such as psychometric curves and psychophysical kernels. Second, we found that sensory samples with large evidence do not contribute disproportionately to subject choices, as predicted by an extrema-detection strategy. Finally, we provide a direct confirmation of temporal integration by showing that the sum of both early and late evidence contributed to observer decisions. Overall, our results provide experimental evidence suggesting that temporal integration is an ubiquitous feature in mammalian perceptual decision-making. Our study also highlights the benefits of using experimental paradigms where the temporal stream of sensory evidence is controlled explicitly by the experimenter, and known precisely by the analyst, to characterize the temporal properties of the decision process.
Making informed decisions in noisy environments requires integrating sensory information 16 over time. However, recent work has suggested that it may be difficult to determine whether 17 an animal's decision-making strategy relies on evidence integration or not. In particular, 18 strategies based on extrema-detection or random snapshots of the evidence stream may be 19 difficult or even impossible to distinguish from classic evidence integration. Moreover, such 20 non-integration strategies might be surprisingly common in experiments that aimed to study 21 decisions based on integration. To determine whether temporal integration is central to 22 perceptual decision making, we developed a new model-based approach for comparing 23 temporal integration against alternative "non-integration" strategies for tasks in which the 24 sensory signal is composed of discrete stimulus samples. We applied these methods to 25 behavioral data from monkeys, rats, and humans performing a variety of sensory decision-26 making tasks. In all species and tasks, we found converging evidence in favor of temporal 27 integration. First, in all observers across studies, the integration model better accounted for 28 standard behavioral statistics such as psychometric curves and psychophysical kernels. 29 Second, we found that sensory samples with large evidence do not contribute 30 disproportionately to subject choices, as predicted by an extrema-detection strategy. Finally, 31 we provide a direct confirmation of temporal integration by showing that the sum of both early 32 and late evidence contributed to observer decisions. Overall, our results provide experimental 33 evidence suggesting that temporal integration is an ubiquitous feature in mammalian 34 perceptual decision-making. Our study also highlights the benefits of using experimental 35 paradigms where the temporal stream of sensory evidence is controlled explicitly by the 36 experimenter, and known precisely by the analyst, to characterize the temporal properties of 37 the decision process. 38
INTRODUCTION 39
Perceptual decision-making is thought to rely on the temporal integration of noisy sensory 40 information on a timescale of hundreds of milliseconds to seconds. Temporal integration 41 corresponds to summing over time the evidence provided by each new sensory stimulus, and 42 optimizes perceptual judgments in face of noise (Bogacz et al. 2006; Gold and Shadlen 2007). 43 A perceptual decision can then be made on the basis of this accumulated evidence, either as 44 some threshold on accumulated evidence is reached, or if some internal or external cue 45 signals the need to initiate a response. 46 Although many behavioral and neural results are consistent with this integration framework, 47 temporal integration is a feature that has often been taken for granted rather than explicitly 48 tested. Recently, the claim that standard perceptual decision-making tasks rely on (or even 49 frequently elicit) temporal integration has been challenged by theoretical results showing that 50 non-integration strategies can produce behavior that carries superficial signatures of temporal 51 integration (Stine et al. 2020). These signatures include the relationship between stimulus 52 difficulty, stimulus duration and behavioral accuracy, the precise temporal weighting of 53 sensory information on the decisions, and the patterns of reaction times. 54 Here, we propose new analytical tools for directly assessing integration and non-integration 55 strategies from fixed-duration or variable-duration paradigms where, critically, the 56 experimenter controls the fluctuations in perceptual evidence over time within each trial 57 (discrete-sample stimulus, or DSS). By leveraging these controlled fluctuations, our methods 58 allow us to make direct comparisons between integration and non-integration strategies. We 59 apply these tools to assess temporal integration in data from monkeys, humans and rats that 60 performed a variety of perceptual decision-making tasks with DSS. Applying these analyses 61 to these behavioral datasets yields strong evidence that perceptual decision-making tasks in 62 all three species rely on temporal integration. Temporal integration, a critical element of many 63 major theories of perception at both the neural and behavioral levels, is indeed a robust and 64 pervasive aspect of mammalian behavior. Our results also illuminate the power of targeted 65 stimulus design and statistical analysis to test specific features of behavior. primates and rodents. Here we focus on experiments in which observers report their choice at 80 the end of a period whose duration is controlled by the experimenter ( Tasks using the DSS paradigm are classically thought to rely on sequential accumulation of 97 the stimulus evidence (Bogacz et al. 2006), which we refer to here as temporal integration. 98 Figure 1A shows an example stimulus sequence composed of n samples that provide differing 99 amounts of evidence in favor of one alternative vs. another ("A" vs. "B"). The accumulated 100 evidence fluctuates as new samples are integrated and finishes at a positive value indicating 101 overall evidence for stimulus category A ( Figure 1B). This integration process can be 102 formalized by defining the the decision variable or accumulated evidence and its updating 103 dynamics across stimulus samples: evidence. More specifically, the observer commits to a decision based on the first sample i in 127 the stimulus sequence that exceeds one of the two symmetrical thresholds, i.e. such that 128 | | ≥ . In our example stimulus, the first sample that reaches this threshold in evidence 129 space is the fifth sample, which points towards stimulus category B, so response B is selected 130 ( Figure 1C). This policy can be viewed as a memory-less decision process with sticky bounds. 131 If the stimulus sequence contains no extreme samples, so that neither threshold is reached, 132 the observer selects a response at random. (Following (Stine et al. 2020), we also explored 133 an alternative mechanism where in such cases the response is based on the last sample in 134 the sequence). 135 The second non-integration model corresponds to the snapshot model (Stine et al. 2020;Pinto 136 et al. 2018). In this model, the observer attends to only one sample i within the stimulus 137 sequence, and makes a decision based solely on the evidence from the attended sample: = 138 if > 0, and = if < 0. The position in the sequence of the attended sample is 139 randomly selected on each trial. In our example, the fourth sample is randomly selected, and 140 since it contains evidence towards stimulus category A, response A is selected ( Figure 1D). 141 We considered variants of this model that gave it additional flexibility, including: allowing the 142 prior probability over the attended sample to depend on its position in the sequence using a 143 non-parametric probability mass function estimated from the data; allowing for deterministic 144 vs. probabilistic decision-making rule based on the attended evidence; including attentional 145 lapses that were either fixed to 0.02 (split equally between leftward and rightward responses) 146 or estimated from behavioral data. We finally considered a variant of the snapshot where the 147 decision was made based on a sub-sequence of K consecutive samples within the main 148 stimulus sequence (1 ≤ < ), rather than based on a single sample. 149 150
Standard behavioral statistics favor integration accounts of pulse-based motion 151 perception in primates 152
To compare the three decision-making models defined above (i.e., temporal integration, 153 extrema-detection, snapshots), we first examined behavioral data from two monkeys 154 performing a fixed-duration motion integration task (Yates et al. 2017). In this experiment, 155 each stimulus was composed of a sequence of 7 motion samples of 150 ms each where the 156 motion strength towards left or right was manipulated independently for each sample. At the 157 end of the stimulus sequence, monkeys reported with a saccade whether the overall sequence 158 contained more motion towards the left or right direction. The animals performed 72137 and 159 33416 trials for monkey N and monkey P respectively, allowing for in-depth dissection of their 160 response patterns. 161 We fit the three models (and their variants) to the responses for each animal individually (see 162 Supplementary the accuracy and psychometric curves were accurately captured by the integration model. In 185 line with Stine and colleagues, we also found that both non-integration models could 186 reproduce the shape of the psychometric curve in monkey N, although the quantitative fit was 187 always better for the integration than non-integration models. By contrast both non-integration 188 models failed to capture the psychometric curve for monkey P ( Figure 2B, bottom row). More 189 systematically, the overall accuracy, which is an aggregate measure of the psychometric 190 curve, clearly differs between models, as the accuracy of the non-integration models 191 systematically deviated from animal data for both animals ( Figure 2C). In other words, all 192 models produce the same type of psychometric curves up to a scaling factor, and this scaling 193 factor (directly linked to the model accuracy) is key to differentiate model fits. better-than-observed accuracy for certain parameter ranges, but these are not the parameters 203 found by the maximum likelihood procedure, probably because they produce a pattern of 204 errors that is inconsistent with the observed pattern of errors. This indicates an inability of the 205 models to match the pattern of errors of the animal (see Discussion). 206 Finally, we assessed quantitatively which model provided the best fit, while correcting for 207 model complexity using the Akaike Information Criterion (AIC, Figure 2A). In both monkeys, 208 AIC favored the integration model over the two non-integration models by a very large margin. 209 We also explored whether variants of the extrema-detection and snapshot models could 210 provide a better match to the behavioral metrics considered above (Supp Figure 2 & 3). We 211 found using the AIC metric that the integration model was preferred over all variants of both 212 non-integration models, for both monkeys. Moreover, these model variants could not replicate 213 the psychophysical kernels as well as the integration model did (Supp Figure 2 & 3). In 214 conclusion, while psychometric curves may not always discriminate between integration and 215 non-integration strategies, other metrics including psychophysical kernels, predicted accuracy 216 and quality of fit (AIC) support temporal integration in monkey perceptual decisions. For one 217 model in one monkey (the snapshot model in monkey P), even the simple metric of overall 218 accuracy compellingly supported temporal integration (Fig. 2C). For the other monkey and/or 219 model, where the distinction was less clear, our model-based approach allowed us to leverage 220 these other metrics to reveal strong support for the temporal integration model ( Fig. 2A-C). 221 Although these data relies only on two experimental subjects, we show below further evidence 222 supporting the integration model in humans and rats. 223 and identify how non-integration models failed to capture them. We started by designing two 246 analyses aimed at testing whether choices were consistent with the extrema-detection model, 247 namely by testing whether choices were strongly correlated with the largest-evidence 248 samples. In the first analysis, we looked at the subset of trials where the evidence provided 249 by the largest-evidence sample in the sequence was at odds with the total evidence in the 250 sequence: we show one example in Figure 3B, where the largest evidence sample points 251 towards response B, while the overall evidence points towards response A. These 'disagree 252 trials' represent a substantial minority of the whole dataset: 1865 trials (2.6%) in monkey N, 253 1831 trials (5.5%) in monkey P. If integration is present, the response of the animal should in 254 general be aligned with the total evidence from the sequence ( Figure 3A, red bars). By 255 contrast, if it followed the extrema-detection model ( Figure 1C), it should in general follow the 256 largest evidence sample ( Figure 3A, green bars). In both monkeys, animal choices were more 257 often than not aligned with the integrated evidence ( Figure Cumming 2007). The extrema-detection model predicts that, in principle, samples whose 270 evidence is below the threshold have little impact on the decision, while samples whose 271 evidence is above the threshold have full impact on the decision. By contrast, the integration 272 model predicts that subjective weight should grow linearly with sample evidence. We 273 estimated subjective weights from monkey choices using a regression method similar in spirit 274 to previous methods (Yang and Shadlen 2007; Waskom and Kiani 2018), taking the form 275 Here f is a function that captures the subjective weight of 276 the sample as a function of its associated evidence. Whereas previous methods estimated 277 subjective weights assuming a uniform psychophysical kernel, our method estimated 278 simultaneously subjective weights ( ) and the psychophysical kernel , thus removing 279 potential estimation biases due to unequal weighting of sample evidence (see Methods). In 280 both monkeys, we indeed found that the subjective weight depends linearly on sample 281 evidence for low to median values of sample evidence (motion pulse lower than 6), in 282 agreement with the integration model (Supp. Figure 4). Surprisingly however, simulated data 283 of the extrema-detection model displayed the same linear pattern for low to median values of 284 sample evidence. We realized this was due to the very high estimated sensory noise (Supp 285 Fig 1), such that, according to the model, even samples with minimal sample evidence were 286 likely to reach the extrema-detection threshold. In other words, unlike the previous analyses, 287 inferring the subjective weights used by animals was inconclusive as to whether animals 288 deployed the extrema-detection strategy. This somewhat surprising dependency reinforces 289 the importance of validating intuitions by fitting and simulating models (
293
Example of an 'agree trial' where the total stimulus evidence (accumulated over samples) and the 294 evidence from the largest-evidence sample point towards the same response (here, response A). In 295 this case, we expect that temporal integration (TI) and extrema-detection (ED) will produce similar 296 responses (here, A). B. Example of a 'disagree trial', where the total stimulus evidence and evidence 297 from the largest-evidence sample point towards opposite responses (here A for the former; B for the 298 latter). In this case, we expect that integration and extrema-detection models will produce opposite 299 responses. C. Proportion of choices out of all disagree trials aligned with total evidence, for animal 300 (black bars), integration (red) and extrema-detection model (green).
302
Choice dependence on early and late stimulus evidence show direct evidence for 303 temporal integration 304 Following model comparisons favoring integration over both snapshot and extrema-detection 305 models, the immediately previous analysis relied on a special subset of trials to provide an 306 additional, and perhaps more intuitive, signature of integration, which ruled out extrema-307 detection as a possible strategy of either monkey. We next employed another novel analysis 308 specifically designed to tease apart unique signatures of the integration and snapshot models. 309 More specifically, we tested whether decisions were based on the information from only one 310 part of the sequence, as predicted by the snapshot model, or from the full sequence, as 311 predicted by the integration model. To facilitate the analysis, we defined early evidence Et by 312 grouping evidence from the first three samples in the sequence, and late evidence Lt, as the 313 grouped evidence from the last four samples. We then displayed the proportion of rightward 314 responses as a function of both early and late evidence in a graphical representation that we 315 call integration map ( Figure 4A). A pure integration strategy corresponds to summing early 316 and late evidence equally, which can be formalized as ( ) = ( + ), where is a 317 sigmoidal function. Because this only depends on the sum + , the probability of response 318 is invariant to changes in the ( , ) space along the diagonal, which leaves the sum 319 unchanged. These diagonals correspond to isolines of the integration map ( Figure 4A, left; 320 Supp Figure 5A). In other words, straight diagonal isolines in the integration map reflect the 321 fact that the decision only depends on the sum of evidence + . Straight isolines thus 322 constitute a specific signature of evidence integration. 323 We contrasted this integration map with the one obtained from a non-integration strategy 324 ( Figure 4A middle panel; Supp Figure 5A). There we assumed that the decision depends either 325 on the early evidence or on the late evidence, as in the snapshot model, with equal probability. 326 This can be formalized as ( ) = 0.5 ( ) + 0.5 ( ). In this case, if late evidence is null 327 ( ( ) = 0.5) and early evidence is very strong toward the right ( ( ) ≃ 1) the overall 328 probability for rightward response is By contrast, the lapse parameters showed no consistent relationship with late evidence 364 ( Figure 4E, right panel). Finally, we directly assessed the similarities between the integration 365 maps from monkey responses and from simulated responses for the three models (integration, 366 snapshot, extrema-detection). The model-data correlation was larger in the integration model 367 than in the non-integration strategies for both monkeys ( Figure 4E; unpaired t-test on 368 bootstrapped r values: p<0.001 for each animal and comparison against extrema-detection 369 and against snapshot model). Overall, integration maps allow to dissect how early and late 370 parts of the stimulus sequence are combined to produce a behavioral response. In both 371 monkeys, these maps carried signatures of temporal integration. For monkey P, the integration 372 model and the data look very similar. For monkey N, there is still a qualitative dependency that 373 deviates from non-integration, but which is not as uniquely matched to the integration strategy 374 (although the imperfect coverage of the two-dimensional space impedes further investigations). 375 Thus, complementing the statistical model tests favoring integration, this richer visualization 376 allows the data to show us that some degree of integration is occurring, albeit not perfect.
Temporal integration in human visual orientation judgments 397
Overall, all our analyses converged to support the idea that monkey decisions in a fixed-398 duration motion discrimination task relied on temporal integration. We explored whether the 399 same results would hold for two other species and perceptual paradigms. We first analyzed 400 the behavioral responses from 9 human subjects performing a variable-duration orientation 401 discrimination task (Cheadle et al. 2014). In each trial, a sequence of 5 to 10 gratings with a 402 certain orientation were shown to the subject, and the subject had to report whether they 403 thought the gratings were overall mostly aligned to the left or to the right diagonal. In this task, 404 the experimenter can control the evidence provided by each sample by adjusting the 405 orientation of the grating. We performed the same analyses on the participant responses than 406 on monkey data. As for monkeys, we found that the integration model nicely captured 407 psychometric curves, participant accuracy and psychophysical kernels ( Figure 5A-C, red 408 curves and symbols). By contrast, both non-integration models failed to capture these patterns 409 ( Figure 5A-C, blue and green curves and symbols). The accuracy from both models 410 consistently underestimated participant performance: 8 and 6 out of 9 subjects outperformed 411 the maximum performance for the snapshot and extrema-detection models, respectively 412 (Supp. Figure 7). This suggests that human participants achieved such accuracy by integrating 413 sensory evidence over successive samples. Moreover, subjects overall weighted more later 414 samples ( Figure 5C), which is inconsistent with the extrema-detection mechanism. A formal 415 model comparison confirmed that in each participant, the integration model provided a far 416 better account of subject responses than either of the non-integration models did ( Figure 5D). 417 We then assessed how subjects combined information from weak and strong evidence 418 samples into their decisions, using the same analyses as for monkeys. As predicted by the 419 integration model, but not by the extrema-detection model, humans choices consistently 420 aligned with the total stimulus evidence and not simply with the strongest evidence sample 421 ( Figure 5E). Finally, the average integration map for early and late evidence within the stimulus 422 sequence displayed nearly linear diagonal isolines, showing that both were integrated into the 423 response ( Figure 5F). Integration maps from participants correlated better with maps predicted 424 by the integration model than with maps predicted by either of the alternative non-integration 425 strategies ( Figure 5G; two-tailed t-test on bootstrapped r values: p<0.001 for 7 out 9 426 participants in the integration vs snapshot comparison; in all 9 participants for the integration 427 vs extrema-detection comparison). Overall, these analyses show converging evidence that 428 human decisions in an orientation discrimination task rely on temporal integration. 429
438
Integration map for early and late stimulus evidence, computed as in Figure 4A
Temporal integration in rat acoustic intensity judgments 444
Finally, we analyzed data from 5 rats performing a fixed-duration auditory task where the 445 animals had to discriminate the side with larger acoustic intensity (Pardo-Vazquez et al. 2019). 446 The relative intensity of the left and right acoustic signals was modulated in sensory samples 447 of 50 ms, so that the stimulus sequence provided time-varying evidence for the rewarded port. 448 The stimulus sequence was composed of either 10 or 20 acoustic samples of 50 ms each, for 449 a total duration of 500 or 1000 ms. We applied the same analysis pipeline as for monkey and 450 human data. The integration model provided a much better account of rat choices than non-451 integration strategies, based on psychometric curves (Fig. 6A), predicted accuracy (Fig. 6B), 452 psychophysical kernel (Fig. 6C) and model comparison using AIC (Fig. 6D). Similar to humans 453 and monkeys, rats tended to select the side corresponding to the total stimulus evidence and 454 not the largest sample evidence in "disagree" trials, as predicted by the integration model (Fig. 455 6E). Finally, the integration map was largely consistent with an integration strategy (Fig. 6F), 456 and correlated more strongly with simulated maps from the integration model ( In all analyses we contrasted predictions from one integration and two non-integration 472 computational models of behavioral responses (Figure 1). For each non-integration model, we 473 considered multiple variants to explore the maximal flexibility offered by each framework to 474 capture animal behavior. For our datasets, evidence in favor of integration was easy to achieve 475 using standard model comparison technique as well as comparing simulated psychometric 476 curves and psychophysical kernels to their experimental counterparts (Figure 2 This overall increased noise level leads to a looser relationship between the stimulus condition 507 and the behavioral responses, which can thus be accounted for by a larger spectrum of 508 computational mechanisms. These issues have been addressed by forcing "pulses" of a 509 certain stimulus strength and/or by performing post hoc analyses to estimate signal and noise 510 (Kiani, Hanks, and Shadlen 2008) but these are partial solutions that DSS paradigms solve by 511 design. This illustrates the benefits of using experimental designs where variability in stimulus 512 information can be fully controlled and parametrized by the experimenter, as these paradigms 513 discriminate more precisely between different models of perceptual decisions. 514 In at least one monkey, although quantitative metrics such as penalized log-likelihood and fits 515 to psychometric curves clearly pointed to the integration model as the best account to 516 behavior, the qualitative failure modes of the non-integration strategies (especially the 517 snapshot model) was not immediately clear. Although we tried variants for each non-518 integration model, there remained a possibility that our precise implementation failed to 519 account for monkey behavior but that other possible implementations would. Note that the 520 extrema-detection and snapshot are two of the many possible non-integration strategies. A 521 generic form for non-integration strategies corresponds to a policy that implements position-522 dependent thresholds on the instantaneous sensory evidence. In this framework, the extrema-523 dependent model corresponds to the case with a position-independent threshold, while the 524 snapshot model corresponds to a null bound for one sample and infinite bounds for all other 525 samples. To rule out these more complex strategies, we conducted additional analyses that 526 specifically targeted core assumptions of the integration and non-integration strategies. 527 First, the extrema-detection model fails to account for the data because it predicts that largest-528 evidence samples should have a disproportionate impact on choices. However, this does not 529 occur, as monkeys and humans tend to respond according to the total evidence and not the 530 single large-evidence sample ( Figure 3C and 5E) -see (Levi et al. 2018) for a similar analysis. 531 All non-integration strategies share the property that on each trial the decision should only rely 532 either on the early or the late part of the trial. We thus directly examined the assumptions of 533 integration and non-integration models by assessing how the evidence from the early and late 534 parts of each stimulus sequence is combined to produce a decision. We introduced integration 535 maps (Figure 4) We present here the most relevant features of the behavioral protocol -see (Yates et al. 2017) 606 for further experimental details. Two adult rhesus macaques (subject N, a 10-year old female; 607 and subject P, a 14-year old male) performed a motion discrimination task. On each trial, a 608 stimulus consisting of a hexagonal grid (5-7 degrees, scaled by eccentricity) of Gabor patches 609 (0.9 cycle per degree; temporal frequency 5 Hz for Monkey P; 7 Hz for Monkey N) was 610 presented. Monkeys were trained to report the net direction of motion in a field of drifting and 611 flickering Gabor elements with an eye movement to one of two targets. Each trial motion 612 stimulus consisted of seven consecutive motion pulses, each lasting 9 or 10 video samples 613 (150 ms or 166 ms; pulse duration did not vary within a session), with no interruptions or gaps 614 between the pulses. The strength and direction of each pulse for trial t and sample i was 615 set by a draw from a Gaussian rounded to the nearest integer value. The difficulty of each trial 616 was modulated by manipulating the mean and variance of the Gaussian distribution. Monkeys 617 were rewarded based on the empirical stimulus and not on the stimulus distribution. We 618 analyzed a total of 112 sessions for monkey N and 60 sessions for monkey P, with a total of 619 72137 and 33416 valid trials, respectively. These sessions correspond to sessions with 620 electrophysiological recordings reported in (Yates et al. 2017) and purely behavioral sessions. 621 All experimental protocols were approved by The University of Texas Institutional Animal Care 622 and Use Committee (AUP-2012-00085, AUP-2015-00068) and in accordance with National 623 Institute of Health standards for care and use of laboratory animals 624 625 Human experiment. 626 9 adult subjects (5 males, 4 females; aged 19-30) performed an orientation discrimination task 627 whereby on each trial they reported in each trial whether a series of gratings were perceived 628 to be mostly tilted clockwise or counterclockwise (Drugowitsch et al. 2016). Each discrete-629 sample stimulus consisted of five to ten gratings. Each grating was a high-contrast Gabor 630 patch (colour: blue or purple; spatial frequency = 2 cycles per degree; SD of Gaussian 631 envelope = 1 degree) presented within a circular aperture (4 degrees) against a uniform gray 632 background. Each grating was presented during 100 ms, and the interval between gratings 633 was fixed to 300 ms. The angles of the gratings were sampled from a von Mises distribution 634 centered on the reference angle ( 0 = 45 degrees for clockwise sequences, 135 degrees for 635 anticlockwise sequences) and with a concentration coefficient = 0.3. The normative 636 evidence provided by sample i in trial t in favor of the clockwise category corresponds to how 637 well the grating orientation aligns with the reference orientation, i.e. = 2 (2( − 638 0 )) . 639 Each sequence was preceded by a rectangle flashed twice during 100 ms (the interval 640 between the flashes and between the second flash and the first grating varied between 300 641 and 400 ms). The participants indicated their choice with a button press after the onset of a 642 centrally occurring dot that succeeded the rectangle mask and were made with a button press 643 with the right hand. Failure to provide a response within 1000 ms after central dot onset was 644 classified as invalid trial. Auditory feedback was provided 250 ms after participant response 645 (at latest 1100 ms after end of stimulus sequence). It consisted of an ascending tone (400 646 Hz/800 Hz; 83 ms/167 ms) for correct responses; descending tone (400 Hz/ 400 Hz; 83 647 ms/167 ms) for incorrect responses; a low tone (400 Hz; 250 ms) for invalid trials. there was only one grating in the sequence, and it was perfectly aligned with one of the 652 reference angles. In the second initiation block, sequences of gratings were introduced, and 653 the difficulty was gradually increased (the distribution concentration linearly decreased from 654 = 1.2 to = 0.3). Invalid trials (mean 6.9 per participant, std 9.4) were excluded from all 655 regression analyses. Long-Evans rats (no genetic modifications; 350-650g; 8-10 weeks-old at the beginning of the 662 experiment), pair-housed and kept on stable conditions of temperature (23 o C) and humidity 663 (60%) with a constant light-dark cycle (12h:12h, experiments were conducted during the light 664 phase). Rats had free access to food, but water was restricted to behavioral sessions. Free 665 water during a limited period was provided on days with no experimental sessions. 666 Rats performed a fixed-duration auditory discrimination task where they had to classify noisy 667 stimuli based on the intensity difference between the two lateral speakers (Pardo-Vazquez et 668 al. 2019; Hermoso-Mendizabal et al. 2020). A LED on the center port indicated that the rat 669 could start the trial by poking in that center port. After this poke, rats had to hold their snouts 670 in the central port during 300 ms (i.e. fixation). Following this period, an acoustic DSS was 671 played. Rats had to remain in the central port during the entire presentation of the stimulus. 672 At stimulus offset, the center LED went off and rats could then come out of the center port and 673 head towards one of the two lateral ports. Entering the lateral port associated with the speaker 674 that generated the larger sound intensity led to a reward of 24 µl of water (correct responses), 675 while entering the opposite port lead to a 5 s timeout accompanied with a bright light during 676 the entire period (incorrect responses). If rats broke fixation during the pre-stimulus fixation 677 period or during the stimulus presentation, the sound was interrupted, the center LED 678 remained on, and the rat had to initiate a new trial starting by center fixation followed by where describes a sensitivity parameter. The deterministic case can be viewed as the limit 733 of the non-deterministic case when all sensitivity parameters diverge to +∞, i.e. when 734 sensory and decision noise are negligible. 735 The overall probability for selecting right choice (marginalizing over the attended sample, 736 which is a hidden variable) can be captured by a mixture model : 737 The mixture coefficients ( = 1, . . , , ) are constrained to be non-negative and sum up to 739 1. In the non-deterministic model, the parameters also include sensitivity parameters . The 740 model is fitted using Expectation-Maximization (Bishop 2006). In the Expectation step, we 741 compute the responsibility , i.e. the posterior probability that the sample i was attended at 742 trial t (for i=L, R, the probability that the trial corresponded to a lapse trial To speed up the computations, in each M step, we only performed one Newton-Raphson 757 update for each sensitivity parameter, rather than iterating the updates fully until convergence. 758 The EM procedure was run until convergence, assessed by an increment in the log-likelihood 759 ( , ) of less than 10 -9 after one EM iteration. The log-likelihood for a given set of parameters 760 is given by ( , ) = ( (0, 2 ). H is the step function. If the stimulus sequence ends and no sample has reached the 784 threshold, then the decision is taken at chance. As described in (Waskom and Kiani 2018), 785 the probability for a rightward choice at trial t can be expressed as: 786 obtained from the Laplace approximation. For psychometric curves, we first defined the 808 weighted stimulus evidence Tt at trial t as the session-modulated weighted sum of signed 809 sample evidence; with the weights obtained from the logistic regression model above 810 = . We then divided the total stimulus evidence into 50 quantiles (10 for human 811 subjects) and computed the psychometric curve as the proportion of rightward choices for 812 each quantile. 813 The boundary performance for the snapshot and extrema-detection models corresponds to 814 the best choice accuracy out of all the parameterizations for each model . In the snapshot 815 model, the boundary performance corresponds to the deterministic version with no-lapse, 816 where the attended sample is always the sample * whose sign better predicts the stimulus 817 category over all animal trials, i.e. * = 1 and = 0 if ≠ * . For the extrema-detection model, 818 the boundary performance corresponds to the lapse-free model with no sensory noise ( = 0) 819 and a certain value for threshold that is identified for each subject by simple parameter 820 search. 821 Finally, model selection was performed using the Akaike Information Criterion = 2 − 822 2 , where p is the number of model parameters and is the likelihood evaluated at 823 maximum likelihood parameters. 824 825 Analysis of majority-driven choices 826 We selected for each animal the subset of trials corresponding to when the largest evidence 827 sample was at odds with the total stimulus evidence, i.e. where ( , | | ≥ | | ∨ ) ≠ 828 ( ). For this subset of trials, we computed the proportion of animal choices that were 829 aligned with the overall stimulus evidence. We repeated the analysis for simulated data from 830 the integration and extrema-detection models. 831 832 Subjective weighting analysis 833 In order to estimate the impact of each sample on the animal choice as a function of sample 834 evidence, we built and estimated the following statistical model 835 As can be seen, this model is equivalent to the temporal integration model under the 837 assumption that f is a linear function. Rather, here we wanted to estimate the function f (as 838 well as the session gain , lateral bias 0 and sensory weight ). Including the session gain 839 was necessary for estimating f accurately from the monkey and rat behavioral data, since the 840 distribution of pulse strength was varied across sessions and could otherwise induce a 841 confound. We assumed that f is an odd function, i.e. (− ) = − ( ). This equation takes 842 the form of a Generalized Unrestricted Model and was fitted using the Laplace approximation 843 method as described in (Adam and Hyafil 2020 , where the 854 weights and session gains correspond to parameters estimated from the temporal 855 integration model (session gains were omitted for human participants). Next we plotted the 856 integration map which represents the probability for rightward choices as a function of ( , ). 857 The map was obtained by smoothing data points with a two-dimensional gaussian kernel. 858 More specifically, for each pair value (E,L), we selected the trials whose early and late 859 evidence values and fell within a certain distance to (E,L), i.e. = (( , )( , )) < 860 2. We then computed the proportion of rightward choices for the selected trials, with a weight 861 for each trial depending on the distance to the pair value = (( , ); ( , ),0. 1 2 ). 862 Because the space (E,L) was not sampled uniformly during the experiment, we represent the 863 density of trials by brightness. For each subject we obtained integration maps both from 864 subject data as well as from model simulations. For each model, we computed the Pearson 865 correlation between the maps obtained from the corresponding simulation and from the 866 subject data. We tested the significance of correlation measures between models by using a 867 bootstrapping procedure: we calculated the correlation measure r from 100 bootstraps for 868 each model and participant, and then performed an unpaired t-test between bootstrapped r. 869 870 Next, we analyzed the conditional psychometric curves, i.e. the psychometric curves for the 871 early evidence conditioned on the value of late evidence, which correspond to vertical cuts in 872 the integration map. To do so, we first binned late evidence by bins of width 0.5. Conditional 873 psychometric curve represent the probability of rightward choices as a function of early 874 evidence , separately for each late evidence bin. For each late evidence bin, we also 875 estimated the corresponding bias , left lapse and right lapse by fitting the following 876 function on the corresponding subset of trials: 877 . Note that this weighting converts the 889 evidence onto the space defined by the preferred direction of the neuron, such that positive 890 evidence signals evidence towards the preferred direction and negative evidence signals 891 evidence towards the anti-preferred direction. We then merged the vectors for normalized 892 spike counts ( ) = ( ) / ( 0 ( ) ), early evidence ( ) and late evidence ( ) across all 893 neurons. The normalized spike counts were binned by values of early and late evidence (bin 894 width: 0.02), and the average over each bin was computed after convolving with a two-895 dimensional gaussian kernel of width 0.1. The neural integration map represents the average 896 normalized activity per bin. 897 Simulations of spiking data for the integration and non-integration models were proceeded as 898 follows. First, the neural integration model corresponds to linear summing with neuron-specific 899 weights which are then passed through an exponential nonlinearity; the spike counts for each 900 trial are generated using a Poisson distribution whose rate is equal to the nonlinear output 901 (Supp Figure 9a, top). This corresponds exactly to the generative process of the Poisson GLM 902 described above. For the extrema detection model (Supp Figure 9a middle), we hypothesized 903 that LIP activity would only be driven by the sample that reaches the threshold (and dictates 904 the animal response). To this end, we first simulated the behavioral extrema detection model 905 for all trials, using parameters ( , , , ) fitted from the corresponding animal, to identify 906 which sample i reaches the subject-specific threshold. We then assumed that the spiking 907 activity of the neuron would follow the stimulus value at sample i (signed by the preferred 908 direction of the neuron ( ) through: 909 ( ( ) ) = ( 0 ( ) + ( ) ( ) /2) 910 Again the spike count were generated from a Poisson distribution with rate ( ( ) ). 911 Finally, for the snapshot model (Supp. Figure 9a bottom), we assumed that the neuron activity 912 would merely reflect the sensory value of the only sample it would attend. We assumed that 913 the probability mass function to attend each of the 7 samples would be neuron-specific, so we 914 used the normalized weights of the Poisson GLM for that specific neuron as defining such 915 probability (weights were signed by the neuron preferred direction so that the vast majority of 916 weights were positive; negative weights were ignored). For each trial, we thus randomly 917 sampled the attended sample i using this probability mass function and then simulated the 918 spike count ( ) from a Poisson distribution with rate ℎ ( ( ) ) = ( 0 ( ) + 919 ( ) ( ) ). 920 We simulated spiking activity for each neuron and for each integration and non-integration 921 model, and then used simulated data to compute neural integration maps exactly as described 922 above for the actual LIP neuron activity. 923 924 Data and code availability 925 All experimental data (behavioral and neural data in monkeys, behavioral data in rats and 926 humans) and code to run the analysis will be made publicly available 927 at [URL] prior to final publication 928
==
Domain: Psychology Biology Medicine
|
[{"added": "2022-10-30T13:44:41.829Z", "created": "2022-10-25T00:00:00.000", "id": "253206227", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2022/10/26/2022.10.25.513647.full.pdf", "pdf_hash": "bb0ff4aa66444a04d1ba6ccbdd523e3cea9c908a", "pdf_src": "BioRxiv", "provenance": "peS2o-0075.json.gz:35131", "s2fieldsofstudy": ["Psychology"], "sha1": "bb0ff4aa66444a04d1ba6ccbdd523e3cea9c908a", "year": 2022}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The propensity to sign-track is associated with externalizing behavior and distinct patterns of reward-related brain activation in youth
Externalizing behaviors in childhood often predict impulse control disorders in adulthood; however, the underlying bio-behavioral risk factors are incompletely understood. In animals, the propensity to sign-track, or the degree to which incentive motivational value is attributed to reward cues, is associated with externalizing-type behaviors and deficits in executive control. Using a Pavlovian conditioned approach paradigm, we quantified sign-tracking in 40 healthy 9–12-year-olds. We also measured parent-reported externalizing behaviors and anticipatory neural activations to outcome-predicting cues using the monetary incentive delay fMRI task. Sign-tracking was associated with attentional and inhibitory control deficits and the degree of amygdala, but not cortical, activation during reward anticipation. These findings support the hypothesis that youth with a propensity to sign-track are prone to externalizing tendencies, with an over-reliance on subcortical cue-reactive brain systems. This research highlights sign-tracking as a promising experimental approach delineating the behavioral and neural circuitry of individuals at risk for externalizing disorders.
Results
Pavlovian conditioned approach. Our primary goal was to establish a Pavlovian conditioning paradigm comparable to those used in animal models to identify sign-tracking and goal-tracking phenotypes in youth. We used the Pavlovian conditioning paradigm described by Flagel et al. 32 for rodents and adapted for humans by Joyner et al. 30 as a basis for the current study (Fig. 1a). An a priori power analysis based on the index of Pavlovian conditioned behavior found in an existing study of sign-tracking and goal-tracking in humans 27 indicated that a sample of 40 would be sufficient to detect a large effect (d = 0.91) with 80% power (see Methods for the full a priori power analysis report). Participants (N = 40, Table 1) completed 40 response-independent conditioning trials consisting of lever (CS) presentation and subsequent reward (US; $0.20 token; all monetary units in USD) presentation. A randomly selected inter-trial interval (ITI) followed each CS-US trial.
Feasibility was assessed by identifying both qualitative ( Supplementary Fig. S1) and quantitative markers of child engagement and learning. We identified individual differences in the attribution of incentive salience to reward-paired cues (sign-tracking or goal-tracking behaviors) via a Pavlovian Conditioned Approach (PavCA) index based on previously used models in animal studies 33 derived from the number and timing of physical contacts to the CS and US during CS-presentation and ITI phases.
Consistent with animal models 33 , we classified categorical phenotypic groups using a PavCA value during the CS-period of 0.5 or greater (STs) and less than −0.5 (GTs). PavCA scores during the CS-period ranged from −0.18 to 0.95 (m = 0.41, sd = 0.40; Supplementary Table S1) and the distribution indicated that participant behaviors were skewed toward either neutral or lever-directed behaviors (Fig. 1b), therefore we were able to identify STs, but not GTs. Nineteen participants were identified as STs (m PavCA = 0.79, sd = 0.13), and since no participants had a PavCA value less than −0.5, we classified the remaining 21 participants as non-sign-trackers (non-ST; m PavCA = 0.06, sd = 0.14) and used these as our comparison groups (see Supplemental Materials To assess learning of conditioned responses to the CS, we examined behavioral responses to the CS using twosided linear mixed effects models with the factors time (Block 1-4), phase (CS, ITI), and phenotype (ST, non-ST) for lever-and reward-directed behaviors ( Supplementary Fig. S2). Because CS lever contacts correlated positively with ITI lever contacts (r = 0. 59 www.nature.com/scientificreports/ for lever-and reward-directed behaviors (contacts and probability) in order to adequately compare between CS and ITI periods. Briefly, we observed that STs demonstrated a higher probability to contact the lever during the CS period and over time (three-way interaction, F 3,266 = 3.84, p = 0.010, η 2 p = 0.04, 90% CI = 0.00 to −0.08). STs and non-STs differed in their probability to contact the reward between phases (phase main effect, F 1,266 = 504.79, p < 0.001, η 2 p = 0.65, 90% CI = 0.60 to 0.70) however there were no significant phenotypic differences during either phase or over time (three-way interaction, F 3,266 = 2.07, p = 0.104, η 2 p = 0.02, 90% CI = 0.00 to 0.05). Of note, there were significant main effects for age (F 1,36 = 6.62, p = 0.010, η 2 p = 0.16, 90% CI = 0.02 to 0.34) and sex (females higher; F 1,36 = 7.43, p = 0.009, η 2 p = 0.17, 90% CI = 0.03 to 0.35) in the probability to contact the reward which may implicate developmental differences impacting goal-tracking behaviors. Together, it appears that much of the individual variation in behavior stemmed from lever-directed behaviors during CS presentation which supports the general skew towards sign-tracking behavior.
We also examined PavCA behavior between phases for each phenotype over time using non-normalized metrics (Fig. 1c). PavCA scores showed significant main effects for phenotype (F 1,36 = 52.82, p < 0.001, η 2 p = 0.59, 90% CI = 0.42 to 0.71) and phase (F 1,265.1 = 1274.34, p < 0.001, η 2 p = 0.83, 90% CI = 0.80 to 0.85), suggesting that participants are behaviorally discriminating between phases, and STs are doing so to a greater degree. STs increasingly approached the lever over time (higher PavCA score; phenotype by time interaction, F 3,265.1 = 2.63, p = 0.050, η 2 p = 0.03, 90% CI = 0.00 to 0.06) and more so during CS presentation (phenotype by phase interaction, F 1,265.1 = 131.46, p < 0.001; three-way interaction, F 1,265.1 = 4.21, p = 0.006, η 2 p = 0.05, 90% CI = 0.01 to 0.09). This appears to reflect learning of the conditioned response and the characterization of sign-tracking by selectively increasing lever approach during the CS period. Conversely, non-STs progressively decreased their lever-directed behaviors over the course of training, further characterizing the distinction between phenotypic responses. PavCA showed a significant main effect for sex (males higher; F 1,36.1 = 13.28, p = 0.001, η 2 p = 0.27, 90% CI = 0.09 to 0.45) but not age (F 1,36 = 1.64, p = 0.208, η 2 p = 0.04, 90% CI = 0.00 to 0.19). Together, these results show that the two phenotypes selectively discriminate between phases and, during CS, PavCA behaviors continually diverge over the course of the session. www.nature.com/scientificreports/ Externalizing behaviors by phenotype. The data presented so far provide evidence for a bias toward responding to reward-paired cues, indicative of sign-tracking, in a subset of youth. Given the well-established characteristic differences between STs and GTs in animal studies 14,21 , we aimed to validate the application of phenotypic distinctions in human youth. Our primary hypothesis was that human STs would demonstrate symptomatic and neurobiological profiles consistent with externalizing characteristics and a reliance on bottom-up processing rather than top-down cognitive control. We examined this using developmentally validated parentreport questionnaires (Child Behavior Checklist, CBCL 34 and Early Adolescent Temperament Questionnaire, EATQ 35 ). It is important to note that these measures do not represent clinical diagnoses, but dimensions of behavior associated with psychiatric symptoms. Measures were tested for normality and all CBCL subscales were log transformed. To directly compare measures between STs and non-STs, we used two-sided Welch two-sample t-tests (Supplementary Table S3; Fig. 2) with FDR p-value corrections used for multiple comparisons. Given that attentional control deficits are a hallmark characteristic in rodent STs 19,21 , and given the shared circuitry between sign-tracking rats 21 and humans with attention-based and impulse control disorders 36 , we expected elevated attention problems in the sign-trackers in our sample. Indeed, when compared to non-STs, STs showed increased symptoms of attention deficit/hyperactivity problems (CBCL; t 30.9 = −2.66, p = 0.014, d = 0.90, s.e. = 0.31, 95% CI = −1.45 to −0.19; Fig. 2a) which supports prior work highlighting differences in reward processing in attention disorders 37 .
Animal literature has further identified characteristic differences in fear responses, such that, when exposed to fear conditioning tasks, STs show exaggerated fear-associated cue reactivity consistent with an increased susceptibility to both substance use and post-traumatic stress 38,39 . When compared to non-STs, STs in our sample had increased reported symptoms of fear (EATQ; unpleasant affect related to anticipation of distress; t 35 = −2.79, p = 0.012, d = 0.90, s.e. = 0.18, 95% CI = −0.88 to −0.14; Fig. 2d), which is in line with previous animal findings implicating STs as having an increased vulnerability to fear-related responses to cues, regardless of context 39 .
Furthermore, inhibitory control deficits have been associated with sign-tracking in rats 18,20,40 as well as an increased vulnerability for impulse control disorders and addiction in humans 41 16,18,48 , this relationship should be further examined within a larger sample.
Beyond the translational utility of capturing sign-tracking tendencies in youth, we aimed to examine symptomatic differences between phenotypes consistent with human-specific psychopathology. Specifically, we expected to see behavioral tendencies in STs that are developmentally characteristic of externalizing disorders. In contrast to non-STs, caregivers reported STs as having increased oppositional defiant problems (CBCL; t 31 = −3.44, p = 0.006, d = 1.16, s.e. = 0.29, 95% CI = −1.58 to −0.40; Fig. 2b) and social problems (CBCL; including items relating to jealousy, not getting along with others, and not being liked by others; t 30.2 = −3.10, p = 0.008, d = 1.05, s.e. = 0.30, 95% CI = −1.56 to −0.32; Fig. 2c). Although these two subscales are positively correlated (r = 0.64, p < 0.001; Supplementary Table S4) suggesting homogenous traits, they are derived from independent items on the CBCL. STs also showed increased levels of negative affect (EATQ composite score of frustration, depressive mood, and aggressive behaviors; t 27.3 = −3.14, p = 0.006, d = 1.14, s.e. = 0.16, 95% CI = −1.89 to −0.22; Fig. 2f). Taken together, the association between the degree of sign-tracking and these measures support the notion that STs exhibit deficits in behavioral regulation and inhibitory control as well as a bias toward affective/reactive responding across multiple diagnostic criteria 42 . Neuroimaging. The results presented above demonstrate behaviors and tendencies specific to STs that may broadly indicate a reliance on bottom-up processing and are consistent with characteristics of both rodent models and theoretical accounts of translation of these paradigms to clinical populations 43 . To further elucidate the neurobiological processes that may contribute to the propensity to sign-track, we employed functional neuroimaging to measure reward processing. Participants completed the Monetary Incentive Delay (MID) fMRI task 44 to determine the brain response to a gain, no gain, or loss predictive cue. We performed a whole-brain voxelwise linear mixed effects analysis with fixed effects for group, condition, age, and sex, a group by condition interaction, and a random intercept for subject. We followed this with planned contrasts investigating a group (ST, non-ST) by condition (win vs neutral or loss vs neutral) interaction. Estimated marginal means were used for post hoc tests with Tukey p-value adjustment for multiple comparisons. A total of 29 subjects (n ST = 12, n non-ST = 17) survived motion correction.
When examining the win-neutral contrasts during the MID anticipatory phase, BOLD activations in the left inferior parietal lobe (IPL) showed a significant group by condition interaction (F 1,27 = 10.30, p = 0.003, η 2 p = 0.28, 90% CI = 0.07 to 0.48, Fig. 3a; see Supplementary Table S5 for details on coordinates and volumes for these and additional significant regions). Post hoc tests using average contrasts extracted from significant voxels in each ROI indicate that non-STs significantly increased activation from neutral to large win conditions (t 27 = −4.15, p = 0.002, d = −1.42, s.e. = 0.37, 95% CI = −2.19 to −0.66), indicating salience-dependent modulation within the IPL. Conversely, there was no evidence for a similar effect in STs and activations did not differ between conditions or from non-STs during the win condition (ps > 0.8). Importantly, this same modulatory pattern is evident in the left IPL for the loss-neutral contrast during the anticipatory phase (group by condition interaction, F 1,27 = 12.31, p = 0.002, η 2 p = 0.31, 90% CI = 0.09 to 0.51; Fig. 3b), which appears to indicate that this effect is not reward-specific but rather salience-driven. This same pattern is consistent across multiple cortical regions implicated in cognitive Table S5) which may indicate that, in comparison to non-STs, STs do not actively engage cognitive control and the salience of the cue is cognitively irrelevant. During the win-neutral anticipation contrasts, BOLD activations in the right amygdala showed a significant group by condition interaction (F 1,27 = 9.04, p = 0.006, η 2 p = 0.25, 90% CI = 0.05 to 0.46; Fig. 3c), such that STs exhibited a significant increase in amygdala activation from neutral to large win conditions (t 27 = −3.05, p = 0.025, Cohen's d = −1.25, s.e. = 0.43, 95% CI = −2.12 to −0.37). These findings indicate salience-dependent modulation within the amygdala only for STs. In contrast to the IPL, amygdala activation for non-STs did not differ between conditions or from STs during either condition (ps > 0.1; see also Supplementary Fig. S5 for additional sensitivity analysis after removing a possible outlier in the ST group). Taken together, the associations between cortical activation during salience processing for non-STs and subcortical activation during salience processing for STs, support two independent processes that contribute to the emergence of sign-tracking and non-sign-tracking behaviors. Whereas non-ST individuals showed greater activation in brain areas that have been associated with executive control, ST youth showed greater activation in the salience network.
To further understand the possible implications of these differential neural modulatory patterns, we used Pearson's r to examine how percent signal change correspond to both externalizing behaviors and Pavlovian conditioned approach indices ( Fig. 4; Supplementary Table S6). Left IPL activation during win-neutral contrasts was negatively related to oppositional defiant problems (r = −0.43, p = 0.020), marginally negatively with negative affect (r = −0.34, p = 0.070), and marginally positively with inhibitory control (r = 0.36, p = 0.060), whereas left IPL during loss-neutral contrasts was negatively related to risk-taking behaviors (r = −0.45, p = 0.020). Although not all statistically significant, these associations support the notion that brain-behavior associations might differ by phenotype. We further examined the correlations between percent signal change and PavCA scores. Activations in both regions during their respective contrasts were significantly correlated with the propensity to sign-track such that higher PavCA scores (sign-tracking) were related to less activity in the left IPL during both win-neutral (r = −0.46, p = 0.010) and loss-neutral (r = −0.45, p = 0.010) as well as greater activity in the amygdala during win-neutral (r = 0.48, p = 0.010). Note that since the ROIs were selected based on differential activation www.nature.com/scientificreports/ between STs and non-STs, these post-hoc correlations with clinical measures are inflated and would be smaller in an independent sample. These findings are in agreement with the pre-clinical literature, suggesting that the behavioral endophenotype of sign-trackers is dominated by subcortical motivational systems; whereas that of non-sign-trackers is dominated by cortical processes.
Environmental influences. Given that the stress response system directly influences both the dopaminergic reward system 47 and animal sign-and goal-tracking behaviors 7,48,49 , we also gathered information regarding environmental markers of adversity and protective factors including relationships and resources available to these youth 50 Table S3). Together, these findings support prior research addressing early life adversity as a potential influential driver for the divergence in phenotypic differences and associated vulnerabilities. While we did not have adequate power to do so, future studies with larger sample sizes would also benefit from examining the question whether social determinants of mental health are an important mediator or moderator for the expression of sign-or goal-tracking tendencies. Oppositional defiance = DSM-oriented subscale from the Child Behavior Checklist (parent report). Scores were log transformed due to non-normality. Externalizing problems = composite Child Behavior Checklist subscale (parent report); Inhibitory control = subscale from the Early Adolescent Temperament Questionnaire (parent report). Surgency = subscale from the Early Adolescent Temperament Questionnaire (parent report); Negative Affect = composite subscale including depressive mood, aggressive behaviors, and frustration on the Early Adolescent Temperament Questionnaire (parent report); Aggression = subscale from the Early Adolescent Temperament Questionnaire (parent report); Sensation seeking = subscale of the Urgency, Premeditation, Perseverance, Sensation Seeking, Positive Urgency scale (youth self-report
Discussion
This study translated a paradigm previously developed to examine sign-and goal-tracking behaviors in rodents in order to determine whether these behavioral distinctions (1) could be observed in human youth, and (2) are associated with externalizing behaviors or reward-related neural activation patterns. First, we present evidence for sign-tracking behaviors in pre-adolescents. Second, STs were distinctive in both externalizing traits and neurobiological patterns consistent with impulse control disorders. Specifically, STs showed greater externalizing characteristics and salience-dependent subcortical reactivity to reward cues, whereas non-STs had fewer externalizing characteristics and more actively engaged cortical control regions. Third, both externalizing traits and the degree of sign-tracking behavior correlated with neural modulation patterns such that those with higher BOLD activation in the left IPL also reportedly had fewer externalizing characteristics and had lower PavCA scores (non-STs) whereas those with increased BOLD activation in the amygdala also had higher PavCA scores (STs). Finally, those displaying a higher degree of sign-tracking behavior also reported increased potential for early life stress (lower income, increased basic needs unaffordability, and fewer protective factors). Taken together, the phenotypic differences in our sample are consistent with rodent models of sign-tracking and support prior findings from humans with impulse control disorders 5 .
The differences in parent-reported behavioral tendencies between STs and non-STs in our sample reflect characteristic patterns in animal studies that are mechanistically tied to dual-systems processing 5 . A fundamental characteristic of an incentive stimulus is its ability to bias attention and elicit approach behaviors suggestive of distinct cognitive/attentional control tendencies specific to this sign-trackers. In humans, attentional capture to reward-paired stimuli (resonant of sign-tracking behaviors) appears to be stronger for those with low cognitive control 51 and directly linked to both impulsivity 52 and substance use 51 . Thus, the presence of increased attentional and executive control problems in STs, in conjunction with cortical activation patterns, suggests that, the attribution of incentive salience to reward-paired cues is driven by inefficient top-down cognitive and executive control 19 . The other symptoms reported here for STs reflect patterns of behavior consistent with the profile of dopamine-driven cue-reactivity in externalizing disorders, namely, defiance, aggression, and impulsivity. Notably, lower activation in cortical regions (i.e., in STs) is also correlated with behavioral reports of increased risk taking, defiance, and to a lesser extent externalizing behaviors, aggression, and lower inhibitory control. Finally, STs were reported to have increased fear-related behaviors, a finding that is supported by rodent models 38,39 . This supports our finding that non-STs are not only reported to have fewer fear-related behaviors but also a neural modulatory profile indicative of emotion regulation and cognitive control.
Animal studies have shown a consistent double dissociation between top-down dominant cortical systems in GTs and subcortical cue-driven systems that dominate in STs 14 . Here, we found concurring evidence in humans that, in contrast to STs, non-STs cognitively modulate their anticipatory responses in the IPL (and consistently across multiple cognitive control regions) but not the amygdala according to cue salience. The IPL has been implicated in probability and reward-related decision making 53 and this pattern for non-STs may reflect cognitive discrimination in anticipatory and preparatory responses according to the salience of the cues. Whereas for STs, the non-modulation in the IPL appears to indicate indiscriminate cortical activation according to cue salience, suggesting that, cognitively, STs interpret all cues as salient and respond relatively equally across conditions. Thus, the selective modulation in this region by non-STs demonstrates distinct cognitive assessment of the cue and preparatory responses selectively applied to highly salient cues. STs do, however, affectively modulate subcortical (amygdala) activity in preparation for reward-paired cues. The amygdala contributes to contextual appraisal of rewards and motivational significance of incentives and is integral to neural processing of reward and reward cues 15 . Therefore, the apparent reliance for STs on primarily subcortical structures to modulate responses and the fact that this salience-dependent modulation does not translate to cognitive action, supports a theory of increased vulnerability to impulse control disorders based on both individual differences in incentive salience attribution and mechanistic differences in reward processing. Our findings are also consistent with a recent fMRI analysis of model-based vs model-free learning in humans demonstrating that increased incentive salience attribution was associated with neural profiles of striatal reward prediction error signals in STs; whereas stronger cortical signals of state-prediction errors were evident in GTs 28 . Together, STs in our sample are categorized both by behavioral Pavlovian responses to reward-paired cues and selectively modulated amygdala activity which points to a likelihood that these patterns are dopamine-dependent and supports previous pre-clinical animal reports of a dual-systems approach.
The dynamic nature of cortical development, top-down control, and neural organization in pre-adolescence 31 highlights the importance of considering developmental trajectories in light of both sign-tracking and externalizing tendencies. Specifically, adolescence is characterized by underdeveloped cortical downregulation of the amygdala that contributes to attentional/executive control deficits 31 . It is likely, therefore, that this developmental variability is, at least in part, impacting both the neuromodulatory patterns and the significant skew in the tendency to sign-track. For instance, the limited goal-tracking behaviors is a notable deviance from rodent models. While age did not significantly differ between phenotypes, given our sample constraints, examination of these traits in larger samples with varying age ranges would help to further elucidate whether this is an effect of developmental stage or another factor not accounted for in our translation of this paradigm. However, in the absence of a significant age effect in PavCA behaviors in our sample, as well as statistical controls for age in place in neuroimaging models, it remains striking that we measured marked individual differences in both brain and behavior. Further, the dynamic nature of both neural and behavioral traits during this age range, underscores the importance of measuring potential predictors of risk for psychopathology, as these characteristics may still be malleable. While more research is necessary to further clarify the developmental trajectory of sign-and goal-tracking and its impact on future psychopathology, measuring and characterizing these phenotypes early www.nature.com/scientificreports/ and longitudinally as they evolve throughout development, underscores the potential utility of this paradigm to delineate possible preventative and interventive measures. The individual differences in both the degree of sign-tracking and the development of impulse control deficits suggests the possibility of additional risk factors at play. Both sign-tracking 49 and impulse control disorders 47 are influenced by stressful early environments. Our finding that STs have lower income households, increased difficulty for affording basic needs, and less social support suggests probable environmental impacts on individual differences in sign-tracking tendencies and the underlying neurobiological mechanisms. The sensitivity of amygdala-frontal development in adolescence to stressful environmental input can bias the reward system to be more reactive to cues 54 and vulnerable to reward-seeking and substance use 55 . The finding that indicators of stressful environments are present more prominently in STs is consistent with this prior work and is supported by the corresponding psychopathological and neural profiles we observe in STs. However, this finding will need to be examined more fully in larger, more demographically diverse samples.
As a whole, these results provide evidence for sign-tracking in youth and a general neuromodulatory and behavioral profile consistent with externalizing traits; however, a number of limitations should be noted. First, our sample size was relatively small and demographically homogenous which limited our variance and power to perform multivariate analyses or complex neuroimaging models. With 40 participants, we were only powered to detect effects larger than roughly 75% of those seen in larger social psychology studies 56 . Additionally, while similarly sized neuroimaging studies have been the norm, reproducibility has been poor and there is even evidence that thousands of participants are necessary to conduct reproducible brain-wide association studies 57 . These two facts lead to relatively high risk of both type I and type II errors, a caveat that could be included in most similarly sized studies. In particular, future, higher powered studies would benefit from statistically addressing environmental factors including income and family history of psychopathology and substance use as well as individuallevel factors such as pubertal stage to further account for possible sex differences. Additionally, a larger cohort would provide the opportunity to address the more limited extremes of sign-and goal-tracking behaviors and further validate if these neural and behavioral distinctions reflect differences in risk for externalizing disorders. Further, our aim was to measure the potential of sign-tracking behaviors as an early biomarker and/or behavioral marker for the risk of later development of psychopathology in a healthy sample. We therefore cannot directly address the application of sign-and goal-tracking to clinical diagnoses or the potential range of sign-and goaltracking behaviors within a clinical sample from this study. Future research is needed to further investigate the direct relevance of sign-and goal-tracking to clinical symptoms. Also of note is our use of monetary rewards for both the PavCA paradigm and the MID task which could influence salience for those participants coming from lower income households. Finally, additional methods of measurement should be considered (e.g., eye tracking, approach behaviors) that could further elucidate whether goal-tracking is, in fact, limited in pre-adolescence, or simply not adequately measured in this sample.
The current data strongly support the feasibility and utility of the sign-tracker/goal-tracker model of reward processing as a useful experimental approach and construct to measure individual differences in the propensity to attribute incentive salience to reward cues in youth. Moreover, these findings reveal a consistent and substantial pattern of neuromodulatory and externalizing trait responses that reliably dissociate bottom-up processing in STs from top-down cognitive control in non-STs. These results provide evidence in human youth for an underlying dual-systems mechanism of neural reward processing between cortical and subcortical control and link the degree of sign-tracking to externalizing aspects of psychopathology that may predispose some individuals to the development of impulse control disorders. The feasibility of this paradigm and the initial delineation of the circuitry associated with the degree of sign-tracking is enhanced by the extensive knowledge base obtained from prior animal studies, which provide neuroscience-based rationale for future studies elucidating the underlying mechanisms of risk for externalizing psychopathology. While there is still more to be discovered, particularly regarding the developmental trajectory, stability, or malleability of these phenotypes in humans, these data offer promise in identifying sign-and goal-tracking phenotypes in humans and present initial evidence detailing corresponding neural profiles and behavioral traits consistent with indicators of impulse control disorders, helping to pave the way for future research with clinical applications.
Materials and methods
Participants and sample selection. The sample consisted of 9-12-year-old youth (N = 40, m = 9.6 years, sd = 0.93; Table 1). All tasks and measures were used in the whole sample. Participants were excluded if they received a diagnosis of a severe learning disorder, Axis 1 psychiatric disorder, Attention Deficit/Hyperactivity Disorder or Pervasive Developmental Disorder (e.g., Autism Spectrum Disorder) as these conditions may affect attentional control and possibly bias the measurement of attention to each stimulus, or if they endorsed MRI contraindications including non-correctable vision, hearing, or sensorimotor impairments, claustrophobia, large body size, or irremovable ferromagnetic metal implements or dental appliances. Participants were recruited via fliers and online advertisements. Youth participated with one parent/caregiver present and all participants were compensated for their time. Caregivers completed questionnaires including demographics, youth behaviors, and family environment. Youth completed the sign-and goal-tracking task at the beginning of each session followed by self-report questionnaires, behavioral and neurocognitive tasks, practice tasks for imaging sessions, and neuroimaging. All study procedures were approved by and performed in accordance with relevant guidelines and regulations of the Western Institutional Review Board. Informed consent and written permission for youth participation (including consent for publication of images/videos) were obtained from caregivers and informed assent was obtained from youth. An a priori power analysis using G*Power software 58 was used to estimate the appropriate sample size based on the index of Pavlovian conditioned behavior found in an existing study of sign-tracking and goal-tracking in humans 27 . We applied power estimation procedures based on these www.nature.com/scientificreports/ values, a minimum effect size of 0.91, and assumed 2-tailed alpha of 0.05. This analysis indicated that a sample of 40 would be sufficient to detect an effect with 80% power. While under-powered for more complex analyses involving additional covariates, these results will be critical for estimating effect sizes with the caveat that effects reported here are likely overestimated. The lower bound on confidence intervals reported here may be appropriate, conservative estimates to use when planning future research. In addition to validation of the sign-tracker/goal-tracker paradigm, the purpose of this study was to identify behaviors and neurological profiles associated with sign-tracking phenotypes in youth. The age range for this sample was selected for multiple reasons. First, the average age of onset for externalizing disorders is 11 years old and many externalizing symptoms emerge during this age range 2 . Second, symptoms of a range of psychiatric conditions (e.g., depression, anxiety, substance use) typically begin to emerge later in adolescence 59 , making pre-adolescence a prime target for mental health screening and development of preventative interventions. Further, the timing of prefrontal cortex development, responsible for decision-making and higher-order executive functions 31 , pre-adolescent youth may be more likely to engage in behaviors influenced by enhanced motivational drive 60,61 than adults, likely impacting the detection of sign-tracking or goal-tracking phenotypes in humans. And finally, we used this age range in an attempt to maximize the quality of neuroimaging data collection in youth.
Sign-and goal-tracking apparatus and paradigm development. The Pavlovian conditioning apparatus used is described in Joyner et al. (2018) and was built to mimic the animal model as closely as possible. The apparatus consisted of two solid-colored response boxes, built to look like building blocks and be appealing to youth but not inherently rewarding. The boxes were: (1) the CS box containing a lever, which illuminated and extended from the box, and (2) the US box containing a small metal tray into which the reward was dispensed. In the CS box, a linear actuator, consisting of a 12 V DC motor (Bühler Motor, GmbH, Nuremberg, Germany) and a worm drive, was used to extend and retract the lever. In the US box, a reward dispenser with infrared sentry (Med Associates, Inc, Fairfax, VT) was used to release the reward. A touch sensor, based on a field-effect transistor, was used to detect a participant's touches to the metal reward tray. The hardware system was operated via an Arduino UNO microcontroller (Arduino, LLC, Somerville, MA, www. ardui no. cc). The microcontroller was programmed to provide three output signals, controlling the lever movement, the lever light, and the reward dispenser. Two Arduino inputs were programmed to record the lever presses and the reward tray touches. The experimental protocol was implemented in custom software written in MATLAB 62 (MathWorks, Inc, Natick, MA, www. mathw orks. com). The code was run by a researcher on a laptop communicating with the Arduino via a USB connection. Response times (in ms) corresponding to all lever presses and reward tray touches in each trial were saved to a file. The CS and US boxes were spaced approximately 12 inches apart to reduce the likelihood of participants simultaneously engaging with both. The left/right positioning of the boxes was counterbalanced between participants to minimize lateral bias.
Response boxes were covered prior to each session and uncovered simultaneously for a demonstration that was given in counterbalanced order at the beginning of each session. To demonstrate the lever box, researchers guided participant attention to the box and manually commanded the program to extend and retract the lever. To demonstrate the reward box, researchers manually commanded the program to dispense one token. Participants were instructed that each token was worth $0.20 and that tokens were theirs to keep and exchange later for money or a prize. During demonstration, participants were encouraged to touch each box at least once and were instructed that during the session, they could interact with either box in any way and as much as they liked. Participants were then instructed to move to the neutrally marked location in between both boxes to begin the session.
The structure of the Pavlovian conditioning paradigm was primarily modeled after rodent studies described by Flagel et al. 32 . This paradigm consists of multiple response-independent trials during which a lever (CS) illuminates and extends, and upon retraction, a reward portion (US) is dispensed. A randomly selected inter-trial interval (ITI) follows each trial, after which the next trial begins. Each of these actions occurs for every trial regardless of subject input or response. In typical animal models 32 , conditioning sessions consist of 25 trials lasting 30-45 min and occur across multiple days. Modeled after Joyner et al. 30 , we condensed training into a single session to make this more feasible for our population and avoid participant burden (e.g., multiple trips to the lab). Within this session, we conducted four blocks of ten trials each, with the total session lasting approximately 20-30 min. Each trial consisted of the lever illuminating and extending for 8.3 s, and upon retraction, the token reward (US) was dispensed into the tray. The ITI period was programmed to last for a randomly selected time, either 8, 16, 24, or 32 s. Each block was followed by a "wiggle break" lasting up to 45 s, during which the child was given the opportunity to relax and reset before the next trial. This setup was intended to maximize participant attention and minimize fatigue, while retaining enough length in the session for associative learning to occur. In order to capture the number of and latency to contacts to each stimulus, we measured these behaviors in line with the traditional measurement of rodent sign-and goal-tracking. The MATLAB 62 program controlling the apparatus recorded the number and timing of contacts to the CS lever and US reward tray during CS presentation and ITI (Fig. 1).
We measured sign-tracking behaviors via a Pavlovian Conditioned Approach (PavCA) index based on previously used models in animal studies 33 derived from the number and timing of physical contacts to the CS and US during CS-presentation and ITI phases. This index was calculated for each trial and consists of the average of three measures: response bias, probability difference score, and latency difference score. Response bias is the probability of contacting the CS versus the US and calculated as: (lever contacts − reward contacts)/(lever contacts + reward contacts). Probability difference score is the probability of contacting the CS minus the probability of contacting the US and calculated as: (lever contacts/(lever contacts + reward contacts)) − (reward contacts/(lever contacts + reward contacts)). Finally, latency difference score is the latency to contact the US minus www.nature.com/scientificreports/ latency to contact the CS for each trial and calculated as: (latency to reward contacts in ms * 0.001 − latency to lever contacts in ms * 0.001)/8.3 s. To estimate the reliability of the PavCA measure, we used a two-way intraclass correlation between even and odd trials for each participant. The reliability was excellent (ICC = 0.97, F(39,39.7) = 56.7, p < 0.001, CI = 0.94 to 0.98).
To assess learning of conditioned responses to the CS, we examined behavioral responses to the CS using twosided linear mixed effects models with the factors time (Block 1-4), phase (CS, ITI), and phenotype (ST, non-ST) for lever-and reward-directed behaviors (Supplementary Fig. S2) and PavCA (Fig. 1c). These models included age and sex as covariates and a random intercept for subject. For post hoc tests we used estimated marginal means. We investigated behaviors both during the CS periods (as seen in animal models) and ITI periods because, unlike animal models, our sample displayed lever-directed behaviors, albeit low levels, during the ITI-period (i.e., when the lever was retracted; CS lever contacts, m = 4.06, sd = 5.73; ITI lever contacts, m = 0.59, sd = 1.53). This may indicate an effect of investigative or exploratory behaviors common during this developmental stage 63 or limited attentional capacity during the longer ITIs and will be an important methodological consideration in human translation of this paradigm in future studies. Further, because human subjects attempted to interact with the lever even during the ITI (i.e., when it was retracted), and because CS lever contacts correlated positively with ITI lever contacts (r = 0.59, p < 0.001; Supplementary Table S2), a normalized frequency score was calculated for lever-and reward-directed behaviors (contacts and probability) in order to adequately compare between CS and ITI periods. To normalize, we divided each measure by the length of their respective phase. A non-normalized score was used for PavCA scores in the main text.
When determining an appropriate reward to use as the US, we took several factors into consideration. First, pilot testing primary food rewards in this age group (chocolate candies, as in Joyner et al. 30 ) demonstrated very minimal incentive toward the reward. Therefore, in order to choose an item that would be (1) rewarding to the study population and (2) small and consistent in shape to both adequately dispense from the machine and reflect the physical characteristics of the candies used in Joyner et al. 30 , we ultimately decided to use colorful wooden beads. Participants were informed that these tokens were worth $0.20 cents each and could be exchanged for their choice of prizes or money (totaling $8) at the end of the session. The use of a secondary reinforcement rather than primary is a notable deviation from rodent models and Joyner and colleagues 30 , however, we believe the benefit of increased participant engagement and reward motivation made this justifiable.
Measures. Parent-reported psychopathology. Two parent-report measures that have been developmentally validated for 9-12-year-olds were used to identify youth psychopathology symptoms. First, the Early Adolescent Temperament Questionnaire (EATQ-R) 35 is a 62-item parent-report measure of youth temperament and selfregulation in 9-15-year-olds. Subscales include effortful control (measures of attention, inhibitory control, and activation control), surgency (measures of surgency, fear, and shyness), negative affect (measures of aggression, frustration, and depressive mood), and affiliativeness. Second, the Child Behavior Checklist (CBCL) 34 measures dimensional psychopathology and is normed by age, sex, and ethnicity. It is a parent-reported youth behavioral questionnaire that includes eight empirically-based subscales for anxious/depressed, withdrawn/depressed, somatic complaints, social problems, thought problems, attention problems, rule breaking, and aggression as well as DSM-oriented subscales for attention deficit-hyperactivity, affective, anxiety, somatic, conduct, and oppositional defiant problems.
Family demographics and functioning. Parents responded to demographic questions from the PhenX survey toolkit 64,65 including household income and 7 items addressing economic adversity and basic needs unaffordability. Additional environmental influences included in our analyses were markers of family environment including the Protective and Compensatory Experiences Scale (PACEs) 50 , a 10-item parent-report scale measuring factors contributing to resiliency in childhood including relationships and resources available to youth. Finally, parents reported on the family environment with the 90-item Family Environment Scale 66,67 . Items were true/false and subscales for cohesion, organization, recreational activities, and conflict were included.
Youth self-report and neurocognitive functioning. Youth reported impulsive and inhibitory behaviors using the Urgency, Premeditation, Perseverance, Sensation Seeking, Positive Urgency (UPPS-P), Impulsive Behavior Scale, and the behavioral inhibition system/ behavioral approach system (BIS/BAS) 68 scale. A modified version of the UPPS-P from PhenX for children was used 64,69 consisting of 20 self-report questions addressing youth impulsive behaviors including subscales for negative and positive urgency, lack of premeditation and perseverance, and sensation seeking. An abridged version of the BIS/BAS scale was used including 24 items with subscales for drive, fun seeking, reward responsiveness, and inhibition 64,68 . Youth neurocognitive functioning was measured using the National Institutes for Health Toolbox Neurocognitive Battery specified for ages 7-17 70,71 . All Toolbox tests were administered measuring executive functioning, episodic memory, language, processing speed, working memory, and attention. For analysis purposes we used the age-corrected composite scores of fluid and crystalized cognitive functioning.
Neuroimaging. Head motion prevention. To prevent and minimize motion: (1) participants watched an age-appropriate informational video explaining MRI safety and the importance of staying still; (2) prior to the scan session, participants completed motion compliance training in mock scanners using head motion detection/feedback; (3) the head was stabilized in head coils; and (4) the MR technologist modeled relaxation techniques. Youth vision was screened and MRI-safe corrective lenses were provided if necessary. 44 . Each trial began with a cue presented for 2 s indicating the trial type, with a pink circle indicating a potential gain of $5 or $0.20, a blue triangle indicating no gain or loss, and a yellow square indicating a potential loss of $5 or $0.20 so that there were 5 trial types: high-loss, low-loss, neutral, low-win, and high-win. A 1.5 to 4 s fixation followed the cue, and then a black target was presented. Participants were instructed to press a button while the target was on the screen, with target duration varying between 0.15 and 0.5 s. On gain trials, participants were rewarded for successfully hitting the target and would neither gain nor lose money for missing it. On loss trials, participants lost the indicated amount when they missed and neither gained nor lost when they hit the target. The outcome of each trial was presented immediately after each response and lasted 2 s minus the duration of the target. Target duration was initialized based on performance during a practice session completed outside the scanner and updated during scanning so that participants would succeed on approximately 60% of trials, leading to mean earnings of $20 with a maximum of $60. Each run contained 50 trials (10 of each trial type) and lasted 5 min 42 s (Supplementary Fig. S3). BOLD imaging during the MID task took place on two identical GE MR750 3 T scanners using multiband acquisition with an acceleration factor of 6 and the following parameters: 60 axial slices, TR/TE = 800/30 ms, FOV/slice = 216/2.4 mm, 90 × 90 matrix producing 2.4 mm isotropic voxels, 419 volumes for 5 min 35 s of scan time per run. Additionally, high-resolution structural images were obtained through a 3D sagittal T1-weighted magnetization-prepared rapid acquisition with gradient echo sequence (TR/TE = 6/2.92 ms, FOV/ slice = 256 × 256/1 mm, 208 sagittal slices).
fMRI data preprocessing. fMRI data were preprocessed using the Analysis of Functional Neuroimaging (AFNI, [URL]:// anfi. nimh. nih. gov) software 72 . Steps included removal of the first 10 volumes to allow for signal stabilization, despiking, slice timing correction, co-registration with the anatomical volume, motion correction, nonlinear warp to MNI space with resampling to 2 mm isotropic voxels, scaling to percent signal change, and application of a 4 mm Gaussian FWHM smoothing kernel. A general linear model was applied with regressors for the six motion parameters, three polynomial terms, and 2-s block regressors for each of the five trial types. Censoring was applied at the regression step so that any TRs with the Euclidean norm of the six motion parameter derivatives greater than 0.3 were removed, along with TRs where greater than 10 percent of brain voxels were outliers (estimated with 3dToutcount). The estimated beta coefficients from this single subject analysis were taken to the group level and are interpreted in terms of percent signal change for each condition.
Statistical analysis. Analyses were conducted in the R System for Statistical Computing 45 . Feasibility of the sign-and goal-tracking task was assessed in multiple ways. First, frequency counts were used to identify qualitative and quantitative levels of child engagement in the paradigm and how well the participants tolerated the task. Second, we assessed behavioral responses during CS and ITI periods using means and standard deviations of each behavior by block. Additionally, Pearson r correlations using the psych package in R 73 were used within behaviors to verify measurement of lever-or reward-directed behaviors. Blocks 1 and 2 were identified as a training phase, therefore the behaviors from Blocks 3 and 4 were averaged for each PavCA index score. To classify categorical phenotypic groups, we used a PavCA value of 0.5 or greater to define sign-tracking and less then −0.5 to define goal-tracking, consistent with categorization used in animal models 74 . Behavioral differences by phenotype were assessed using two-sided Welch two-sample t-tests in the R package stats 45 . Finally, we assessed the demonstration of learning a conditioned response to the CS by examining behavioral responses to the CS over all four Blocks (40 trials total). We used linear mixed effects models using the R package lme4 75 to test for three-way interactions between time (Block 1-4), phase (CS, ITI), and phenotype (ST, non-ST) for lever-and reward-directed behaviors. The R package emmeans 76 was used to assess planned post hoc comparisons. Error bars in figures represent standard error of the mean calculated in the Rmisc 77 package in R.
Recent research using animal models has shown that ITI duration may also impact the likelihood of displaying each CR and sign-tracking behavior appears to be more likely during a longer ITI periods 78 due to a weakened association between contacting the location of the US and receiving a reward. Additionally, the responses during ITI periods in our sample were relatively high compared to typical animal behaviors, therefore, to adequately compare between CS and ITI periods, scores were normalized by dividing each score by the length of each respective phase (8 s for CS, and 8, 16, 24, or 32 s for each ITI).
Behavioral outcome measures. We conducted Welsh independent samples t-tests using the R package stats to examine phenotypic differences in symptoms and environmental variables. Tests were p-value corrected for multiple comparisons using the FDR corrections. All variables were tested for normality and transformed using the optLog 79 package in R where necessary. CBCL subscales were all log transformed. Whole brain analyses. We performed a whole-brain voxelwise linear mixed effects analysis using AFNI's 3dLME 80 with fixed effects for group, condition, age, and sex, a group by condition interaction, and a random intercept for subject. We followed this with planned contrasts investigating a group (ST, non-ST) by condition (high-win vs neutral or high-loss vs neutral) interaction. 3dFWHMx (with the newer -acf option) was used to estimate the smoothness of the residuals of the group model, and this smoothness was used along with 3dClust-Sim to perform cluster-wise correction. The ACF parameters were estimated to be 0.09, 6.63, and 1.36 which indicates an effective FWHM smoothness of 2.8 mm. Significant clusters are reported using a voxelwise p-value threshold of 0.005 and α < 0.05 at the cluster level (N = 11.73 voxels). www.nature.com/scientificreports/ with data processing and technical assistance, and E. Pribil for data collection. The funders had no role in the study design, data collection or analysis, decision to publish, or preparation of the manuscript.
==
Domain: Psychology Biology Medicine
|
[{"added": "2022-02-01T16:08:32.912Z", "created": "2022-01-29T00:00:00.000", "id": "246437681", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": null, "oa_url": null, "pdf_hash": "3e76c608ac9496979ab389bdbbfa2421dd9816a6", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:32020", "s2fieldsofstudy": ["Psychology", "Biology"], "sha1": "24aa8db9fb77840700ff86175a307b687451cb3c", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Reward-Related Dorsal Striatal Activity Differences between Former and Current Cocaine Dependent Individuals during an Interactive Competitive Game
Cocaine addiction is characterized by impulsivity, impaired social relationships, and abnormal mesocorticolimbic reward processing, but their interrelationships relative to stages of cocaine addiction are unclear. We assessed blood-oxygenation-level dependent (BOLD) signal in ventral and dorsal striatum during functional magnetic resonance imaging (fMRI) in current (CCD; n = 30) and former (FCD; n = 28) cocaine dependent subjects as well as healthy control (HC; n = 31) subjects while playing an interactive competitive Domino game involving risk-taking and reward/punishment processing. Out-of-scanner impulsivity-related measures were also collected. Although both FCD and CCD subjects scored significantly higher on impulsivity-related measures than did HC subjects, only FCD subjects had differences in striatal activation, specifically showing hypoactivation during their response to gains versus losses in right dorsal caudate, a brain region linked to habituation, cocaine craving and addiction maintenance. Right caudate activity in FCD subjects also correlated negatively with impulsivity-related measures of self-reported compulsivity and sensitivity to reward. These findings suggest that remitted cocaine dependence is associated with striatal dysfunction during social reward processing in a manner linked to compulsivity and reward sensitivity measures. Future research should investigate the extent to which such differences might reflect underlying vulnerabilities linked to cocaine-using propensities (e.g., relapses).
Introduction
Deficits in impulse control and reward processing are hypothesized to initiate and sustain cocaine dependence [1,2,3,4], which is characterized by favoring immediate rewards of drug use over delayed non-drug rewards, despite potential negative consequenc-es [5]. Mesocorticolimbic circuits, involving the dopaminergically innervated ventral and dorsal striatum as well as orbitofrontal and anterior cingulate cortices, are crucially involved in reward processing, and dysregulation in these circuits is implicated in both impulsivity and cocaine dependence [1,6,7,8,9,10].
Abuse of substances, including cocaine [11] and alcohol [12,13,14], has been associated with reduced ventral striatal activity during non-drug reward anticipation or receipt, but the number of such studies is small and the origins of such hypoactivity are still poorly defined. One explanation for diminished mesocorticolimbic activation involves the rewarddeficiency syndrome (RDS) hypothesis [15], which conjectures that drugs of abuse, due to their potent dopaminergic effects, normalize ventral striatal dopamine levels, whereas non-drug related rewards fail to do so, leading RDS individuals to seek cocaine or other abused drugs. Long-term, chronic cocaine abuse has been shown, however, to exacerbate underlying nondrug reward response deficiencies, through remodeling of neural circuitry [6,16,17,18]. This so-called 'hijacking' of the reward system leads abusers to attribute even greater value to drugrelated rewards at the expense of non-drug rewards [19,20].
Two recent studies found, however, in apparent contradiction to the RDS hypothesis, greater ventral striatal activity in cocainedependent [21] or substance-dependent [22] individuals, when compared with healthy subjects, during non-drug reward anticipation or receipt. These studies lend support to the alternative 'impulsivity hypothesis' [23], related to opponent process theory, and which contrasts with the RDS hypothesis in that it predicts greater sensitivity even to non-drug rewards, along with an insensitivity to punishments [22,24,25]. Such discrepancies in recent studies of substance abuse and neural response to rewards clearly indicate that more research is needed to elucidate the effects of cocaine-abuse on the reward system, and the ventral striatum in particular, during non-drug reward anticipation and receipt.
As drug use and other addictive behaviors become habitual, striatal involvement may shift from ventral to dorsal [5,26,27,28,29]. Dorsal striatal-related networks are implicated in habitual behaviors [30], including cue-driven drug use and craving, and are theorized to contribute to compulsive cocaine use and relapse [26,27,28,31,32]. In support of this concept, two recent functional magnetic resonance imaging (fMRI) studies found that increased dorsal striatal activity was directly related to cocaine craving, either induced by psychological stress (in abstinent cocaine-dependent individuals in treatment) [33] or cocaine imagery (in actively abusing cocaine-dependent individuals) [5,10]. Therefore, although cocaine use may be initiated by factors including trait impulsivity and supported by cocaine's rewarding effects, habitual use in the later stages of addiction may depend less on the experience of cocaine reward and instead be both impulsively and compulsively driven, via ventral and dorsal striatum respectively, by cocaine cues that previously signaled reward [5,6].
Impulsivity (e.g., delay discounting and impaired inhibition) and theoretically related constructs (e.g., compulsivity, risktaking and sensation-seeking) can be quantified using self-report or laboratory-based measurements [34,35]. Three recent studies demonstrated that impulsivity-related measures were associated with diminished ventral striatal activation during reward anticipation in individuals with pathological gambling [36], in detoxified alcohol-dependent subjects [12], and in nonalcoholic individuals who were family history positive (FHP) for alcoholism [14]. In the third study, impulsivity and related constructs were assessed using a principal-component-based factor analysis of five self-report and two laboratory-based impulsivity-related measures [37], which revealed that the multifaceted nature of impulsivity and related constructs was optimally described by five factors. One of the five factors loaded most strongly on two self-report measures (i.e., the Sensitivity to Punishment and Sensitivity to Reward Questionnaire and the Padua Inventory assessing compulsivity [38]) and was found to correlate negatively with ventral striatal activity. Such a factorial approach should provide a means to behaviorally quantify impulsivity domains in users of other addictive drugs as well, including cocaine, and explore correlates with brain function. However, a correlational analysis of a broad range of impulsivity-related measures with reward circuitry function in both former and current cocaine dependent individuals has not yet been reported.
Examining reward response in the context of risk-taking is important, as increased risk-taking behavior is commonly observed in individuals with the greatest likelihood to develop substance abuse or dependence [23]. The Domino fMRI task was originally introduced to examine the role of the amygdala in signaling prospective negative outcomes during risk-taking [39,40], but has also been shown to strongly activate the reward system, particularly the ventral striatum, as a result of gains during gameplay [40,41]. Therefore, in studies of substancedependent individuals, the Domino task is uniquely suited to examine both dysfunctional brain reward processing and behavioral risk-taking, as well as their relationships with the five impulsivity-related constructs derived from the factor analysis described above. Examination of reward system dysfunction in individuals in long-term cocaine abstinence, to our knowledge, has not yet been attempted. The inclusion of FCD individuals, who have been cocaine abstinent for at least six months, in addition to CCD individuals and healthy controls, is important because chronic cocaine use can induce changes in brain structure and function [27,42]. Because the extent to which neural and behavioral recovery occurs with sustained abstinence is unclear, inclusion of FCD subjects permits direct comparisons with CCD individuals and might demonstrate in these subjects at least partial recovery of brain reward functioning to pre-morbid status.
Our goals in this study were as follows: 1) examine striatal activity, both ventral and dorsal, during reward receipt in both former and current cocaine-dependent individuals versus healthy controls, 2) assess risk-taking behavior in these same groups during Domino gameplay and 3) find relationships between neural reward activity and behavioral risk-taking with five impulsivity-related factors derived from a factor analysis study. In keeping with these goals, we hypothesized that both FCD and CCD subjects as compared with HC subjects would exhibit dysfunctional reward processing, assessed using fMRI during Domino gameplay. In particular, in accordance with the RDS hypothesis, we predicted that both ventral and dorsal striatum would show less activation in response to gains versus losses during gameplay in CCD and FCD compared with HC. We predicted, however, that FCD would exhibit less hypoactivity in these same brain regions than CCD, due to at least partial recovery of reward neurocircuitry to baseline. We also anticipated that both CCD and FCD groups, due to greater trait impulsivity, would exhibit greater risk-taking behavior during Domino gameplay than HC individuals, but the FCD group would exhibit less risk-taking than the CCD group, due to reduced impulsivity from cessation of chronic drug use. Lastly, we determined whether activity in these same striatal regions correlated with subjects' impulsivity-related measures, hypothesizing that gain-related ventral striatal activation would correlate inversely with impulsivity-related factors across groups, based on findings in previous studies [12,14,36].
Ethics Statement
The study was approved by the Hartford Hospital and Yale University Institutional Review Boards and carried out at Hartford Hospital Institute of Living. All study participants provided written informed consent after the study had been fully explained to them. Participants were paid for participating in the imaging study.
Study Participants
Participants included n = 36 CCD, n = 28 FCD and n = 33 HC subjects. After removing subjects for excessive reaction times (discussed later), 30 CCD, 28 FCD and 31 HC subjects remained. Table 1 (top) shows demographic data; groups were matched for age, gender and IQ. All subjects were right-handed. Participants were recruited by word-of-mouth, flyers, newspaper, online advertisement and drug abuse programs.
All FCD and CCD subjects met DSM-IV criteria for dependence (for FCD, during previous use) based on initial screening. For inclusion into the study, FCD subjects were required to have ceased all cocaine use at least 6 months prior to the beginning of the study. Abstinence in FCD individuals was confirmed by self-report, urine toxicology screening, and where available, information from the substance abuse long-term followup groups that a proportion of subjects attended at the Institute of Living. For CCD subjects, participation required a positive cocaine urine test (last use ,72 hours) on the day of screening. Frequency of self-reported use by CCD subjects in the 30 days prior to enrollment was as follows: ,2 days: 3 subjects; 2-5 days: 6 subjects; 6-12 days: 9 subjects; 13 or more days: 10 subjects. Cocaine use data were not available for two CCD and one FCD subjects. Table 1 (middle and bottom) indicates substance abuse and Axis-I diagnoses, respectively.
Exclusion criteria for all subjects included: current nonsubstance Axis I disorders (e.g., schizophrenia), as assessed by structured clinical interview (SCID; [43]), current/past major neurological/physical illness, history of head trauma causing loss of consciousness, metallic objects in the body and estimated fullscale WAIS IQ ,70. FCD were excluded for positive urine drug screen on the day of the scan and HC excluded for meeting DSM-IV criteria for substance abuse or dependence, except nicotine, or if their urine tested positive for recreational drugs on the day of testing. Nine CCD subjects had comorbid opiate use due to recruitment of cocaine dependent subjects from an outpatient drug abuse program (see Table 1). Cocaine use data were assessed using a self-report substance abuse questionnaire based on timeline follow-back methods [44]. Current and prior cocaine use data are in Table 1.
Domino Task
The Domino task is an event-related, two-player competitive computerized game modified from Kahn et al [39] described previously [41]. Figure 1 provides a graphical overview of the Domino task.
Participants practiced the game outside the scanner prior to scanning. Scanning began when the experimenter recognized that participants understood the game's rules. A thorough debriefing was conducted immediately after scanning, where participants were asked about their emotions and strategies while playing. Open-ended questions and a Likert scale questionnaire were used where participants rated responses from 1 (least) to 5 (greatest) for agreement with statements (see example in ''Results'' section).
To characterize players' game decisions, a Risk Index was defined as the ratio between the number of times a player chose a non-matching chip, only when a choice between non-matching and matching chips was available, to the total number of chips played (again, only when a choice between non-matching and matching chips was available).
The scanned subject is the player while a computer randomly generates the opponent's responses. Subjects were told, however, that they were playing against a human opponent. Thus, from their perspective, subjects were playing in an interpersonal competitive context. Each game contains a pool of 28 dominolike game pieces. At the beginning of each game, 12 random domino chips are assigned to the player's bank (shown face up on the computer screen), four undisclosed chips are randomly assigned to the opponent's bank and a randomly chosen Master chip shows face-up on the board. The remaining 11 chips are not used. Each of the player's chips is either a matching chip (has one of the two numbers on the Master chip) or a non-matching chip (has neither number on Master chip).
The player's goal is to discard all assigned chips before the game ends (4 min) and, if they attain this goal, they are awarded $10, paid in cash at the end of the session. Thus, for the purposes of this study, during each round, discarding chips will be referred to as 'gains', while acquiring chips will be referred to as a 'losses.' Furthermore, as will be explained below, playing a matching chip is considered a 'safe' move, while playing a non-matching chip is considered a 'risky' move or 'bluff'. It is only possible to win and to collect the resulting monetary bonus by occasionally 'bluffing' (i.e., playing a non-matching chip).
During each round of the game, the player selects a chip to play, places it face down adjacent to the Master chip and awaits the opponent's response. The opponent can either challenge the player by asking him/her to reveal the chosen chip, or not challenge, allowing him/her to move on to the next round. Each round progresses according to the following commands, presented to the player both visually and aurally: (a) Choose instructs the subject to mentally select a chip to be played. The player can decide to pick either a matching or a non-matching chip. This is the 'Decision-making' interval; (b) Ready instructs the subject to move a cursor (using his/her dominant hand) to their chosen chip. This is the 'Ready' interval. These first two intervals each last 4 s; (c) Go instructs the player to press a button, as quickly as possible (note: the Go event duration is the subject's reaction time (RT)), to put the chosen chip face down next to the Master chip.
The player then awaits the opponent's response. This is the 'Anticipation of Outcome' interval with a 'jittered' duration of either 3.4, 5.4 or 7.4 s (5.462.0 s) [45]. The opponent's response is either (d) Show or No-Show. The former command exposes the player's selected chip, revealing whether they played 'safe' or 'bluffed', while the latter allows the player to discard his/her chip without exposing its value. This is the 'Response to Outcome' interval and its duration is also jittered (5.462.0 s). The next round of the game starts when the 'Response to Outcome' interval ends. The Choose command is then again presented to the player.
Based on the player's choice and opponent's response there are four possible consequences per round revealed during the 'Response to Outcome' interval: (1) Show Matching chip: a matching chip is exposed and the player is rewarded by discarding the selected chip plus one additional chip randomly chosen from his/her bank. This is an absolute gain. (2) Show Non-Matching chip: a non-matching chip is exposed, and the player is punished by acquiring the chip just played, plus two additional chips (from either the opponent's bank or his/her previously discarded chips, thus not chosen by the player), for a total of three chips. This is an absolute loss; (3) No-Show of a Non-Match chip: a non-matching chip remains unexposed and then discarded, so the players successfully 'bluff', i.e., get away with a non-matching choice. This is a relative gain; and (4) No-Show of a Matching chip: a matching chip is not exposed and then discarded, so the player is relatively punished as he/she could have discarded another chip. This is a relative loss.
Rounds continue until the players win (by discarding all of their chips) or lose, when either 240 seconds have elapsed, or they have acquired all 16 available chips from the bank and the board. Participants played Domino games over two scan runs of 15 min each for a total of 8.3960.61 games (Mean 6 SD). Participants were told they were playing against the experiment-er, whom they met prior to the scan, outside the scanner. The experimenter talked to the participant after each run making competitive comments about the games just played (such as ''you really got me this time …''). To ensure that players were engaged in the game and believed that winning was possible, if they did not win during the first run, the first game of the second run was not automated and the experimenter ''threw'' the game, ensuring that the player won. Twenty-nine of the total of 89 players played a non-automated game (HC, 13 games; FCD, 6 games, CCD, 10 games); these games were excluded from the analysis. Games shorter than one minute (5.62% of all games) were also not analyzed.
Other Axis-I C P C P C P
Behavioral Data Analysis
Likert scale score answers to the Domino debriefing statements were analyzed using non-parametric tests (Wilcoxon rank sum, Kruskal-Wallis) where appropriate, except for the case when determining if a given group response to a statement was statistically significantly greater than the middle score of 3. Here, a parametric one-sample t-test was used. SPSS TM software (v15, SPSS Inc., Chicago, IL) was used for all behavioral statistical analyses.
Subject Exclusion
Subjects were excluded from analysis for mean reaction times (RTs) (during the Go event) in excess of 3.4 seconds, a value chosen based on the minimum stimulus duration of the 'Anticipation of Outcome' interval, is jittered (5.462.0 seconds). On the basis of excessive mean RTs, six CCD and two HC subjects were excluded.
Impulsivity-related Measures
In a previous study that included many current study subjects [37], domains for multiple behavioral and self-report measures of impulsivity and related constructs were examined in At-Risk/ Addiction subjects (individuals either family history positive for alcohol dependence, or current/former cocaine dependent) as well as healthy controls (14 CCD, 9 FCD and 21 HC from that study participated in the current study). Briefly, five widely-used, reliable and valid self-report questionnaires were used, and were described in that study: (i) the Behavioral Inhibition/Activation System (BIS/BAS) [46], (ii) the Barratt Impulsivity Scale (BIS-11) [47], (iii) the Sensitivity to Punishment and Sensitivity to Reward Questionnaire (SPSRQ) [48], (iv) the Sensation Seeking Scale (SSS Form V) [49] and (v) the Padua Inventory [50], plus two computer-based behavioral laboratory tests: (i) the Balloon Analog Risk Task (BART) [51] and (ii) the Experiential Discounting Task (EDT) [52].
In our current study, we used the previously identified Factor structure [37], i.e., we did not perform a new factor analysis, but rather used the five factor structure implicit in the Component Score Coefficient (CSC) Matrix calculated in the original Factor Analysis using Statistical Package for Social Sciences (SPSS), version 15.0 (SPSS Inc., Chicago, Illinois). Subjects taking the impulsivityrelated test battery since the previous study, including subjects in this study, were added to the original database of n = 176 subjects and then new Z-scores for each impulsivity-related test were computed for all subjects (final n = 246). Updated Z-scores were multiplied by the CSC matrix to calculate new factor scores for our HC, FCD and CCD subjects.
Functional MRI Acquisition
Blood oxygenation level dependent (BOLD) data were collected with a T2*-weighted echo planar imaging (EPI) sequence (TR/ TE = 1860/27 msec, Flip angle = 70u, Field of view = 22 cm with a 64664 acquisition matrix) using a Siemens Allegra 3 Tesla scanner. Thirty-six contiguous axial functional slices of 3 mm thickness with 1 mm gap were acquired, yielding 3.463.464.0 mm voxels. Overall, 492 images were acquired during each run, including six 'dummy' images at the beginning to allow global image intensity to reach equilibrium, which were excluded from data analysis.
fMRI Data
Preprocessing. Imaging data were preprocessed using SPM2 (Wellcome Department of Cognitive Neurology, London, UK). Each individual's data set was realigned to the first 'non-dummy' T2* image using the INRIAlign toolbox (A. Roche, EPIDAURE Group; [URL]) to compensate for any subject head movement. Movement parameters for each subject were then screened for excess head movement (.4 mm). The resulting images were spatially normalized to the Montreal Neurological Institute standard template [53] and spatially smoothed with a 9 mm isotropic (FWHM) Gaussian kernel. A high-pass filter with a cutoff of 128 s was applied to correct for EPI signal low-frequency drift.
Events and Regressors
Functional MRI data were analyzed using a general linear model (GLM) approach using SPM5 (Wellcome Department of Cognitive Neurology, London, UK). As described previously [41], from the four game intervals described in detail in the online supplement, we created ten first-level (subject-level) regressors: (1) choose-match and choose-nonmatch from the 'Decision-making' interval; (2) ready from the 'Ready' interval; (3) pick-match and picknonmatch from the 'Anticipation of Outcome' interval; (4) showmatch, show-nonmatch, noshow-match and noshow-nonmatch from the 'Response to Outcome' interval, each regressor corresponding to the four consequences; (5) and finally a misc regressor for events of non-interest including Go events and between-game events, during which the subject learned whether they won or lost the last game and then waited for the next game to being. The misc regressor also included all events occurring during games of less than oneminute duration. These events were not analyzed.
Regressors were modeled as boxcar functions convolved with the SPM5 canonical hemodynamic response function (HRF) and included HRF temporal derivatives. Regressors also included the six movement parameters (translation: x, y and z and rotation: pitch, roll and yaw).
In a fixed-effects first-level analysis, individual statistical parametric maps and contrast images were calculated and were composed of contrasts from the four 'Response to Outcome' interval regressors. These include the Gain contrast, a linear combination of the show-match and noshow-nonmatch regressors compared with an implicit baseline, the Loss contrast, a linear combination of the show-nonmatch and noshow-match regressors compared with an implicit baseline, and finally the Gain-Loss difference contrast.
Statistical Analyses
We delineated the reward network using a random-effects second-level analysis one-sample t-test of the Gain-Loss contrast across all subjects (n = 89). We thresholded the resulting statistical parametric map at p,0.05 family-wise error (FWE) rate wholebrain corrected. We used the across-all-subjects Gain-Loss contrast map as an inclusive mask in all subsequent analyses, such that they examined group differences within the brain reward network only. We hereafter refer to this inclusive mask as the Reward mask, and the brain regions it defines as the Reward network.
To explore group differences in the Gain-Loss contrast we performed a random-effects repeated-measures between-subjects analysis of variance (ANOVA) in which we compared the three HC, FCD and CCD groups. We were specifically interested in between-group differences in Gain-Loss activation in targeted regions of interest (ROIs), ventral and dorsal striatum, according to our hypotheses. We first applied the Reward mask, as described previously, to the results from the one-way ANOVA betweengroup Gain-Loss contrast comparison. The statistical threshold for clusters occurring in ROIs was then set to p,0.05 uncorrected (due to an anticipated loss of power from a between-group ANOVA second-level analysis) with a minimum cluster size of k = 10 contiguous voxels (.270 mm 3 ). The SPM5 toolbox 'rfxplot' was used to calculate effects sizes for the random effects group SPM5 analyses and was also used to create subsequent effect size bar graphs [54].
Correlation of fMRI with Impulsivity-related Measures
To determine the relationship between impulsivity-related measures and reward-related brain activity, we performed a random-effects multiple regression analysis in SPM5 to obtain the correlation of each subject's impulsivity-related factor score with the corresponding Gain-Loss contrast at each voxel. Only impulsivity-related factors (from the set of five factors described earlier) that were significantly different among the three groups were subject to correlation analysis.
As with the between-group ANOVA analysis, we first masked the multiple regression analysis using the Reward mask. We then conducted a between-group ANOVA on the multiple regression results to determine clusters with significant between-group differences for each impulsivity-related measure versus Gain-Loss contrast correlation. As with main-effect ANOVA described previously, we present only those between-group ANOVA correlation clusters occurring in ROIs, and set the statistical threshold to p,0.05 uncorrected with a minimum cluster size of k = 10 voxels.
Post hoc analyses were conducted to determine which group(s) differed in correlation between subjects' impulsivity-related factor scores and corresponding Gain-Loss contrast. Statistical thresholds for post-hoc regression, performed on each group separately, were set at q,0.05 false-discovery rate (FDR) corrected (restricted to those voxels included in the Reward network) with a minimum cluster size of k = 10 contiguous voxels. Regression data for presentation for each group, as before, were further restricted to voxels in clusters located in ROIs. Data were obtained from the mean Gain-Loss contrast value of all voxels within a 5 mm radius of the peak voxel and were extracted using the eigenvariate option in SPM5.
Domino debriefing. Four statements on the Likert scale questionnaire were intended to reveal subject's emotional reactions to the four possible outcomes of each move during the 'Response to Outcome' interval (see Table 2). Kruskal-Wallis tests indicated no differences among the three groups in mean responses to any statement. One-sample t-tests showed that, for all three groups, Likert scale responses to both absolute and relative gains (''I felt glad when …'') were statistically significant with regard to agreement with the statements (i.e., mean response .3) while responses to both absolute and relative losses (''I felt unhappy when …'') were not (see Table 2), suggesting that subjects were more motivated by gains than losses, consistent with previous results [41].
For the HC, FCD and CCD groups, paired Wilcoxon signedrank tests comparing subjects' responses to statements regarding absolute versus relative gains (i.e., AS1 versus AS2), and absolute versus relative losses (i.e., AS3 versus AS4), showed no significant differences (AS1 vs. AS2, P.0.100 and AS3 vs. AS4, P.0.100, for all three groups; see Table 2 for AS definitions). Therefore, we concluded that subjects did not perceive absolute gain and loss events as more emotionally salient than the relative events. For this reason, in our fMRI GLM analyses, we grouped the relative and absolute gain event regressors together as the contrast 'Gain', and, likewise, we grouped relative and absolute losses event regressors together as the contrast 'Loss'.
Risk behavior. A two-way ANOVA (364) of Risk-Index with the two factors Group (i.e., the three groups, HC, FCD and CCD) and Time (elapsed minutes into game, binned into four one minute intervals), respectively, revealed a significant main effect of Time (F = 13.802, df = 3,258, P,0.0001). Players tended to 'bluff' their opponent more towards the game end than beginning. There was no significant effect of Group (F = 0.723, df = 2,86, P = 0.488) and no significant interaction between Group and Time (F = 0.539, df = 6,258, P = 0.778).
fMRI Analyses
We focused fMRI analyses exclusively on brain activity during the 'Response to Outcome' interval, both for the overall group analysis and for between-group comparisons, with the principal contrast of interest being Gain-Loss.
One-sample t-test:
Across-all-subjects reward network. A one-sample t-test for the Gain-Loss contrast across all subjects from the three groups (n = 89), showed strong activity in reward-related brain regions including bilateral ventral striatum, right dorsal striatum (caudate) and left and right lateral orbitofrontal cortex (see Figure 2, p,0.05, FWE whole-brain corrected). Table 3, top provides the locations (MNI coordinates) and t-scores for each region of significant activity. One-way ANOVA: Main effect of group. The masked oneway ANOVA analysis of the between-group differences for the Gain-Loss contrast yielded only one brain region with a group difference in activity, an ROI cluster (k = 18 voxels) in the right dorsal caudate (p,0.05 uncorrected, minimum cluster size k = 10; see Figure 3, panel A). Table 3, bottom, provides the location (MNI coordinates) and statistics for the right dorsal caudate region of significant between-group difference in activity. Effect sizes for Gain, Loss, and for Gain-Loss, for each of the three groups at the peak main effect of Group activation in the right dorsal caudate are shown in Figure 3, revealed several clusters within the Reward network at p,0.05 uncorrected, minimum cluster size, k = 10 voxels (see Table 4). One cluster was located in an ROI, the right dorsal caudate, as shown by the block arrow in Figure 4, panel A, at essentially the same location as the right dorsal caudate cluster in the betweengroup ANOVA main effect shown in Figure 3, panel A (peak cluster voxels at x,y,z = 15,18,0 vs. 18,18,9).
Impulsivity-related scores:
Post-hoc correlation analysis. A post-hoc analysis of the between-group ANOVA multiple regression results indicated that FCD was significantly different than HC and CCD. Only FCD had surviving clusters within the Reward network at a threshold of q,0.05 FDR corrected (minimum cluster size k = 10 voxels). All correlation clusters within the Reward network for the FCD group, including those not located in ROIs, are described in Table 5. The ROI cluster in the right dorsal caudate (see Figure 4, panel B, coronal slice y = +18 mm), in particular, overlaps the same right dorsal caudate ROI cluster found in the main effect of Gain-Loss (Figure 3, panel A). In this right dorsal caudate ROI cluster, FCD Factor 2 impulsivity scores were significantly negatively correlated with Gain-Loss effect size (R = 20.641, p = 0.0002; 5 mm radius sphere centered at x,y,z = 18,18,0) (see plot, Figure 4, panel C). Additional findings. No correlation was found between abstinence durations for FCD subjects with Gain-Loss effect size in any ROI. Additionally, we found no difference in HC, CCD and FCD subjects' risk-taking behavior during gameplay as measured using the Risk Index.
Prior psychiatric disorders: Post hoc analysis. To determine if prior psychiatric disorders in the FCD and CCD groups had an impact on our findings, we repeated our analyses excluding the eight FCD and three CCD subjects who had prior psychiatric diagnoses of mood disorders or PTSD. There were, however, no qualitative differences between our original analysis and the post hoc analysis. The post hoc analysis still revealed relative hypoactivity for the FCD group in the right dorsal caudate (peak voxel at x,y,z = 18,18,9; post hoc vs. original: F = 3.73 (df A. Statistical parametric F-maps (sagittal, coronal and axial) of the Gain-Loss contrast for one-way ANOVA between-group main effect (masked with the Reward mask). Crosshairs overlaid on brain slices are at located at x,y,z = 18, 18,9 (peak voxel). Glass brain at top right shows that the cluster in the right dorsal caudate is only surviving cluster. Threshold was set at p,0.05 uncorrected, minimum cluster size k = 10 voxels. B. Effect sizes for Gain (red), Loss (blue) and Gain-Loss (green) contrasts for HC, FCD and CCD groups at x,y,z = 18,18,9. Black bar represents standard error of the mean.
Discussion
Our principal finding was that while engaged in the Domino task, involving risk-reward decision-making, FCD, CCD and HC subjects showed largely similar behavioral and brain responses. However, a neural between-group difference was found in the dorsal caudate, such that FCD subjects compared to HC and CCD showed reduced BOLD activation during response to gains. Interestingly, in the same region (i.e., right dorsal caudate), only FCD subjects' Gain-Loss activity correlated inversely with their scores on impulsivity-related Factor 2, 'Self-reported compulsivity and sensitivity to reward and punishments.' Scores on this factor were significantly greater in FCD and CCD than in HC. Gain-Loss activity in bilateral ventral striatum also correlated inversely with impulsivity-related Factor 2 for FCD subjects, but not CCD or HC subjects. Therefore, contrary to our hypothesis, self-reported impulsivity did not relate to ventral striatal activation in CCD. Also contrary to our hypothesis, we found no difference between groups in risk-taking behavior as assessed by subjects' Risk-Index scores.
We predicted that, in accordance with the reward deficiency syndrome (RDS) hypothesis, both ventral and dorsal striatal activity would be reduced in both CCD and FCD groups, with CCD showing the greatest reductions in ventral striatal activity. We did not find, however, any significant activation differences between HC and CCD for the Gain-Loss contrast in any brain region, including the ventral striatum, in contrast to both the RDS and impulsivity hypotheses. We also found no significant ventral striatal Gain-Loss activity differences between HC and FCD. It is important to note that although much research on drug abuse has focused on the ventral striatum due to its suggested involvement in reward processing and responses to abused drugs [6,55], such efforts have yielded conflicting results in comparing drug users to healthy controls during non-drug reward anticipation and receipt, with some studies showing ventral striatal hypoactivity [11,12,13] while others reveal either no differences or ventral striatal hyperactivity in drug users [21,22,56]. Here, we present two possible explanations for such conflicting results in these non-drug reward-related studies of substance abuse might include study differences in: 1) subject populations and 2) fMRI tasks.
With regard to differences in subject populations, some of the above-cited studies examined subjects primarily dependent on drugs other than cocaine, including alcohol [12,13,22] or marijuana [56]. Clearly, comparison of such studies with fMRI studies of cocaine-dependent individuals, such as our study, presents significant difficulties. Additionally, the cocaine-dependent participants in the studies of Asensio et al. [11] and Jia et al. [21] were recruited from treatment centers and therefore were actively seeking cocaine abstinence, whereas many of our CCD participants were not. Thus, participants in these two prior studies of cocaine-dependence would, in terms of recentness of cocaine use and/or withdrawal, lie somewhere between the CCD and FCD groups in our study. Lastly, etiological heterogeneity in cocaine dependence might be a factor, such that in some individuals, reward-system deficiency might best explain the initial motivation to use cocaine (i.e., the RDS hypothesis), while in other individuals, the initial drive to use cocaine might arise more from risk-taking and/or sensation-seeking personality traits (i.e., the 'impulsivity hypothesis'). According to this theory, the two groups would show opposite changes in ventral striatal activity in response to non-drug rewards. Studies of cocaine dependence thus far might have included individuals of both etiologies, perhaps leading to seemingly conflicting findings.
With regard to differences in fMRI tasks, the Asensio et al. study involved activation of the reward system with erotic stimuli, and therefore was not only non-drug, but also non-monetary based, whereas the Jia et al. study used a monetary-based paradigm, the Monetary Incentive Delay task (MIDT). Similar to the MIDT, our Domino task involves monetary rewards, but the reward is received only at the end of several games of up to four minutes duration, and only if the game is won. Furthermore, in our study brain activity was measured during domino chip acquisition or disposal during ongoing gameplay, not at the end of the game. Therefore, the gains and losses in the Domino task represent a level of abstraction removed from the already abstract monetary gains and losses in the MIDT, which, in turn, contrasts with fMRI reward-related studies that use ''concrete'' non-drug rewards, such as food (juice) [57,58,59] or sex (erotic stimuli) [11,60]. Therefore, despite studies reporting deleterious consequences of chronic cocaine use on the mesocorticolimbic reward system [61,62,63], and due to noteworthy inconsistencies in findings thus far, it remains unclear if cocaine-dependent individuals in a given study sample will demonstrate ventral striatal hypoactivity or hyperactivity during anticipation or receipt of all types of non-drug rewards (e.g., abstract or concrete rewards), or only under specific circumstances (e.g., during early withdrawal, in the context of continued cocaine abuse, etc.). Our own findings show no differences between CCD and HC individuals, further demonstrating the lack of consistency in findings in cocaine studies examining striatal responses to non-drug rewards. Taken together, these studies indicate that more research is needed to understand the apparent differences in findings across studies and the extent to which specific individual differences might be contributing to these findings.
We can at this stage only speculate about the possible causes for our results showing relative FCD hypoactivity in the right dorsal caudate for the Gain-Loss contrast and the inverse correlation of Gain-Loss contrast effect size with impulsivity-related Factor 2 in that same region. Our finding of right dorsal caudate hypoactivity in FCD subjects during reward receipt will require replication in future studies, and the interpretation of such results will also necessitate the collection of additional data on treatment strategies (both behavioral and cognitive) used by FCD individuals to maintain abstinence.
We present here, however, three possible explanations for our findings in the right dorsal caudate of FCD individuals that future studies might explore. One such explanation is that prior to cocaine use, FCD compared with CCD subjects, had a different neural 'makeup' within the striatum, including dorsal caudate, or other mesocorticolimbic regions, such that even after years of abuse FCD subjects were able to achieve abstinence more easily than CCD. A second possibility is that individual variation in predisposition to cocaine dependence might involve dorsal caudate activity differences that are normalized by cocaine use. Prolonged abstinence then leads to an ''unmasking'' of some these preexisting differences in dorsal caudate activity. In other words, once cocaine use achieves a certain chronicity, continued drug use might be necessary to 'normalize' the reward system [64]. Therefore, when cocaine use ceases, underlying dorsal caudate deficits, such as those involving reduced dopamine D2-like receptors and elevated dopamine transporters [65], might manifest during receipt of non-drug rewards. A third possible explanation is that FCD subjects, all of whom had abused cocaine long-term (.3 months, average 133 months), might have achieved lasting abstinence through the development of cognitive and behavioral strategies (e.g., group or individual psychotherapy, self-restraint/ willpower) necessary to substantially reduce and resist cocaine craving. Such cognitive strategies might have included acquiring the ability to inhibit craving-and habit-related signals arising from the dorsal caudate during reward receipt. This proposed explanation is consistent with recent studies in which substance dependent individuals were able to use cognitive and behavioral training to reduce self-reported cravings and ventral striatal activity when presented with drug cues [66,67,68]. It is also consistent with recent studies demonstrating that pharmacological (GABA-receptor agonist) blockade of activity in the dorsal striatum significantly reduced cue-induced cocaine-seeking in rats, suggesting that similar reductions in dorsal striatal activity would reduce cocaine-craving in human cocaine dependence [32,69].
Therefore, reduced dorsal caudate activity during reward receipt observed in our FCD subjects, rather than representing a dysfunctional response, might instead represent a successful cognitive strategy that allowed these individuals to remain cocaine abstinent by not activating drug-habit-associated brain regions. Our finding that FCD subjects' scores on self-reported compulsivity and reward-punishment sensitivity correlated negatively with right dorsal caudate activity for the Gain-Loss contrast is also consistent with Volkow's cognitive control hypothesis [66]. The inverse correlation of self-reported compulsivity/reward-punishment sensitivity scores with right dorsal caudate activity might suggest that the most compulsive FCD subjects need to exert the greatest cognitive control to maintain abstinence. Additionally, this inverse correlation indicates that the less compulsive/impulsive FCD subjects tended to mitigate the between-group difference for the Gain-Loss contrast in that brain region. CCD subjects scored similarly to FCD on impulsivity-related Factor 2 (and both scored significantly greater than controls), and yet did not show Gain-Loss contrast hypoactivity in the dorsal caudate, nor Gain-Loss correlation with Factor 2. This result might indicate an unmasking of underlying dysfunction in the FCD group through long-term abstinence. Although the ideas presented here are at this stage largely speculative, and these results will require replication in future studies, when taken together with findings that impulsivity measures correlate inversely with ventral striatal activation in substance-dependent and pathological gambling populations [12,13,36], these concepts resonate with models of ventral-todorsal striatal function underlying impulsive-to-compulsive aspects of addictions [26,29,31,70,71].
Study limitations include reliance on CCD and FCD individuals' self-reports of durations, amounts and frequencies of cocaine use, and, in the case of the former users, the length of abstinence. CCD and FCD subjects, however, provided at least two urine samples to verify current users were cocaine-positive and former users were not. Another possible limitation is that nine CCD subjects had comorbid opiate abuse and/or dependence treated with stable methadone doses at the time of testing. Re-analysis of CCD subjects' fMRI data with opiate users removed, however, revealed no changes from results presented in Figure 3. Effect size plots were essentially identical to those depicted in Figure 4, panel B, and peak difference occurred at the same voxel (x,y,z = 18, 18,9) in a similarly sized cluster in the right dorsal caudate. Other past psychiatric co-morbidity, which were more prevalent in the FCD group, might be confounding our results. However, the fact that removing these participants from the analysis did not have a significant impact on the finding of relative hypoactivity in the right dorsal caudate of FCD group during reward receipt indicates that cocaine use rather than psychiatric co-morbidity is the principal determinant of our results. Also, both current and former cocaine dependent subjects reported varying amounts of weekly cocaine use and lifetime durations of use and differences in age at first use that could potentially impact brain function. Future, larger studies could examine these factors directly.
We should also note the significant difference in weekly spending on cocaine between the FCD and CCD groups. While self-reported weekly spending in the FCD group is considerably higher than the CCD group, we believe that this difference might, at least in part, reflect under-reporting by the CCD group. A study by Harrison and Hughes [72] found that, with respect to recent cocaine users (i.e., the CCD group), ''over the course of the 17week clinical trial, subjects reported cocaine use on 20 percent of occasions, but tested positive for cocaine (qualitatively) on 68 percent of occasions.'' Similar studies on self-reported usage in current cocaine abusers have also shown that such users tend to under-report [73,74,75]. Therefore, we speculate that many of the CCD individuals in our study may have been under-reporting their current usage. This under-reporting might be due to the desire of CCD individuals to conceal the magnitude of their ongoing illicit drug use from both law enforcement and from healthcare professionals. The FCD group in our study, in contrast, might not have such reservations in self-reporting past cocaine usage. However, we acknowledge that the significant difference in self-reported weekly spending on cocaine between the former and current cocaine groups is a potential confound.
Finally, we should note that we were unable to determine precisely the last use of cocaine by CCD subjects before their fMRI scan. A positive urine test for cocaine generally indicates that the individual had used cocaine within the previous 72 hours [76]. Due to cocaine's short half-life of 40-60 minutes [77], this implies that some urine-positive CCD subjects in our study might have been in a state of early withdrawal, with symptoms including depressed mood, fatigue, and psychomotor retardation or agitation [78] that might have impacted their reward system response. On the other hand, some urine-positive CCD subjects might have used cocaine comparatively recently prior to scanning (,3-6 hours), so as to have hypothetically 'normalized' their reward system at the time of scan. Hence, in CCD participants, variation in amount and recency of cocaine use and state of intoxication/withdrawal at the time of fMRI scan might have added variance or 'noise' that obscured underlying differences in reward system function in these subjects versus FCD and HC.
Conclusions
In summary, during receipts of rewards versus punishments in an interpersonal competitive game involving risk-taking, FCD but not CCD subjects showed altered striatal activation compared with HC subjects. Furthermore, these activation differences were greatest in right dorsal caudate, a region associated with cocaine craving and habit-based behavior such as occurs in drug addiction and, for FCD individuals only, were negatively correlated with an impulsivity-related factor associated with compulsivity and sensitivity to reward and punishment. To our knowledge, our study is one of the first to examine brain reward system function in FCD subjects, i.e., individuals with long term cocaine abstinence (.6 months), thus filling an important gap in the study of cocaine addiction. Future studies should be directed towards determining explanations for the persisting significance of the impulsivityrelated factor of self-reported compulsivity and reward-punishment sensitivity in cocaine-dependent individuals, even after prolonged cocaine abstinence. Finally, future research should examine the extent to which dorsal striatal function might represent a target for treatment development in cocaine dependence.
==
Domain: Psychology Biology Medicine
|
[{"added": "2016-05-04T20:20:58.661Z", "created": "2012-05-14T00:00:00.000", "id": "3827248", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0034917&type=printable", "pdf_hash": "91ac58dcc55d120a891f1c40a15eb34ab1106674", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:36241", "s2fieldsofstudy": ["Psychology", "Biology"], "sha1": "91ac58dcc55d120a891f1c40a15eb34ab1106674", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Genome-wide significant regions in 43 Utah high-risk families implicate multiple genes involved in risk for completed suicide
Suicide is the 10th leading cause of death in the United States. Although environment has undeniable impact, evidence suggests that genetic factors play a significant role in completed suicide. We linked a resource of ~ 4500 DNA samples from completed suicides obtained from the Utah Medical Examiner to genealogical records and medical records data available on over eight million individuals. This linking has resulted in the identification of high-risk extended families (7–9 generations) with significant familial risk of completed suicide. Familial aggregation across distant relatives minimizes effects of shared environment, provides more genetically homogeneous risk groups, and magnifies genetic risks through familial repetition. We analyzed Illumina PsychArray genotypes from suicide cases in 43 high-risk families, identifying 30 distinct shared genomic segments with genome-wide evidence (p = 2.02E-07–1.30E-18) of segregation with completed suicide. The 207 genes implicated by the shared regions provide a focused set of genes for further study; 18 have been previously associated with suicide risk. Although PsychArray variants do not represent exhaustive variation within the 207 genes, we investigated these for specific segregation within the high-risk families, and for association of variants with predicted functional impact in ~ 1300 additional Utah suicides unrelated to the discovery families. None of the limited PsychArray variants explained the high-risk family segregation; sequencing of these regions will be needed to discover segregating risk variants, which may be rarer or regulatory. However, additional association tests yielded four significant PsychArray variants (SP110, rs181058279; AGBL2, rs76215382; SUCLA2, rs121908538; APH1B, rs745918508), raising the likelihood that these genes confer risk of completed suicide.
Introduction
Suicide is the 10th leading cause of death in the United States; over 44,000 individuals die by suicide in the United States every year [1]. Although environmental variables have undeniable impact, evidence suggests that genetic factors play a role in completed suicide, with heritability of close to 50% [2,3]. Recent growth in the number of suicide genetic studies has resulted in promising findings from candidate gene and genome-wide association studies [4], though many remain to be replicated. Replication is hampered by sample differences across studies, including differences in demographics and primary diagnoses of study samples, as many studies of suicide risk have been conducted within cases ascertained for specific psychiatric disorders [4]. In addition, most studies of suicide have focused on suicidal ideation and behaviors; these phenotypes are much more common than completed suicide, allowing for ascertainment of sufficiently powered samples, but suicidal behaviors can be difficult to quantify, and represent individuals with a range of risk for later suicide. In addition, evidence suggests important differences in the etiology of suicidal behaviors versus the less ambiguous but much rarer outcome of completed suicide [5].
We implemented a unique study design to investigate genetic risk for suicide through the collection of DNA samples on > 4500 consecutive individuals who died by suicide in the state of Utah, providing an unparalleled population-based genetic resource. This sample results from a long-term collaboration with the Utah State Office of the Medical Examiner. The records from these cases have been linked to the Utah Population Database (UPDB, [URL] lthcare.utah.edu/huntsmancancerinstitute/research/updb), a comprehensive database including multi-generational genealogies, as well as death certificates, demographic data, and current medical information on over eight million individuals. Through this linking, we have identified very large families (7-9 generations) with significantly elevated suicide risk. Familial aggregation across distant relatives in these families minimizes the impact of shared environment on risk. High-risk families also provide more genetically homogeneous risk groups, increasing statistical power to detect familial variants associated with disease risk. The Utah extended family study design has already shown success in the study of other complex genetic diseases of extended families of similar size (e.g., colon cancer [6], breast cancer [7], and cardiac arrhythmia [8]).
This study reflects an analysis of 43 very large Utah families at significantly elevated risk for completed suicide.
The focus on completed suicide, statistically concentrated in these high-risk families, optimizes power to reveal regions of the genome likely to contain risk variants. Our design investigates genetic risk using the suicide cases in the extended families regardless of co-occurring psychopathology, and continues with follow-up studies from our population-wide ascertainment of all suicide deaths in Utah, again without regard to co-occurring psychopathology. We recognize that psychiatric diagnoses are critically important in suicide risk [9,10]; it is likely that findings from our study are related to these associated risks. However, because of the familial aspect of our design, it is possible that our results may reveal risk variants that crosscuts specific psychiatric diagnoses [11][12][13][14].
Genetic studies of psychiatric disease have revealed associations with multiple rare and common risk variants with reduced penetrance, which may interact in complex ways with each other, with background genetics, and with environmental risks [15]. Based on results to date, we expect that suicide will follow this complex genetic architecture. In this study, the familial analyses use a new statistical method (Shared Genomic Segments, SGS [16]) that is well-powered to identify rare genetic variants in large families, evidence that can then be used to prioritize searches for additional variants contributing to risk in other case samples. This design is complementary to the Genome-Wide Association Study (GWAS) approach in large casecontrol samples, which can also produce statistical evidence for risk genes to be followed up in independent case samples.
Using genome-wide single-nucleotide polymorphism (SNP) variants matched to the same variants in publicly available population control data, we identified regions of the genome that segregate in suicide cases within high-risk families. These statistically significant regions provide compelling genes as targets for follow-up. Although such follow-up studies would ideally use comprehensive sequence data, the SNP array platform used in this study contains putatively functional variants of high interest to psychiatric and medical disorders, both of which may share overlapping suicide risk. Functional content of the Psy-chArray was investigated first within familial cases responsible for the significant regions, and then within1 300 additional Utah suicide cases unrelated to the original extended families analyzed.
This study adds to the growing knowledge of genetic risk for suicide. First, we have identified genes in regions of significant familial segregation in large high-risk families, providing replication for previously reported genes of high interest, and identifying target genes for additional followup. Second, we have identified novel risk variants using a large follow-up association analysis of PsychArray variants with predicted functional impact using a populationmatched resource of suicide cases.
Sample
This project is possible because of a collaboration with the Utah State Office of the Medical Examiner (OME), which has spanned two decades. With Institutional Review Board (IRB) permissions from the University of Utah, the Utah Department of Health and Utah Intermountain Healthcare, we have collected de-identified DNA samples from consecutive suicides since 1997. The collection numbers 4585 (3632 males and 953 females). DNA was extracted from blood using the Qiagen Autopure LS automated DNA extractor (www.qiagen.com). Identifying information from cases with DNA was linked to data within the UPDB's secure computer servers. All identifying data were then stripped before providing data to the research team; suicide cases and family structure data are referenced by anonymous IDs. DNA for this research project is shared with the NIMH Repository and Genomics Resource, project number 315 (2880 samples are now at the repository; additional samples are being sent on an ongoing basis).
Determination of familial risk, selection of families/ cases
Genealogical data in the UPDB was used to construct family trees and identify those families at high risk for suicide. Beyond the suicide cases with DNA, the UPDB contains records of all known suicides from Utah death certificates dating from 1904 (N = 14,288). All 14,288 cases were used to estimate familial risk of suicide. To determine the extended families at highest risk, we used the Familial Standardized Incidence Ratio (FSIR) statistic [17], calculated by comparing the incidence of suicide in each extended family to its expected incidence determined by the statewide distribution for suicide stratified by sex and age. We identified 241 high-risk families containing significant excess of suicides (p < 0.05) and at least three suicides with DNA. We selected 43 of these families for analysis (Table 1) based on significance of the FSIR risk statistic, number of cases with DNA, and overall count of meioses between these cases (see Table 1). The 43 families included 2.04-4.41 times the expected number of suicide cases as reflected in the FSIR statistic (range p = 0.003-1E-12, average p = 0.0007). The average number of cases per family with genotyping was 6.2 (range 3-13), and the average number of meioses between analyzed cases was 29.6 (range 15-70; see Fig. 1 for an example of how meioses are counted in a family of moderate size from our resource). Family-specific significance thresholds for genomic sharing (see analysis section below) depend upon family size, structure, dispersion of cases, and number of cases analyzed. Permission for use of family structure data were granted by the Resource for Genetic and Epidemiologic Research (RGE, [URL]), the oversight committee for use of UPDB data.
The 198 families not selected for analysis in this study exhibited less-significant risk, and/or had too few cases with DNA, and/or had insufficient distance between cases. The average number of suicide cases per family with DNA in these 198 families was 3.23 (SD = 0.88, range 3-5). The average p value associated with these FSIRs was 0.0028 (SD = 0.0074, range 0.0458-1E-4).
Diagnostic data
In addition to basic demographic and cause of death information, we had access to diagnostic data for psychiatric conditions associated with suicide using electronic medical records data through the UPDB. Codes were linked to case numeric case identifiers within the Utah Population Database; de-identified results were provided for analysis. Conditions were defined by groups of diagnostic codes aggregated according to the International Classification of Diseases (ICD) system (www.icd9data.com; see Supplemental Table S1 for the list of codes used to define diagnostic categories in this study). Importantly, because our cases are not derived from a clinical population, cases can exhibit with no co-occurring diagnoses. Missing diagnostic data can occur for many reasons, including: (1) existence of diagnostic codes other than the 359 codes in Table S1; (2) a case who did not seek medical attention for the psychiatric disorders in question owing to stigma, lack of insurance, cultural barriers, or other lack of access to services, agerelated lack of recognition of pathology, or symptoms not perceived to require medical attention; (3) diagnostic data not contained in the UPDB, including diagnoses prior to the storage of electronic diagnoses, or diagnoses given out of state or outside the~85% coverage of electronic medical records data available in the UPDB. We treated missing data as unknown rather than assuming the absence of pathology.
Molecular data
The SGS analyses used variants from the Illumina Infinium PsychArray platform, version 1.0 ( [URL]. com/products/by-type/microarray-kits/infinium-psycharray. html) genotyped on 216 suicide cases in the 43 selected families. This PsychArray includes 265,000 common informative tag SNPs, 245,000 variants selected from exome sequencing studies of medical and psychiatric conditions, and 50,000 rare variants associated specifically with psychiatric conditions. Supplemental Figure S1 shows the use of genotype data for the study. Genotyped array content was oriented to 1000 Genomes Project data. For initial analyses of familial sharing, we included all variants contained in 1000 Genomes Project control data, omitting variants where orientation was ambiguous, and variants which were not polymorphic. Using PLINK [16], we also removed 17,058 variants with > 5% missing calls and 176 variants that failed Hardy-Weinberg equilibrium (p < 0.001). In addition, one case from family 553615 was removed owing to a low call rate (> 5% missing). Our initial familial analyses used 237,415 variants from 215 completed suicide cases to reveal familial variation that defines the boundaries of the segments shared among related cases. Rare putatively functional array variants meeting QC criteria, including psychiatric and medical disease-specific variants, were used to follow-up additional variants in the shared regions.
Analysis
(See Supplemental Figure S1 for a flow diagram of the study). We began by using a new analytical method, Shared Genomic Segments (SGS) [18] developed for analyzing large high-risk families to identify subsets of cases that share regions beyond sharing expected by chance. SGS identifies excessive lengths of consecutive SNPs with allelic sharing between relatives to infer genomic segments that are inherited. Theoretically, chance inherited genomic sharing in distant relatives is extremely improbable; thus, the method has power in large families such as those in our study [19]. The significance of each shared segment is assessed empirically using gene-drop simulations (independent of case status) to create a null distribution of expected sharing within each family. The method assigns haplotypes to family founders according to a publicly available linkage disequilibrium map from 1000 Genomes European data, followed by simulated segregation through each specific family structure, repeated a minimum of 500,000 times. The observed sharing is compared with simulated sharing to determine significance. See Supplemental Figure S2 for a hypothetical simplified example of SGS sharing. Genome-wide significance thresholds are calculated specific to each family, as statistical power varies with family structure, number of cases, and distance between cases. Significance thresholds account for multiple testing and linkage disequilibrium, and also adjust for within-family heterogeneity by including adjustment for all possible subsets of within-family sharing among cases. Model fitting to determine theoretical genome-wide thresholds used these distributions of gene-drop results. The overwhelming majority of the genome will be null (does not contain a suicide risk variant); we acknowledge a slight conservative bias as these distributions also contain a small number of true positives [16]. The genome-wide significant threshold corresponds to a false-positive rate of 0.5 per genome per family, whereas the suggestive threshold corresponds to one false-positive result per genome per family. In this study, we report regions with family-specific genome-wide significant evidence, and regions overlapping in more than one family where family-specific evidence was at least genome-wide suggestive. P values for these overlapping multiple-family regions were approximated using Fisher's combined probability test [20]. The SGS analysis software is freely available ( [URL]. edu/huntsman/labs/camp/analysis-tool/shared-genomicsegment.php). Power of SGS was previously investigated for a range of genetic models involving rare variants in extended family data [19], showing appropriate power for large families with at least 15 meioses between cases. For all scenarios considered in this study, genome-wide association studies would have had negligible power. Given these results, we selected only extended families with at least 15 meioses between cases (see Table 1 and Fig. 1).
Follow-up analyses
All follow-up work focused on the targeted set of genes identified by the significant SGS regions. Genes were considered within significant segments for follow-up if coding or regulatory sequence (defined using Genomic Regions Enrichment of Annotations Tool [21]) fell within the shared segment. Genes determined in previous research to be highly likely to represent false-positive results [22] were deleted (in our data, a cluster of 56 olfactory receptor genes in one region on chromosome 11, FAT1, CTC-432M15.3, and TRIM51). Three phases of follow-up work were pursued (see Figure S1).
Corroborating evidence from the literature
As a first investigation of the genes indicated by significant SGS regions, we conducted a comprehensive search of the literature for all suicide-related risk genes against which to compare with SGS location-specific evidence. Suicide risk was identified by searching the Web of Science database for the terms: suicid* and gene* (captures variants including suicide, suicidal, suicidality, gene, genetic, etc.). We included reviews on the genetics of suicide as well as linkage, GWAS, candidate gene, expression and epigenetic studies. As secondary information about the suicide-related genes, we also queried DisGeNET [23,24] for gene associations or involvement with neuropsychiatric disorders and inflammation owing to their known association with suicide risk [9,10,25].
PsychArray variants in cases subsets in high-risk families with SGS sharing
The next phase of follow-up comprised a search of the specific familial cases that generated each SGS region using available array variants within each region to determine whether any particular array variant could be responsible each result. Although the variants available to us on the PsychArray are far from complete, a search of relatively rare coding-region variants provides an efficient, immediate, potentially interpretable screen of our results in lieu of large-scale sequencing data [26]. We therefore checked for sharing of the minor allele of non-synonymous variants within specific cases responsible for SGS results, and strictly within region boundaries. Because the SGS method is most powerful for the detection of rare familial variants, we selected a minor allele frequency < 10% in the publicly available Exome Aggregation Consortium (ExAC, www.exac.broadinstitute.org) European, non-Finnish data (matching in ancestry to our sample).
Gene-based evidence in additional Utah suicide cases
The final follow-up phase focused on genes from SGS regions as targets for further study in additional sample resources. Although familial variants may be private to the extended family/families producing the SGS evidence, it is also possible that the evidence implicates genes with additional risk variants in independent case samples (allelic heterogeneity). We screened an independent cohort of Utah suicides for potential functional variants in SGS-targeted genes; this case sample most closely matches the discovery families, as it was derived from the Utah population, and is comprised of completed suicides. Owing to the same population ascertainment source, it is predicted to match the familial discovery sample regarding demographics and diagnostics. To maximize statistical power in our relatively small follow-up cohort of 1300 completed suicide cases, we focused on the potential to discover moderately penetrant, potentially interpretable functional causal variation. To this end, we used the following criteria to select variants: (1) in coding sequence of genes identified by the significant SGS regions, (2) non-synonymous and predicted to be damaging from either PolyPhen [27] or Sift [28], (3) minor allele frequency < 20% in ExAC European, non-Finnish data. We tested for significant allelic association compared with ExAC European, non-Finnish data using Fisher's exact test, or with chi-square tests for variants with > 10 observed chromosomes with the minor allele in cases and controls. Tests for additional variants within SGS regions excluded suicide cases responsible for original sharing evidence in that region. Significance was adjusted for multiple tests. Suicide cases are not as evident in upper generations because suicide status from death certificates is only available back to 1904. Note that gender is disguised and sibship order is randomized in order to protect the privacy of family members. Family size: there are 34 total meioses between the seven cases in this family; this counting is shown in purple on the drawing. SGS requires a total of at least 15 meioses between cases for adequate statistical power. Shared segments: three genomic segments provided significant evidence of sharing between cases in this family. The pattern of segregation of each segment is shown. Cases 2, 4, 5, 6, and 7 share region 1 (red). Cases 1, 2, 3, 4, 5, and 6 share region 2 (gold). Cases 1, 2, 3, 4, and 7 share region 3 (blue). Essential segregation is shown; however, when cases do not share, the region can actually be lost at any meiosis above the case in the family tree. The exact point of this loss is unknown to identify genomic shared regions are done within family, we included cases each time they occurred under each founder, as we do not know a priori where true sharing may occur. It is possible that cases share risk variant(s) from one set of founders with other cases in that family, but then also share other risk variant(s) with cases in a second family through connections with the other founding couple. The complexities in relationships may allow for future studies of gene × gene interactions once risk variants have been established.
High-risk families
Descriptive characteristics of the 215 independent discovery cases from the 43 families were compared with the other 4370 unselected Utah suicide cases with DNA. Within the 215 high-risk familial cases, 172 were male (80.0%), similar to the 79.2% rate in the unselected sample. Average age at death in the family sample was 34.28 years (standard This region was significantly shared by 553615 on its own, but a smaller overlapping region was also shared by 553615, 603481, and 176860 c Two cases were omitted from family 553615 to satisfy the independence requirement for computing Fisher's combined p value [20]. Significance thresholds were re-computed for family 553615 eliminating these cases. Person 112304 is a descendant of both 553615 and 603471, and person 95765 is a descendant of both 553615 and 176860 We found similar percentages of cases with presence of diagnoses from electronic records in the 215 familial cases as compared with the unselected 4370 cases, with the exception of an increase in cases with personality disorders and an increase in prior attempts/suicidal ideation in the familial sample. Percentages of cases with at least one code in each of the diagnostic categories vs. the unselected sample were as follows: depression, 41.4% vs. 36.2%; bipolar, 13.0% vs. 10.6%; anxiety, 23.3% vs. 22.8%; psychosis, 2.3% vs. 2.9%; substance use/abuse, 9.8% vs. 12.4%; personality disorders, 14.4% vs. 9.7% (chi-square = 5.06, p = 0.02); ADHD, 4.7% vs. 3.7%; previous attempts/ideation, 37.2% vs. 28.9% (chisquare = 6.82, p = 0.009).
SGS results
SGS analyses revealed 16 single-family regions with genomewide significance ( Table 2). Several families generated more than one region; these were the larger families, where there was more opportunity for multiple different case subsets to show sharing evidence (see Fig. 1 for a specific example of sharing in family 66,494; see Supplemental Figure S4a for drawings of all families with genome-wide significance). Table 2 also presents 15 regions where sharing evidence overlapped across more than one family ( Figure S4b); in each of these regions, the single-family evidence was at least at the genome-wide suggestive level. For the region on chromosome 5q23.3-q31.1, person 112,304 is a descendant of both 553,615 and 603,471, and person 95,765 is a descendant in both 553,615 and 176,860. To satisfy the independence requirement for computing the Fisher's combined p value [20], we computed the p value for this region omitting these cases. There are 207 genes with coding or regulatory sequence in the 31 SGS regions (Table S2). Follow-up studies (see Figure S1 for overview) 1. Supporting literature evidence: we did not find any overlap between significant SGS regions and genomic regions identified by previous family-based linkage studies of suicidal behaviors (Table S3). At the gene level, we reviewed the 207 SGS-targeted genes, first investigating specific supporting evidence of suicide risk. From our comprehensive literature search, a total of 755 genes have been associated with suicide with varying levels of statistical support (Table S4). Eighteen SGStargeted genes were among these 755 suicide-risk genes (see Table 3; also highlighted in Table S2; a detailed description of these 18 genes follows Table S2). Given an estimated number of~19,000 genes in the human genome [29], we estimate that 755/19000 = 4% of genes in the genome have current evidence associated with suicide risk. If the SGS regions were a random sample of the genome and unassociated with the suicide phenotype, we would expect that only~4% of the genes in SGS regions ( When we analyzed the selected 352 potentially damaging, relatively rare array variants within genes targeted by significant SGS regions, we found four variants with significantly increased presence of the minor allele compared with ExAC European non-Finnish frequencies, adjusting for multiple testing correction (Table 4: rs181058279, p = 5.45E-06; rs76215382, p = 8.48E-05; rs121908538, p = 3.14E-12; rs745918508, p = 5.40E-29). Specific characteristics of cases with each of these rare variants are described in Table S5. This evidence suggests rates of psychopathology similar to rates seen in the overall follow-up genotyped sample. Demographics were also similar, though cases with the AGBL2 variant were significantly more likely to be female (chi-square = 7.82, p = 0.003).
Discussion
We have ascertained and studied a unique resource of 43 extended families at high risk for suicide. The design uses the distantly related, high-risk cases to magnify genetic effects, enrich for genetic homogeneity, and minimize shared environmental effects. Families were identified from cases sampled from population-wide ascertainment, resulting in a study design independent of specific psychiatric diagnosis. Cases in the high-risk families were significantly younger at death, by 5.73 years on average, perhaps reflecting enhanced familial genetic risk over and above accumulated environmental risks that may play a greater role in suicide at later ages. The follow-up sample of~1300 genotyped cases more closely matched this familial discovery sample. The mean age at death was similarly young, 5.05 years younger than the unelected sample. The matching of the replication cohort may be due to the fact that we have thus far targeted our overall genotyping efforts to cases with increased evidence of at least one other extended relative who is at suicide risk. Diagnostic data, when present, suggested similar rates of psychopathology across our entire research resource, with somewhat elevated rates of personality disorders and of suicidal ideation and previous attempt in both the family discovery sample and the follow-up genotyped sample.
Cases in families were analyzed with a statistically powerful method, SGS [18], resulting in the identification of genome-wide significant regions likely to harbor risk variants. This family evidence implicated 207 genes for targeted follow-up. We found significant overlap with a comprehensive survey of 18 genes implicated in suicide, lending further support for these genes. Of note, 15 of these 18 genes also show previous associations with inflammatory conditions (Table 3), supporting accumulating evidence for a cross-association between inflammation and suicide risk [30]. Because our method discovers familial genomic regions, we also reviewed prior family linkage studies of suicide risk, but did not find overlaps. This result is perhaps not surprising owing to differences in ascertainment and outcome measures in these previous studies.
The additional rare disease-associated content of the array did not immediately reveal functional rare variants shared across cases responsible for the familial sharing. This result is likely due to the limited number of potentially riskcausing variants captured on the array; sequencing will be required to discover the causal variants shared across the high-risk discovery cases. Alternatively, one or more regions may be false positives.
SGS also provides target genes for other follow-up studies. Genes truly associated with suicide risk may harbor multiple risk-associated variants (allelic heterogeneity). By focusing our follow-up studies to find additional risk alleles to the much reduced number of high-interest target variants in genes identified by SGS, statistical power is increased. An independent population-based cohort of~1300 Utah completed suicide cases, well matched for ascertainment, resulted in four variants associated with suicide (SP110, AGBL2, SUCLA2, and APH1B). SP110 is part of a leukocyte-specific nuclear body protein complex, and likely plays a role in gene transcription [25]. It has been implicated in pathogen resistance and immunodeficiency [31,32], and may relate to suicide risk through a growing body of evidence implicating immune risk and inflammation [30]. AGBL2 is an ATP/GTP binding protein implicated in brain structure and function [33]. SUCLA2 is a mitochondrial tricarboxylic acid cycle protein recently implicated in energy supply to the synapse [34], and is possibly associated with recent findings linking suicide risk and hypoxia [35]. APH1B is a transmembrane protein associated with risk of Alzheimer's [36] and Parkinson's [37] diseases.
Characteristics of cases with each of these four variants did not reveal any striking patterns of association with specific psychopathology, though cases with the AGBL2 variant were significantly more likely to be female. Followup in additional research cohorts will be required to clarify diagnostic associations, and to replicate the association with gender found with the AGBL2 variant.
Limitations
Suicide cases were predominantly of Northern European ancestry, as verified with genotype data, so results may be limited to this race/ethnic group. The genome-wide background of the PsychArray contains~265,000 common variants, which is relatively sparse for a genome-wide array.
A denser array could have provided additional precision to region boundaries, or may have revealed that some regions were false positives. Diagnostic data were limited to available diagnoses in the electronic medical record. Cases with no diagnostic data are not assumed to have an absence of psychopathology. Rather, missing data more likely reflect either diagnoses outside the scope of our data resources, or lack of connection to services owing to insurance, stigma, cultural factors, or a perception that symptoms did not warrant treatment.
Conclusions
Our study has found significant associations using only on the relatively rare, potentially functional variants captured on the PsychArray; these results have been discovered through a rigorous statistical prioritization and variant selection based only on functional annotation and frequency. As new data on our resource become available, it is likely that additional potential risk variation will be found. However, the current work has produced several important lines of evidence. First, the genome-wide significant SGS regions identify 207 target genes for suicide risk. Second, follow-up analyses of these regions in an independent population-based cohort of suicides highlighted four genes with potential functional risk variants, pending replication. Finally, the SGS regions contained 18 genes with corroborating evidence for suicide risk, suggesting these as strong candidates for future work.
Compliance with ethical standards
Conflict of interest HC had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. HC, LI, and ED are partially supported by a research contract from Janssen Research, LLC. QSL is an investigator at Janssen Research, LLC. The remaining authors have no conflicts of interest relevant to the content of this manuscript, including no financial interest, relationships, or affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit [URL]. org/licenses/by/4.0/.
==
Domain: Psychology Biology Medicine
|
[{"added": "2018-11-10T06:20:13.851Z", "created": "2017-09-28T00:00:00.000", "id": "53030497", "metadata": {"extfieldsofstudy": ["Biology", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://www.nature.com/articles/s41380-018-0282-3.pdf", "pdf_hash": "5ef419053d19cf03d04315cbb75f8c1ee915a6ba", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:35062", "s2fieldsofstudy": ["Psychology", "Biology"], "sha1": "d6c7b39098d1f204ddaa72794b9d5ffbb19d75a7", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Entrainment of heterogeneous glycolytic oscillations in single cells
Cell signaling, gene expression, and metabolism are affected by cell-cell heterogeneity and random changes in the environment. The effects of such fluctuations on cell signaling and gene expression have recently been studied intensively using single-cell experiments. In metabolism heterogeneity may be particularly important because it may affect synchronisation of metabolic oscillations, an important example of cell-cell communication. This synchronisation is notoriously difficult to describe theoretically as the example of glycolytic oscillations shows: neither is the mechanism of glycolytic synchronisation understood nor the role of cell-cell heterogeneity. To pin down the mechanism and to assess its robustness and universality we have experimentally investigated the entrainment of glycolytic oscillations in individual yeast cells by periodic external perturbations. We find that oscillatory cells synchronise through phase shifts and that the mechanism is insensitive to cell heterogeneity (robustness) and similar for different types of external perturbations (universality).
C ellular signaling, gene expression, and metabolism are determined by chemical reactions within the cell. The discrete nature of molecular reactions as well as environmental fluctuations and heterogeneity cause fluctuations in these processes. The effect of such noise on cell signaling and gene expression has recently been studied intensively using single-cell analysis [1][2][3][4][5] . However, despite its importance the role of noise and heterogeneity in metabolism 6 is not yet well understood. Heterogeneity is very important in systems where cell-cell communication may cause the cells to synchronise their metabolic oscillations. Cell-cell communication is important because it is a prerequisite for organisation of cell communities and is necessary for evolution to proceed from unicellular to multicellular behaviour.
One of the most intensively studied metabolic dynamics is that of glycolytic oscillations in yeast cells Saccharomyces cerevisiae 7-10 , for a review see 11 . Remarkably, yeast cells in dense populations appear to be able to synchronise their glycolytic oscillations 9,12-15 and macroscopic oscillations have been observed in yeast populations [7][8][9][10] . This effect appears quite universal in that different signaling molecules/receptors can cause synchronisation 9,[12][13][14][15][16][17][18][19] . It is thus expected that similar synchronisation mechanisms are at work in other oscillatory cell types, possibly via different metabolic species. However, the empirical data pertain to dense cell cultures of millions of cells and previous attempts to investigate the oscillatory behaviour in individual cells have proven challenging 8,20,21 . Since observations of macroscopic oscillations do not distinguish between oscillation and synchronisation such measurements neither allow to deduce the microscopic mechanism of synchronisation nor to investigate the role of cell-cell variations.
There is a long history of theoretical investigations of oscillations in glycolytic networks 22,23 and it is now well understood that sustained oscillations may occur under a wide range of conditions [24][25][26][27][28][29] . Synchronisation, by contrast, appears to be much more difficult to achieve in computer simulations of model systems [24][25][26][27] . Whether or not synchronisation occurs appears to very sensitively depend upon the model parameters 24 . This is at variance with the apparent robustness of the phenomenon observed in dense cell cultures.
In order to understand the synchronisation mechanism and to elucidate the role of heterogeneity it is necessary to quantify how individual cells respond to periodic perturbations of their environments, either due to an externally applied perturbation or through interactions with other cells. To achieve a qualitative and quantitative understanding of the phenomenon requires us to answer three fundamental questions.
First, what is the mechanism of synchronisation? Experimental studies of macroscopic oscillations indicate that phase synchronisation may play a role 15 : Figure 3(A) in Ref. 15 shows that the macroscopic phase shift due to a sudden change in environmental conditions depends upon the value of the macroscopic phase just before the perturbation, see also 9 . In order to quantify the effect and to unequivocally establish whether the synchronisation can be achieved by phase changes alone it is necessary to follow how an individual cell is entrained by a periodic perturbation. One must determine whether and at which frequencies cells may continue to oscillate after the perturbation has been switched off, and, most importantly, whether the amplitude changes during entrainment. Last but not least an important question is how the frequency and amplitude of oscillation in the absence of perturbations affect the propensity of the cell to be entrained when the periodic perturbation is switched on. Existing experiments 9,14,15 do not answer these questions because (i) only the macroscopic response is measured, and (ii) the response is only measured at a single perturbation pulse. Some model calculations show that mainly non-oscillatory cells become entrained by perturbations 30 . A very important open question is how the phase of an entrained cell relates to the phase of the perturbation. Do cells typically oscillate in phase with the perturbation or not? Theoretical models can show in-phase or out-of-phase entrainment, sensitively depending on model parameters 26 . Macroscopic experiments do not allow resolving this question, because subpopulations oscillating out-of-phase will only lead to a lowering of the amplitude of the macroscopic signal. In order to determine the mechanism of synchronisation, these experiments must be performed on individual cells.
Second, how robust is the synchronisation mechanism? In a theoretical model for phase synchronisation, the efficiency of the mechanism is determined by the heterogeneity of the cells as well as the strength of the entrainment 31 . This is very important because no two cells are alike, and different cells respond differently to external perturbations. In order to quantify how this heterogeneity may prevent synchronisation, we must study the response of an ensemble of independent individual cells with different properties. Can a periodic perturbation entrain cells that initially oscillate at substantially different frequencies? Theoretical analysis using a recent model has shown that perturbations with the synchronising metabolite acetaldehyde (ACA) are not sufficient to entrain cells when they have different frequencies 32 . Most importantly, in an ensemble of many cells, how a large fraction of the individual cells becomes entrained? To answer this question single-cell studies are required, because it is impossible to distinguish between full and partial synchronisation in experiments that only measure macroscopic properties. In order to determine how quickly the mechanism entrains it is necessary to investigate responses of individual cells over many subsequent periods of the periodic perturbation. The degree of synchronisation is quantified by computing an order parameter 31 .
Third, how universal is the synchronisation mechanism? Entrainment involves the entire glycolytic network, and not just a single reaction or chemical species. The effect might in fact be the result of the combined response to several different chemical species 32 . The kinetics leading to synchronisation is thus very complicated, but entrainment appears to occur for a wide range of different conditions and types of perturbations 9,[12][13][14][15][16][17][18][19] . The question is whether the mechanism leading to synchronisation is the same for all species.
In order to answer the questions raised above we report on the results of novel experiments subjecting individual yeast cells to periodic perturbations. Using an experimental setup that combines an optical trap for cell positioning with microfluidics for precise spatial and temporal control of the extracellular environment we induce glycolytic oscillations in individual cells and investigate their response to periodic perturbations. These experiments allow us to address the questions raised above.
To enforce strict periodicity of the perturbations, a microfluidic device is used to achieve complete and reversible changes of the extracellular milieu. This makes it possible to experimentally investigate how the phase, the frequency and the amplitude of an individual-cell oscillation are affected by periodic external perturbations. Analysing these parameters from individual cells allows us to determine the mechanism of entrainment, and thus to answer the first question raised above.
To answer the second question, we calculate the order parameter of the phases of the oscillations of the individual cells. This reveals the degree of synchronisation. Studying individual cells allows us to determine which fraction of cells becomes entrained by the perturbation. To ensure that no additional perturbations from cell-cell interactions complicate the analysis, we position the cells far apart from each other by means of an optical trap. Cell-cell interactions are further reduced by flushing out intercellular signaling agents by the flow in the microfluidic chamber 28 .
To answer the third question we investigate through which biochemical reactions in the glycolytic reaction network the perturbation causes entrainment. Starting out, we investigated the individual cell responses to periodic addition of ACA to gain information about the single-cell responses at conditions previously studied in populations 13,15 (see Supplementary Fig. S1 online). Strikingly, we found that removal of cyanide without any addition of ACA caused a similar response, in contrast to the results shown in previous population studies 9 . In this study we therefore chose to investigate the effect of periodic cyanide perturbations. Induction of oscillations has mostly been achieved by addition of glucose and subsequent addition of 2-8 mM cyanide 26,33,34 . The most common explanation for macroscopic oscillations to be detected only for this cyanide concentration range is twofold. Firstly, in this range respiration is inhibited by cyanide binding to cytochrome c oxidase. Secondly, in this range the ACA concentration is lowered to levels where the cells become sensitive to ACA secretion by other cells 9,13,15,[33][34][35] . The reason is that cyanide binds ACA and forms lactonitrile 36 .
Results
Entrainment by cyanide removal. Alternating between a solution containing 20 mM glucose 1 5 mM cyanide and a solution containing only glucose, we show that periodic removal of cyanide entrains the oscillations of individual cells in a similar way as addition of ACA ( Fig. 1a and Supplementary Fig. S1 online). In control experiments where only the flow rates in the microfluidic flow chamber are changed but the concentration of chemicals in the solutions are kept constant, no entrainment can be seen (Fig. 2a). The results of the control experiment are in agreement with a previous study, which showed that the oscillatory behaviour is affected only by changes of chemicals and not by changes of flow rates in the microfluidic flow chamber 37 .
Mechanism of synchronisation. To investigate the mechanism responsible for the observed entrainment, the instantaneous phases were calculated (Figs. 1b and 2b) by means of a standard procedure based upon evaluating the Hilbert transform of the signal 38 . To reveal temporal information about the synchronisation mechanism, the phases before and 9 s after cyanide removal were plotted for perturbations 1, 2, 4 and 8 ( Fig. 3a-d). Data from experiments where entrainment is induced by cyanide removal is shown in red, while data from control experiments is shown in black. The corrected phase shifts at the first perturbation were fitted with second-degree polynomials ( Fig. 4 and Table 1). The cells in the control experiments show the expected, constant phase shift, independent of the phase of the oscillations at the perturbation. The cells exposed to the periodic cyanide removal, on the other hand, show a significant phase dependence of the phase shift. The results show that cells that have phases close to -p or p become negatively phase shifted, while cells that have phases close to that of the perturbation remain relatively unaffected by the perturbation. Three important conclusions can thus be drawn from these graphs; (1) there is a phase dependence on the phase shifts in the experiment where cells become entrained (2) most of the entraining phase shifts occur during the first perturbations and at the consequent perturbations the cells have already become entrained (3) there is no phase dependence on the phase shift in the control experiment. Phase synchronisation assumes that the physical properties of the entrained oscillator remain essentially unchanged and that entrainment is due to phase shifts caused by the perturbation. To test whether the frequencies of the oscillations become affected by the perturbations, the mean and standard deviation of the frequencies were calculated before and after the perturbation interval (Fig. 6a). No significant difference could be seen between the experiments where entrainment was induced and the control experiment, neither with regard to the actual frequencies of the oscillations, nor with regard to the distribution of the frequencies. The amplitudes of the oscillations also remained relatively unaffected by the perturbation for oscillatory cells. However, also non-oscillatory cells became entrained by the perturbation (Fig. 5). For this response to be detectable, also the amplitude must become affected. This indicates that phase changes are not sufficient to entrain non-oscillatory cells. In contrast to results of model calculations 26 , under no circumstances could entrainment out-of-phase with the periodic perturbation be observed, neither for oscillatory nor non-oscillatory cells.
Robustness of the synchronisation mechanism. The degree of synchronisation is quantified by the so-called order parameter r(t) (see Methods). It increases very rapidly in the perturbation interval and already within the first minute of perturbations, the order parameter reaches a value close to unity, corresponding to complete synchrony (Fig. 1c). In the absence of external perturbation the order parameter is expected to tend to zero, assuming that there are no cell-cell interactions that could stabilise synchronisation. However, this is true only provided that the number of cells N used in the analysis is large. In general, we expect the order parameter to be of the order of 1= ffiffiffiffi N p , corresponding well with the measured value in the absence of perturbations (Figs. 1c and 2c).
How quickly the order parameter decays after the periodic perturbation is switched off depends upon the distribution of the frequencies of the individual cells. We expect that the order parameter decays as r(t) 5 r 0 e 2Dvt , see Methods. This expectation is borne out by our experiments (Fig. 6b-d). Note also that since the oscillations are initially induced simultaneously in all cells, it is expected that the order parameter should be high at the start of the experiment and then decay if there are no cell-cell interactions (Figs. 1c and 2c). The decay of the order parameter at the start of the experiment cannot, however, be well described simply by the distribution of frequencies at the start of the experiment. It can instead be explained by a measured difference in duration of the initial spike of NADH, causing cells with the same frequency to become out of phase already within the first oscillation period.
Universality of the synchronisation mechanism. It remains to determine how cyanide can work as a synchronising agent and whether this mechanism is likely to be universal. There are several possible explanations as to why removal of cyanide may have a synchronising effect. One reason could be that respiration is no longer inhibited when cyanide is removed. To cause the observed response, which occurs within a few seconds, it would be necessary that cyanide be released very rapidly from its interaction with cytochrome c oxidase. To investigate whether respiratory reactions are responsible for the synchronisation, the experiment was repeated by changing from glucose/cyanide to glucose using a hypoxic glucose solution. However, also here entrainment could be seen (data not shown). This indicates that the main cause of entrainment at cyanide removal is not due to respiratory reactions. However, care has to be taken when performing experiments using hypoxic media, since yeast cells can respire also in very oxygen poor environments. Even low concentrations of oxygen in the hypoxic medium could cause cell responses when cyanide is removed.
A second reason for the response could be that the pH is altered when cyanide is added. In previous studies, Poulsen et al. induced oscillations using 4 mM KOH and hypoxic solutions instead of cyanide 35 . In their studies, the addition of KOH caused a similar pH increase as the addition of cyanide, raising the pH value 0.06 units. Measurements of the pH of our solutions showed that addition of 5 mM cyanide increased the pH value with 0.09 units. To determine whether the cells respond to such a change in pH, the measurement was repeated with perturbations with a hypoxic glucose solution without cyanide, where the pH was adjusted using KOH. This experiment showed that the cells still became entrained by a change from glucose/cyanide to glucose at hypoxic conditions, also when the pH was kept constant (data not shown). It thus seems unlikely that the pH difference is causing the response at cyanide removal.
A third reason for the observed synchronisation could be cyanide binding to intracellular ACA. The phase dependence of the phase shift differs for different chemical species 13 . Our data reveals that the (1) and Table 1 for parameter values), which reveals a significant difference between the values of the quadratic terms. Cells that have phases close to 2p or p become negatively phase shifted when cyanide is removed, while cells which have phases close to that of the perturbation remain relatively unaffected by the perturbation. phase response to cyanide removal is comparable to previous studies of synchronisation by ACA addition in populations 15 and in single cells ( Supplementary Fig. S1 online), while population studies of perturbations with e.g. glucose reveals a different phase relation 9 . This supports the hypothesis that the removal of cyanide causes a similar response as addition of ACA.
Discussion
The experiments reported here show that the mechanism behind the synchronisation of individual oscillatory yeast cells is phase synchro-nisation rather than frequency or amplitude modulation. Our data show that the strongest phase response is achieved when the concentration of NADH is at a minimum, where the removal of cyanide prolongs the subsequent period roughly by one third. Oscillatory cells continue to oscillate with frequencies and frequency distributions largely unaffected by the periodic perturbation. Also the amplitude of the oscillations remains relatively unaffected by the perturbations. These results support the paradigm originally put forward by Kuramoto and his collaborators (for a review see 39 ): that periodic perturbations can entrain oscillators by affecting just the phases of the oscillators, resulting in a robust and universal mechanism. Theoretical models tend to be very sensitive to changes in parameter values and using different parameters can result in in-or out-of-phase synchronisation of the oscillations. In contrast to some model results 26 , our results show no cases of entrainment out-ofphase, neither for oscillatory nor non-oscillatory cells.
Our experiments show that non-oscillatory cells also become entrained by external periodic perturbations, so these cells must also contribute to macroscopic oscillations observed in e.g. Ref. 40. However, in contrast to the entrainment of oscillatory cells, which relies solely on phase synchronisation, entrainment of non-oscillatory cells requires a change of the amplitude (as well as the frequency). In this case the entrainment cannot be explained solely in terms of phase shifts.
The graphs in Figure 3 reveal temporal information about the entrainment of the individual cells, where the clustering of phases can be seen to occur already during the first few perturbations. This is consistent with the rapid increase of the order parameter (Fig. 1c). Heterogeneity is present in the phase response of individual cells, but despite the heterogeneity the majority of oscillatory cells eventually become entrained by the perturbation. The final phase distribution is within 20% of the original distribution. This shows that the synchronisation mechanism is very robust with regard to cell heterogeneity. This result is in stark contrast with results of model calculations 24,25,30 .
Our results indicate that cyanide can work as a synchronising agent due to its ability to bind intracellular ACA. Removal of cyanide will then have the same effect as addition of ACA, where the response is transduced to the first part of glycolysis via NADH/NAD 1 and ATP/ADP 9,40,41 . Here it is thus impossible to completely separate the cell responses due to cyanide removal and ACA addition. Different externally applied chemical species thus appear to synchronise individual cells by similar mechanisms, giving further support to the hypothesis that the synchronisation mechanism is universal. Intracellular cyanide reactions are rather complex. That cyanide can act in more ways than by binding ACA and inhibit respiration has been indicated in previous studies, where the oscillatory tendency was stronger when cyanide was added than under hypoxic conditions 34 and under hypoxic conditions where ACA was flushed away by a flow 42 . And indeed, recently Hald et al. showed that cyanide also reacts with other metabolites, namely pyruvate and dihydroxyacetone phosphate (DHAP), and that cyanide might affect the behavior of glycolytic oscillations in more ways than just by binding ACA and inhibiting respiration 36 . Previous studies have also shown that cyanide causes longer trains of oscillations than other inhibitors of respiration, such as antimycin A and azide 43,44 and that oscillations disappear if both cyanide and azide are present 35 . The role of cyanide inhibiting respiration by binding to cytochrome c oxidase and the contribution of respiratory reactions to the oscillatory behavior have recently been discussed by Schrøder et al., who found that oscillations disappear for strains with deletions of a gene coding for subunit VI of cytochrome c oxidase 45 . This finding was unexpected, since the general assumption is that cytochrome c oxidase is completely inhibited at cyanide concentrations used for oscillation studies and should not contribute at all to the oscillatory behavior of the cells. Cyanide reactions with glucose has, however, been shown to be negligible in the pH range of our experiments 36 . These results highlight the need for further studies to fully understand intracellular cyanide reactions. In summary, we report on experimental results that allow us to determine the mechanism, the robustness, and the universality of synchronisation of glycolytic oscillations in yeast cells. Our results pose challenges to theoretical modeling, namely to reproduce our results by simulations of the dynamics of glycolytic reaction networks, and to explain the robustness and universality of the synchronisation mechanism. A successful model must show (1) the phase responses to periodic perturbations reported in this work, (2) the robustness of phase synchronisation also for the cell heterogeneity reported here, and (3) the universality of the mechanism; a wide range of periodic perturbations entrain the oscillations. Our experiments are the first that show entrainment of individual yeast cells by periodic perturbations. They support the paradigm originally put forward by Kuramoto: that periodic perturbations can entrain oscillators by affecting just the phases of the oscillators, resulting in a robust and universal mechanism.
Methods
Cell preparation. Yeast cells (Saccharomyces cerevisiae, X2180), were grown in a medium comprising 10 g/l glucose, 6.7 g/l yeast nitrogen base and 100 mM potassium phthalate (pH 5.0) on a rotary shaker at 30uC until glucose was exhausted, as described in 46 . The cells were harvested by centrifugation (3500 g for 5 min), washed twice in 100 mM potassium phosphate buffer (pH 6.8) and subsequently glucose starved in the buffer solution on a rotary shaker for 3 h at 30uC. The cells were then washed in the buffer solution and stored at 4uC until use.
Experimental procedure. The experimental setup and procedure was described in detail in 42 . The cells were introduced into a microfluidic flow chamber with four inlet channels 42 and positioned at the bottom of the chamber in arrays of 535 cells with a cell-cell distance of 10 mm using optical tweezers 42,47,48 . By positioning the cells far apart, cell-cell interactions by e.g. ACA was avoided 28 . The flows in the channels were laminar and solutions from different channels could only mix by diffusion. The extracellular environment in the cell array area could thus be controlled both spatially and temporally by adjusting the flow rates in the different inlet channels, causing the cell array area to be covered by the chemicals from the intended channel. A complete change of environment was achieved within 4 s of a change in flow rates. All channels contained 100 mM potassium phosphate buffer (pH 6.8) and the cells were introduced in the lowest channel. Glycolytic oscillations were induced by a 20 mM glucose 1 5 mM KCN solution, introduced in the second highest channel for 10 min before the periodic perturbations started and for 15 min after the perturbations had ended. Entrainment was investigated by periodically introducing the solution of interest in the top channel for 20 s and the glucose/cyanide solution for 20 s, giving a total period time of the perturbation of 40 s. The perturbations continued for 15 periods (10 min). For perturbations with ACA, a solution comprising 20 mM glucose 1 5 mM KCN 1 1 mM ACA was used. For experiments performed with hypoxic solutions, the solutions were bubbled through with N 2 -gas for 10 min before the experiment. The cell responses were studied by imaging the NADH autofluorescence from the individual cells using an epi-fluorescence microscope (DMI6000B, Leica Microsystems) and an EM-CCD camera (C9100-12, Hamamatsu Photonics), where images were acquired every other second 42 .
Data analysis. The acquired images were processed and analysed as described previously 42 . The resulting signals of the NADH fluorescence intensity of individual cells were then analysed using MATLAB (The MathWorks, Inc.). Running averages of the NADH signals were calculated using a window of approximately two periods (55 data points) and were subtracted from the NADH signals to reduce spurious trends and drift ( Supplementary Fig. S2 online). This facilitated extraction of the phases of the oscillations. The instantaneous phases of the oscillations of the individual cells, w n (t), were extracted from the NADH signals by means of a standard procedure that is based upon evaluating the Hilbert transform of the signals in MATLAB 38 . The phases are defined in the interval [2p, p] and a phase of 0 represents a maximum in the concentration of NADH (Figs. 1b and 2b). In Figures 1 and 2, the ten cells that showed the most stable oscillations before the perturbation where chosen. The phases before, w before , and 9 s after cyanide removal, w after , were plotted for perturbations 1, 2, 4 and 8, for the experiments where entrainment was induced ( Fig. 3a-d, N 5 20, red dots) and for control experiments where only the flow rates were changed but not the chemicals in the solutions (Fig. 3a-d, N 5 32, black dots). The 9 s delay was chosen to allow the cells time to respond to the perturbation.
The oscillation frequency of the individual cells, v n , was calculated in time intervals 0-5, 5-10 min and 20-25 min as the frequency at the maximum peak of the Fourier transforms of the NADH signals multiplied by 2p, and is given as mean 6 standard deviation for experiments where entrainment was induced (Fig. 6a, N 5 24, red bars) and for the control experiments (Fig. 6a, N 5 20, black bars).
The phase shifts of the data sets at the first perturbation were calculated as Dw 5 w after 2 w before . If jDwj . p, Dw was adjusted with 62p to move it into the interval [2p, p]. The phase shifts were then corrected by subtracting the expected phase shifts of each cell, v n Dt, where Dt is the time between the measurement of the phase before and after the perturbation. This corrected phase shift was fitted by a second-degree polynomial on the form: Dw{v n Dt~c 2 w 2 zc 1 wzc 0 , ð1Þ where the parameter values are estimated as mean values and their uncertainty (due to the spread of data points and the finite sample size) is expressed in terms of their 95% confidence interval ( Fig. 4 and Table 1). The degree of synchronisation was characterised by the order parameter r(t), defined as 31,49 where N is the total number of cells in the experiment (N 5 10 in Figs. 1, 2 and 6c-d, and N 5 14 in Fig. 6b). An order parameter close to unity indicates a high degree of synchronisation, while an order parameter close to zero indicates large heterogeneity in phases among the individual cells and thus low entrainment by the external, periodic perturbation. When the cells are independent and there is no external perturbation, the order parameter is expected to decay as where r 0 is the order at t 5 0 and Dv is the standard deviation of v n . The frequencies were calculated in time interval 20-25 min and the decay was set to start 18 s after the end of the last perturbation ( Fig. 6b-d).
==
Domain: Physics Biology Chemistry Medicine
|
[{"added": "2015-04-01T08:03:35.000Z", "created": "2015-03-24T00:00:00.000", "id": "16408086", "metadata": {"extfieldsofstudy": ["Biology", "Physics", "Chemistry", "Medicine"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/srep09404.pdf", "pdf_hash": "dff454f5a770f635fbe21749cb0b7ca3fa40203f", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:38549", "s2fieldsofstudy": ["Biology"], "sha1": "660e1b548b565f7b0e1656fc2759b2e91709b072", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of Heterogeneity on Cancer: A Game Theory Perspective
In this study, we explore interactions between cancer cells by using the hawk–dove game. We analyze the heterogeneity of tumors by considering games with populations composed of 2 or 3 types of cell. We determine what strategies are evolutionarily stable in the 2-type and 3-type population games and what the corresponding expected payoffs are. Our results show that the payoff of the best-off cell in the 2-type population game is higher than that of the best-off cell in the 3-type population game. When these mathematical findings are transferred to the field of oncology they suggest that a tumor with low intratumor heterogeneity pursues a more aggressive course than one with high intratumor heterogeneity. Some histological and genomic data on clear cell renal cell carcinomas is consistent with these results. We underline the importance of identifying intratumor heterogeneity in routine practice and suggest that therapeutic strategies that preserve heterogeneity may be promising as they may slow down cancer growth.
Introduction
Precision oncology loses much of its efficiency when significant genetic differences appear between different regions of the same tumor and remain hidden because of incomplete sampling. In terms of therapy this means that some areas of the tumor will respond to treatment while others will not. The reliable identification of such intratumor heterogeneity (ITH) so as to assure the efficacy of current therapies is a cornerstone of modern oncology (Middleton et al. 2021). From a genomic perspective, temporal and spatial heterogeneity develop in every tumor, following different evolutionary models. To date, four different patterns have been identified: Linear, branching, neutral, and punctuated (Davis et al. 2017). Branching and punctuated patterns have been extensively analyzed in clear cell renal cell carcinomas (CCRCC), an aggressive variant of renal cancer (Turajlic et al. 2018b).
It is widely agreed that the punctuated pattern, corresponding to tumors that display low ITH, shows a worse prognosis than the branching pattern, which corresponds to high ITH. Motivated by this finding, we set out to design a game theoretical model that represents it well in oncology and to see whether the solution to that model is consistent with it.
Game theory has recently been identified by 33 expert oncologists as a key model for understanding tumorigenesis and potentially guiding therapy (Dujon et al. 2021). Several games (or a combination of games) have been used to study cancer, including the prisoner's dilemma (West et al. 2016), the hawk-dove game (Tomlinson 1997;McEvoy 2009;Kareva and Karev 2019;Swierniak et al. 2019Swierniak et al. , 2020, coordination games (Bayer et al. 2022) and public good games (Nogales and Zazo 2021).
In particular, evolutionary game theory is a powerful tool for studying cancer as it models how types of cells compete with each other for resource and space, and how their strategies evolve over time. See for instance, Wölfl et al. (2022) for a review of this literature. An evolutionarily stable strategy (ESS) guarantees that no mutant strategy can invade the population (Maynard-Smith and Price 1973;Maynard-Smith and Parker 1976;Maynard-Smith 1982). That is, an ESS is a strategy which, when adopted by a population of individuals, cannot be invaded by any alternative strategy. 1 In the context of cancer, ESS can be used to understand the evolution of tumor cell populations.
In this paper we use a non cooperative game and the ESS as a solution concept. This approach has several advantages. From a mathematical perspective, it avoids having to explain possible evolutionary trajectories that might conflict with empirical oncological findings. From a medical perspective, it is not feasible to conduct a large number of biopsies over time in patients: The evolutionary trajectories of cells over time cannot be assessed in the clinical practice due to deontological reasons. From a game theory perspective, the ESS can be applied in games in normal form -the archetypal model of interactions.
Our contribution consists of developing a game theoretical model that relates ITH to the prognosis of the tumor. More specifically, we start by choosing the hawk-dove game because of its simplicity and its affinity to cell behavior. 2 Hawk-dove games are defined by two parameters: The resource (v) and the cost (c). The resource in tumors consists of a varied spectrum of diffusible factors released into the medium by tumor and/or non-tumor cells. An example is the fibroblast activation protein (FAP) produced and released by a specific sub-type of cancer-associated fibroblasts (Errarte et al. 2020). The cost is the energy required to obtain the resource. The Atkinson level enables the energy spent by any cell, normal or neoplastic, to be measured. It takes into account the relative cytoplasmic concentrations of adenosine tri-(ATP), di-(ADP), and mono-(AMP) phosphate (the essential molecules fueling all cellular processes) (De la Fuente et al. 2014). The hawk-dove game is applied under the assumption that cells do not recognize their own type but detect the type of their opponents. Examples of how cancer cells behave differently depending on the recognition of their respective cellular contexts are available in the literature (Maruyama and Fujita 2017).
Then we use a heterogeneous game (Inarra and Laruelle 2012), i.e. we consider heterogeneous populations composed of two or three types of cells: The 2-type population model represents low ITH (punctuated pattern) and the 3-type population model represents high ITH (branching pattern). Clinical practice suggests that encountering more than three different types of cells in a tumor is quite unusual. Therefore, we do not consider populations composed of four or more types.
Finally, we compare the ESS for the 2-type and 3-type populations models. In the ESS, different types of cells obtain different levels of payoffs. The cells that obtain the highest expected payoffs are the fittest ones, i.e. those that divide at the highest rate, increasing their proportion over time. Thus, these cells determine how aggressive the tumor is. Therefore, we compare the expected payoffs of the fittest cells for the two population games to assess the prognosis of tumors. The findings obtained are consistent with what is observed in histological and genomic studies on CCRCC (Turajlic et al. 2018a, b;Manini et al. 2022).
The paper is organized as follows. Section 2 presents a case study that enables us to develop our 2-type and 3-type game theoretical models. Section 3 sets out the game theoretical models and gives the main results. Section 4 contains a discussion and our conclusions.
A Case Study
In clinical practice, tumors are classified using two types of approaches: Histological and genomic. In the former, tumor classification is based on the morphological features of tumor cells under the microscope when pathologists assign tumor grading following internationally accepted criteria. Most tumors in the body are graded from 1 to 3, or more rarely from 1 to 4, where grade 1 means the lowest aggressiveness and grade 3 (or 4) means the highest aggressiveness. In such analyses, a distinction can also be drawn between low and high ITH.
In genomic analysis, a distinction can be drawn between punctuated and branching cancer evolutionary patterns as discussed in Davis et al. (2017). With the recent development of next-generation sequencing (NGS)-based platforms, it has become feasible to conduct concurrent analysis on hundreds of genes or even the entire genome using small quantities of tumor tissue collected by needle biopsy.
The two types of tumor analysis agree that low ITH (corresponding to the punctuated evolutionary pattern) is more aggressive than high ITH ( corresponding to the branching pattern) (Manini et al. 2022;Turajlic et al. 2018b).
To study the effects of low and high ITH in CCRCC, Manini et al. Manini et al. (2022) have recently reviewed a series of 28 exhaustively sampled CCRCC focusing specifically on the variability of tumor grades. The follow-up on all patients varied between 5 and 10 years. More than 1500 tumor samples were assessed, averaging more than 50 samples per case. In 5 of the 28 cases a single grade was observed in all the samples analyzed. In 15 cases two different grades were observed; and in 8 cases three grades were observed. Cases with two different grades across all the regions analyzed were considered to show low ITH, whereas cases with three different grades were considered as high ITH.
The 23 cases that display some heterogeneity are summarized in Table 1. 3 For each case we give the level of heterogeneity, the percentage of the highest grade and the outcome at last contact. It can be observed that 9 of the 15 patients with tumors showing low ITH died of the disease, 5 were alive with the disease, and 1 was alive without the disease at last contact. By contrast, all 8 patients with tumors showing high ITH were alive without the disease at last contact. Moreover, with the exception of Cases 1 and 16, all tumors showing low ITH also show high proportions of the highest grade cells.
The prognosis of patients with low ITH and high proportions of the most malignant type of cells is bad. Indeed all but two of them died of the disease. Conversely, all the patients with tumors showing high ITH are alive and show small proportions of the highest grade cells.
The Hawk-Dove Game in Tumors
In this section we present our game theoretical approach based on the case study in the previous section. We first introduce the hawk-dove game in a homogenous population. To adapt our modeling to the ITH problem we refer to cells rather than players, and seek to justify every assumption that we make in the cancer setting. Then we introduce the hawk-dove game in populations with 2 different types of cells (referred to as 2type populations) and 3 different types of cells (referred to as 3-type populations) exemplifying interactions within cells in low ITH and high ITH tumors, respectively. Outcome at last contact-AWD, alive with disease; AwoD, alive without disease; DOD, died of disease
Homogenous Population
We consider a tumor formed by a population of n cells. Encounters between cells are bilateral and in each encounter a cell can behave aggressively, like a hawk, or passively, like a dove, to acquire a resource v. If one cell is aggressive and its opponent is passive, the former obtains the resource and the latter gets nothing. If both cells are aggressive there is a fight and the winner gets the resource while the loser bears a cost c > v.
Assuming that they both have the same probability of winning, the expected reward, ("expected payoff" hereafter) for each cell is (v − c)/2. If both cells are passive, one withdraws and gets nothing while the other takes the resource. Assuming that they both have the same probability of withdrawing, the expected payoff for each cell is v/2. These contingencies are summarized in the following payoff matrix.
A strategy denoted by α represents the probability of a cell being aggressive on meeting another cell. A cell can choose to play hawk (α = 1), dove (α = 0) or a mixed strategy (0 < α < 1). The expected payoff of a cell that plays α when its opponent plays β is given by u(α, β): The solution concept applied to solve this game is the evolutionarily stable strategy (ESS). An ESS is a strategy that maximizes the expected payoff of a player when its opponent chooses the same strategy, i.e. a symmetric Nash equilibrium. An ESS also guarantees that no mutant strategy can invade the population. It is known that strategy v/c is the only ESS (Maynard-Smith 1982) when the population is homogeneous, i.e. composed of a single type of cell.
Heterogeneous Population
To analyze the effect of ITH on the fitness of tumors, we need to compare cells with identical capacities in terms of both the level of resources to which they are exposed and their energy cost in fighting for those resources. Notice that a priori types do not confer any advantage: The same payoff matrix is played in every encounter.
Tumors are usually composed of different types of cells. In our modeling we follow the above case study, and consider two levels of ITH: Low and high. For notation purposes, we define A-cells as the best-off cells, i.e. those with the strictly largest expected payoff, thus those that reproduce fastest and define the aggressiveness of a cancer tumor. A tumor with low ITH is composed of two types of cells,
say A-cells and B-cells. A tumor with high ITH is composed of three types of cells, say A-cells, B-cells and E-cells.
As mentioned in the Introduction, cells are sensitive to their environment and recognize the type of their opponent, but not their own type. Hence, a cell sees itself as I -cell, i.e. it does not know its own type, which may be A-cell or B-cell in a 2-type population game or A-cell, B-cell, or E-cell in a 3-type population game.
2-Type Population Game
Consider a heterogeneous tumor formed by A-cells and B-cells. A 2-type population game is denoted by 2 (v, c, x A ), as it can be defined by parameters v, c, and the proportion of A-cells, x A . Cells cannot adopt a different strategy according to their own type but are able to choose a different probability of playing aggressively when facing any type of opponent. 4 Thus, a strategy of a player is a pair (α A , α B ), where α I denotes the probability of behaving aggressively on meeting an I -cell (I = A, B). In other words, α I indicates the level of aggression received by an I -cell. Game 2 (v, c, x A ) is analyzed in Inarra and Laruelle (2012). It is shown that no strategy with α A = α B is evolutionarily stable. The game has two evolutionarily stable strategies in which cells of one type receive less aggression and obtain a larger expected payoff than cells of the other type. 5 We focus on the ESS where the A-cells have the largest expected payoff and face the least aggression. This ESS depends on the proportion of A-cells. We denote it by α * * A , α * * B ESS :
The level of aggression toward the latter increases when the proportion of Acells increases, until full aggression ( α B = 1) is reached (for x A >x A ). Also, if the proportion of A-cells is below a threshold (x A <x A + 1/n), A -cells suffer no aggression (α A = 0) and aggression is concentrated only on B-cells.
Above that threshold, however, A-cells do receive some aggression. Since only the expected payoff of the best-off cells is of interest in this context, we focus our analysis on the expected payoff of those cells in the ESS, denoted by U * * A (x A ), which is given by: Observe that if the proportion of A-cells is below a threshold (x A <x A ), in the ESS the larger x A is, the larger the expected payoff of the A-cells is (see "Appendix A"). Indeed, A-cells receive no aggression while B-cells receive more aggression as x A increases. In consequence, an A-cell more frequently obtains resource v, thus increasing its expected payoff. The expected payoff of an A-cell peaks when x A =x A and starts decreasing thereafter. This finding suggests that A-cells start fighting among themselves for resource v. Figure 1
Consider a heterogeneous tumor formed by A-cells (in a proportion of x A ), E-cells (in a proportion of x E ) and B-cells (in a proportion of x B = 1 − x A − x E ). A 3-type population game is denoted by 3 (v, c, x A , x E ), as defined by parameters v, c, and the proportions of A-cells and E-cells.
A strategy is denoted by (α A , α B , α E ) where α I is the probability of behaving aggressively when facing an I -cell (I = A, B, E). If an I -cell plays (α A , α B , α E ) against an opponent playing (β A , β B , β E ) its expected payoff is the sum of the probability of meeting a cell of its own type, i.e. (nx I − 1)/(n − 1), multiplied by u(α I , β I ), and the probability of meeting a cell of each other cell type, i.e. (nx J )/(n − 1), multiplied by u(α J , β I ). The expected payoff of an I -cell denoted by U I , gives the following expressions: Moreover, as a cell does not know its own type, what can be maximized is the expected payoff of a generic cell, denoted by U . This is computed as the sum of the probability of being I multiplied by the expected payoff of an I -cell with I ∈ {A, B, E}: Substituting (1) into (3) and then substituting (3) into (4), the expected payoff of a generic cell can be written as follows: As in the 2-type game, whenever cells choose the same strategy the cell that receives the largest payoff is the cell that receives the least aggression. That is, U I > U J is equivalent to α I < α J (for I , J ∈ {A, B, E} and I = J ). See "Appendix B" for the proof.
We focus on the ESS where the three different types receive different expected payoffs. To make the comparison with the 2-type game, we set U A > U B > U E . The following proposition, proven in "Appendix C", indicates that the ESS depends on the proportions of the A-cells and E -cells. We denote it by α * * * A , α * * * B , α * * * E ESS .
As expected, A-cells receive less aggression than B-cells, which in turn receive less aggression than E-cells (α A < α B < α E ). The best-off cells receive no aggression, B-cells receive some aggression and E-cells receive full aggression. Figure 2 plots the aggression suffered by B-cells in the ESS for different proportions of A-cells and E-cells, showing that: (i) The larger x A is, the greater the aggression suffered by B-cells is; and (ii) the larger x E is, the lesser the aggression suffered by B-cells is (see "Appendix D").
Recall that the expected payoff of the best-off A-cells is of interest in this context, so we present the expected payoff in the ESS of these cells, denoted by U * * * A (x A , x E ), which is given by: Observe that the larger x A is, the larger the expected payoff of each A-cell is (see "Appendix E"). Indeed, aggression towards B-cells increases while A-cells receive no aggression (and E-cells receive full aggression). As a result, an A-cell obtains resource v more often and its expected payoff therefore increases. By contrast, the larger x E is, the smaller the expected payoff of each A-cell is. As x E increases two opposite effects come into play: On the one hand there are more E -cells, which receive full aggression, so overall aggression levels increase, which is beneficial for A-cells. On the other hand, as x E increases the aggression received by B-cells decreases, which reduces overall aggression, which is detrimental for A-cells. This negative effect turns out to outweigh the positive effect, so the expected payoff of each A -cell decreases with x E .
An important point to highlight here is that there is no ESS with U A > U B > U E when the proportion of A-cells or E-cells exceeds thresholdsx A orx E . However, having x A >x A or x E >x E does not rule out the existence of ESS. In these cases two cell types receive the same expected payoff: Either U A > U B = U E or U A = U B > U E , reproducing the outcomes of the 2-type population game described above. This can be illustrated as follows for the case U A > U B = U E .
In this ESS, the E-cells are not distinguished from the B-cells (α B
= α E ). The level of aggression received and the expected payoffs are identical to those obtained in the ESS in the 2-type population game. This is equivalent to a high ITH tumor evolving into Table 1 every case in which the proportion of the highest grade is large shows a tumor with low ITH, except those of the patients described in Cases 1 and 16.
Comparison of Heterogeneous Tumors
As mentioned, the main question here concerns the comparison of the expected payoff of the best-off A-cells in the 2-type and 3-type population games. Clearly, the comparison is pertinent for equal proportions of A-cells in the two tumors under analysis. The following proposition (see "Appendix F") indicates when the expected payoff of an A-cell in ESS is strictly larger in the 2-type population game than in the 3-type population game. Fig. 3
Discussion
Below, we discuss the consistency of our results obtained from extending the classic hawk-dove game to a heterogeneous population as outlined in Sect. 3 with the observations in the histological study on CCRCC described in Sect. 2.
To bridge the results from Sects. 2 and 3, recall that following the empirical oncology study in Manini et al. (2022), ITH is considered as low when the tumor has two different cell populations (grades) measured by histological parameters. Such tumors are represented in our model by 2-type population games. When the tumor has three or more identifiable cell populations, ITH is considered to be high. This case is represented by 3-type population games. In Table 1 the proportion of the highest grade can be interpreted as the proportion of the best-off type, that is, x A .
Proposition 1 says that in a 3-type population game an ESS exists when the proportions of the best-off and worst-off cells do not exceed certain thresholds. Beyond those thresholds, for the best-off and worst-off cells, the ESS has two types of cells that receive exactly the same aggression and therefore have the same expected payoff (see Proposition 2). In accordance with these results in the case study we do not expect to find patients with tumors that show high ITH and a large proportion of the highest grade cells. This is indeed the case and is reflected in Table 1. Indeed, high ITH is linked to proportions of x A ≤ 20%. From a genomic point of view, this result could be interpreted as a branching-type tumor becoming punctuated. Such an evolution has in fact been detected in several types of malignant tumors (Bao et al. 2021).
Proposition 3 says that the best-off cell type obtains a larger expected payoff in 2type population games than in 3-type population games. This means that if two tumors with the same proportion of the highest grade are compared, the one with the higher ITH should be less aggressive than the one with the lower ITH. In other words, for the same proportion of the highest grade a three-cell-type tumor pursues a less aggressive course than a two-cell-type one. This result is also consistent with the findings in the case-study: Patients with cancer showing high ITH seem to have a better prognosis than those with cancers showing low ITH. Figures 1 and 3 can be interpreted as follows: the higher the proportion of the highest grade cells, the more aggressive a tumor is (and the worse the prognosis is). Moreover, the rate is increasing. This means that the larger the proportion is, the larger the effect on the worsening of the prognosis is. More detailed data would be useful to check this conjecture.
At a practical level, our results suggest the idea that it is appropriate to preserve high ITH in tumors as a promising therapeutic strategy. Thus, a therapy using the tumor containment strategy instead of the conventional of maximum tolerable dose could be more effective since the former procedure forces tumor cells to divide its energy expenditure in two different tasks to survive, i.e. maintaining the tumor growth on one hand, and developing resistances to therapy on the other. Since the total amount of energy into the cell is limited by definition, both tasks will slow down allowing longer cancer survival rates in patients [24]. Moreover, the maximum tolerable dose has already been questioned using a game theory approach (Archetti 2021).
The extension of the classic game of hawk and dove to heterogeneous populations seems to support the results presented in the case study. We believe that our contribution opens up a path for future research into developing case studies where low and high ITH in other types of cancer are analyzed in a way that enables the robustness of the hawk-dove game model to be confirmed for heterogeneous populations.
From a clinical point of view, the results highlight the importance of therapies focused on maintaining high levels of intra-tumor heterogeneity, with a greater diversity of cancer cells, in order to try to slow down the progress of cancer and decrease its clinical aggressiveness.
given that x E < (1 − 1 > 0 given that Thus, for x A <x A and x E <x E , we have U * * A (x A ) > U * * * A (x A , x E ).
==
Domain: Physics Mathematics Biology Medicine
|
[{"added": "2022-02-07T02:15:13.509Z", "created": "2022-02-04T00:00:00.000", "id": "246607823", "metadata": {"extfieldsofstudy": ["Biology", "Physics", "Medicine"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://link.springer.com/content/pdf/10.1007/s11538-023-01178-9.pdf", "pdf_hash": "d93d64f50c11d2645b847f7c7c24457c0eda9ae5", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39472", "s2fieldsofstudy": ["Medicine", "Mathematics", "Biology"], "sha1": "0294a07cc156455c18c019b923ed2866c3f5918b", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Mechanisms of Synchrony in Heterogeneous Inhibitory Interneuronal Networks for Type 1 versus Type 2 Excitability
PV+ fast spiking basket interneurons are often implicated in gamma rhythms. Here we focus on mechanisms present in purely inhibitory networks. Neurons with type 1 excitability can fire arbitrarily slowly, whereas those with type 2 excitability cannot fire below a minimum frequency. We systematically examine how excitability type affects synchronization of individual spikes to a population rhythm in the presence of heterogeneity and noise, using model neurons of each type with matched F/I curve, input resistance, time constant and action potential shape. Population synchrony in noisy heterogeneous networks is maintained because neurons either fire within a tight time window or skip that cycle. Type 2 neurons with hyperpolarizing inhibition skip cycles due to their intrinsic dynamics; we show here the cycle skipping mechanism for type 1 neurons or type 2 neurons with shunting inhibition is synaptic and not intrinsic. Type 2 neurons are more resistant than type 1 to partial and complete suppression in networks with hyperpolarizing inhibition that exhibit network gamma. Moreover, type 2 neurons are recruited more rapidly and more completely into theta-nested gamma. In contrast, type 1 networks perform better with shunting inhibition on both counts, because the nonlinear dynamics in that case favor suppression of type 2 compared to type 1 neurons. Conductances that control excitability type may provide a therapeutic target to improve spatial and working memory and other tasks that rely on gamma synchrony or phase amplitude coupling. Author Summary The collective, synchronized activity of neurons produces brain rhythms. These rhythms are thought to subserve cognitive functions such as attention and memory encoding and retrieval. Faster rhythms are nested in slower rhythms as a putative way of chunking information. A subset of neurons called fast spiking basket cells tend to inhibit other neurons from firing. These neurons play an important role in oscillations, and in the coupling of faster oscillations to slower ones. In some brain regions these neurons can fire arbitrarily slowly (type 1 dynamics) whereas in others they cannot fire below a minimum cutoff frequency (type 2 dynamics). Mathematically, these distinct origins of rhythmic firing are signatures of very different dynamics. Here, we show that these distinct excitability types affect the ability of networks of these neurons to synchronize their fast oscillatory activity, as well as the ability of slower oscillations to modulate these fast oscillations. The exact nature of the inhibitory coupling, which may vary between brain regions, determines which type synchronizes better and is modulated better.
Abstract PV+ fast spiking basket interneurons are often implicated in gamma rhythms. Here we focus on mechanisms present in purely inhibitory networks. Neurons with type 1 excitability can fire arbitrarily slowly, whereas those with type 2 excitability cannot fire below a minimum frequency. We systematically examine how excitability type affects synchronization of individual spikes to a population rhythm in the presence of heterogeneity and noise, using model neurons of each type with matched F/I curve, input resistance, time constant and action potential shape. Population synchrony in noisy heterogeneous networks is maintained because neurons either fire within a tight time window or skip that cycle. Type 2 neurons with hyperpolarizing inhibition skip cycles due to their intrinsic dynamics; we show here the cycle skipping mechanism for type 1 neurons or type 2 neurons with shunting inhibition is synaptic and not intrinsic. Type 2 neurons are more resistant than type 1 to partial and complete suppression in networks with hyperpolarizing inhibition that exhibit network gamma. Moreover, type 2 neurons are recruited more rapidly and more completely into theta-nested gamma. In contrast, type 1 networks perform better with shunting inhibition on both counts, because the nonlinear dynamics in that case favor suppression of type 2 compared to type 1 neurons. Conductances that control excitability type may provide a therapeutic target to improve spatial and working memory and other tasks that rely on gamma synchrony or phase amplitude coupling.
Author Summary
The collective, synchronized activity of neurons produces brain rhythms. These rhythms are thought to subserve cognitive functions such as attention and memory encoding and retrieval. Faster rhythms are nested in slower rhythms as a putative way of chunking information. A subset of neurons called fast spiking basket cells tend to inhibit other neurons from firing. These neurons play an important role in oscillations, and in the coupling of faster oscillations to slower ones. In some brain regions these neurons can fire arbitrarily slowly (type 1 dynamics) whereas in others they cannot fire below a minimum cutoff frequency (type 2 dynamics). Mathematically, these distinct origins of rhythmic firing are signatures of very different dynamics. Here, we show that these distinct excitability types affect the ability of networks of these neurons to synchronize their fast oscillatory activity, as well as the ability of slower oscillations to modulate these fast oscillations. The exact nature of the inhibitory coupling, which may vary between brain regions, determines which type synchronizes better and is modulated better.
Gamma frequency oscillations are thought to serve as substrates for working memory, conceptual categorization, and attention [9], and are altered in psychiatric disorders, for example schizophrenia, dementia, and autism [10]. PV+ FS basket cells play an important role in theta nested gamma [11]. Moreover, nesting of gamma within theta has been proposed a substrate for episodic memory [12] and disruption of theta nested gamma has been proposed to explain deficits in spatial memory in temporal lobe epilepsy [13,14].
Here we focus on mechanisms present in purely inhibitory networks, and the role of excitability type of the FS interneurons in gamma frequency synchronization and cross frequency synchronization with theta oscillations. Neurons with Hodgkin's type 1 excitability are able to spike arbitrarily slowly, whereas those with type 2 excitability have an abrupt onset of repetitive firing that cannot be maintained below a threshold frequency [15,16]. Our recent work [17] shows that FS neurons likely exhibit type 2 excitability in the medial entorhinal cortex, and others have shown that they exhibit type 2 excitability in striatum [18] and neocortex [19]. However, previous work supports type 1 excitability in hippocampal area CA1 [20][21][22]. We will focus on reduced, 2D models here.
The advantage of 2D models is that they can be visualized in a two-dimensional phase plane that reveals the essential features of the intrinsic dynamics. For type 2 excitability, there is a circular flow near the associated subcritical Hopf bifurcation that results in subthreshold oscillations, postinhibitory rebound inherent in the spike generating mechanism, and a resonant spiking frequency.
These features are absent in neurons with type 1 excitability. In contrast, type 1 neurons rely on a saddle-node-on-an-invariant-circle (SNIC) bifurcation, which creates a slow channel in the phase plane where the net current is very small, allowing the neurons to spike arbitrarily slowly.
Previously [17], we noted that cycle skipping was a way for a network to preserve synchrony of individual spikes with the population in the presence of heterogeneity and noise. In other words, spikes that would have occurred too late to be in synchrony with the population are suppressed.
In the case of type 2 excitability, this can be accomplished via the intrinsic dynamics, because in all regimes (excitable, bistable and oscillatory) there is a boundary between trajectories that lead to spiking and those that lead to cycle skipping. However, we did not calibrate the models we compared in such a way as to maximize the fairness of the comparison. Moreover, we only looked at hyperpolarizing and not shunting inhibition.
Fast, ionotropic inhibition in the central nervous system in generally mediated by GABAA receptors, with chloride ions as the charge carrier. The reversal potential of these channels depends on the intracellular concentration of chloride. For quiescent neurons, for hyperpolarizing inhibition the synaptic reversal potential that is negative to the resting potential, whereas for shunting inhibition the reversal potential is more positive than the resting potential but negative to the spike threshold. For oscillatory neurons, the membrane potential during the interspike interval substitutes for the resting potential [23]. Shunting inhibition is sometimes defined as an increase in synaptic conductance in the absence of an obvious change in membrane potential [24]. In our model neurons, a reversal potential of -75 mV produces hyperpolarization whereas -65 mV does not, so we classify the latter as shunting in our context. The reversal potential for chloride in fast spiking basket cells in vitro is about -52 mV [25] in the dentate gyrus, which implies shunting inhibition in those cells. Previous studies in CA1 and CA3 concluded that inhibition between basket cells was hyperpolarizing [26,27]. However, the Cl-concentration can be modulated [28], and there can be intracellular gradients in chloride concentration [29]. Therefore, there is sufficient uncertainty regarding the precise reversal potential of chloride at synapses between interneurons to warrant a systematic study of both types of inhibition.
Here, we systematically examine how excitability type affects synchronization of individual spikes to a population rhythm in the presence of heterogeneity and noise. Simple 2D model neurons of each type were calibrated to have a very similar F/I curve, input resistance, time constant and action potential shape in order to isolate the consequences of excitability type alone on the robustness of synchronization in inhibitory interneuronal networks.
Phenomenological model of type 1 and type 2 excitability
We began with the standard 2D reduction [30] of the 4D Hodgkin-Huxley model [31]. The activation of the voltage-dependent sodium current was assumed to be fast and set to its steady-state value with respect to the membrane potential ( . The inactivation variable for the voltagedependent Na + current ( h ) was yoked to the variable ( n ) for the activation of the delayed rectifier K + current, under the assumption that the slow time scale for these variables was similar. This 2D system with one fast variable (membrane potential, v ) and one slow variable ( n ) is amenable to phase plane analysis under fast/slow assumptions [32]: Figure 1 shows a phase plane analysis of the bifurcations underlying type 1 (Fig. 1A) and type 2 ( Fig 1B) excitability. In the phase plane, fixed points occur at the intersection of the n-nullcline (gray curve), which is simply the steady state activation curve for the n activation variable, and the voltage nullcline at rest (red curve), which is the set of values of n and v for which the net ionic current plus any applied current is zero. The voltage nullcline is N-shaped with three branches, but in order to emphasize the bifurcation point, only the stable left and unstable middle branch and their associated fixed points are shown in Fig 1. For this fast/slow system the leftmost branch is stable, and the middle branch is unstable due to the autocatalytic sodium current. At rest (with no applied current), the fixed point (filled circle) on the left branch is stable and determines the resting potential at -68 mV for both the type 1 (Fig. 1A1) and type 2 ( Fig. 1B1) cases.
The first important difference is that for type 2 there is a single fixed point at all values of applied current, whereas for type 1 the number of fixed points depends upon the level of the applied current.
For type 1 at rest (Fig. 1A1), in addition to the stable fixed point that determines the resting potential, there is an unstable fixed (open circle) point on the middle branch, as well as another one on the unstable branch that is not shown. As the applied current is increased to the stable and unstable fixed points shown collide (Fig. 1A2) and form a saddle node (circle half filled). This collision coincides with the birth of limit cycle surrounding the rightmost unstable fixed point on the unstable branch (not shown), therefore the onset of spiking occurs at a saddle-nodeon-an-invariant-limit -cycle (SNIC) bifurcation [16,33]. As the applied current is increased further and the colliding fixed points annihilate each other. The neuron is left with no stable resting potential. The intersection of the limit cycle with the saddle node gives rise to a trajectory with infinite period. As the applied current is increased further, the trajectory must pass through a gap between the two nullclines; the rate of passage though this bottleneck is arbitrarily slow with increasing proximity to the bifurcation, hence the arbitrarily slow frequencies obtainable by type 1 neurons shown in Fig. 2A (black dots).
In contrast, for type 2 excitability the Hopf bifurcation occurs when the lone stable fixed point corresponding to the rest potential loses stability (Fig. 1B1). When the applied depolarizing current , the fixed point (Fig. 1B2, open circle) moves past the trough of the nullcline onto the unstable branch of the voltage nullcline. The Hopf bifurcation is subcritical [16,33,34] because in the range of input currents between 2 1.74, 2.11 / I A cm the model exhibits bistability between stable limit cycle corresponding to repetitive spiking and a stable fixed point ( Fig. 2A, overlap in the red cross at 0 frequency and at a nonzero frequency). These attractors are separated by unstable limit cycle (not shown). The stable and unstable limit cycles collide and annihilate each other in a saddle node of periodics at . Excitability type is defined as the response of a quiescent neuron as the applied current is increased [15]; the sudden onset of spiking as the applied current is increased occurs at about 30 Hz when the quiescent state loses stability as shown in Fig. 1B.
Steady State Synchrony in Oscillatory Networks
We constructed networks of 300 sparsely and randomly connected neurons with heterogeneity in frequency by distributing the bias current uniformly along a region of the F/I curve that spanned a 20 Hz range (see black bars in Fig. 2A). The conduction delays between neurons were also uniformly distributed between 0.7 and 3.5 ms. In our previous work [35] in a homogeneous network, delays on the short end of this range favor a solution with two subclusters in antiphase, whereas delays at the longer end of the range favor global synchrony of a single cluster. Using a mixture of delays results in solutions that are not obviously one or two clusters, but are transitional between these two extremes. Fig. 3A and B) in is much flatter for type 1 than type 2, because the type 2 histogram is skewed to higher participation rates.
Steady State Synchrony in Oscillatory Networks-Hyperpolarizing Inhibition
This tendency for greater suppression of type 1 neurons with hyperpolarizing inhibition was pre- Fig. 1A1) at the intersection with the n-nullcline. The unstable branch of the V-nullcline (arrow indicating firing threshold points to this branch) separates neurons into two groups. Neurons whose trajectory has already moved to the right of this branch will fire an action potential, but those that fall on the left side of this boundary when the channel closes will skip this network cycle. In general, the neurons that skip are the slower neurons; they will move leftward towards the stable fixed point. The two groups are clearly shown in For example, there is no slow channel. Instead, as explained in Fig. 2A, near a subcritical Hopf bifurcation, a bistable region exists in which quiescence at a stable fixed point (arrow) co-exists with repetitive spiking at the same value of net applied current. Whether the neuron is silent or active depends on recent history, meaning the current location of the trajectory. Specifically, the red closed curve in Fig. 4B1 is an unstable limit cycle that divides the neurons into two groups.
Those inside the red curve will spiral into the stable fixed point at its center, whereas those outside will curve around it to fire an action potential. Thus the mechanism for cycle skipping has a component that results from the intrinsic dynamics. There is no bias toward faster cells, all neurons are almost equally likely to fall outside of the quiescent zone and fire an action potential. Fig. 4B2 shows that as neurons move to the right and fire actions potentials, inhibition accumulates as in
Steady State Synchrony in Oscillatory Networks-Shunting Inhibition
The results in the previous section were for hyperpolarizing inhibition. For both type 1 and type 2, the bifurcation that gives rise to spiking occurs at about -64 mV. Hyperpolarizing inhibition with a reversal potential of -75 mV hyperpolarizes the membrane at most points during the interspike interval except during the trough of the AHP. Shunting inhibition with a reversal potential of -65 mV does not produce big changes in the membrane potential. 4B2 is that the quasithreshold described above (red curve in Fig. 6B2) is now visible as the "ghost" of the unstable limit cycle at the inhibition midpoint. Moreover, the inset in Fig. 6C2 neurons the cutoff frequency is also variable. We incorporated this additional dimension of variability into our simulations using a scale factor for the model dynamics as described in the Methods.
In order to determine the effect of heterogeneity in this parameter, we scaled the dynamics of both variables for each neuron in order to achieve uniformly distributed cutoff frequencies in the range from 10 to 60Hz for type 2 neurons, and used the same range of scale factors for type 1 neurons. We then redid the heatmaps for each type with hyperpolarizing and with shunting inhibition (supplementary figures 1 and 2). Although both types show a reduction in synchronization, the qualitative results that type 2 networks are more resistant to partial and complete suppression for hyperpolarizing but not shunting inhibition still hold.
We also checked to make sure that a saddle-node-not-on-an-invariant-circle (SN) would have qualitatively similar synchronization tendencies. We adjusted the parameters ( 0 3. ms ) of the single type 1 model neuron such that the stable spiking limit cycle is not tangent to the saddle node at the bifurcation. Nonetheless, at synaptic conductance strengths sufficient to create a stable fixed point during network activity, a "bottleneck" or slow channel emerges with the same functional role in organizing network activity as if there were a saddle node on an invariant circle (not shown).
Phase Amplitude Coupling
In the hippocampus, gamma power is maximal when nested in theta oscillations [37]. In order to determine the relative abilities of networks of neurons with type 1 excitability versus type 2 excitability to produce theta-nested gamma, we drove these networks with perfectly sinusoidal inhibitory waveforms at a fixed frequency in the theta range, in the presence of the constant heterogeneous depolarizing bias currents distributed as described in the Methods. The depolarization mimics tonic activation of metabotropic glutamatergic/cholinergic receptors. PV+ basket cells in freely moving rats fire at about 7 Hz during low oscillatory periods, but that rate triples to 21 Hz during theta oscillations [38], presumably due to greater tonic excitation. The sinusoidal drive mimics phasic inhibition from the septum. The amplitude of modulation is set to be the same for each neuron within the population, to reflect the common source of the modulation. The noise and conductance parameters were selected in a regime in which all four types of networks synchronized well. The attributes of the network oscillations for the selected parameter regime are given in Table 3. In additional sets of simulations (not shown), we confirmed that the results shown below are qualitatively similar for wide range of synaptic conductance and noise levels and are not specific for the chosen values. Figure 7C shows the results in the two dimensional parameter space of modulation depth and modulation frequency, with the color in panel C1
Phase Amplitude Coupling-Hyperpolarizing Inhibition
and C2 indicating the un-normalized vector strength as described in the methods. If the vectors were normalized to reflect only how tightly locked the LFP envelope was to the theta drive, the two types perform equally. Removing the normalization reveals the greater recruitment of the population into the nested gamma in type 2 networks. The heatmaps in Fig. 7C1 and C2 show that low frequencies and shallow modulations are most effective, especially for type 2. The heatmap in Fig. 7D show that Type 2 always outperforms type 1 because the red color is always greater than zero.
Almost no theta modulation is evident in the membrane potential traces of individual neurons Fig 7A and B. As the network recovers from inhibitory modulation, the firing rate in the population increases, therefore network inhibition fills in when the external inhibitory drive wanes. This feedback mechanism is also responsible for keeping neurons near the bifurcation at minimal inhibition in steady-state oscillations (see Figs. 4A1 and B1 and 6A1 and B1).
In order to explain the superior performance of type 2 for hyperpolarizing inhibition, we can refer back to the phase plane in Fig. 4A. In order to synchronize, type 1 relies on the accumulation of inhibition as neurons escape through the slow channel, whereas there is no slow channel for type 2 (Fig. 4B). Instead, all neurons have access back to the strong vector fields away from the fixed point and an opportunity to spike. We hypothesize that the improved recruitment of neurons into the gamma oscillation in type 2 networks was because type 1 networks need more time to establish synchrony. We tested this idea in a network in which all neurons had the same applied bias current and therefore the same intrinsic frequency (see Figure 8A). The speed of synchronization is clearer for the homogeneous frequency case because participation is more uniform. A headto-head comparison over two cycles of sinusoidal revel reveals that indeed synchronization begins earlier in the type 2 homogeneous network compared to the type 1 homogeneous network (see leftmost vertical gray bar), which accounts in part for the larger vector strength for theta phase gamma amplitude coupling for type 2 with hyperpolarizing inhibition. The second gray bar shows that the synchrony persists longer for type 2 as well.
The superiority of type 2 for hyperpolarizing inhibition derives from the ability to more quickly recruit the activity of sufficient neurons to establish synchrony at gamma frequencies. The presumption is that the interneurons are in an excited, oscillatory state during theta activity, and are rhythmically inhibited by the septum. For modulation that is too shallow, gamma activity is ongoing and theta power is weak (bottom blue strips on Fig.7 C1 and C2). For modulation that is sufficiently deep to establish theta power, the modulation must be sufficiently slow or shallow that there are still long enough windows of time above the spiking threshold to recruit enough active neurons to establish gamma synchrony (red areas in bottom left half of Fig. 7 C1 and C2).
Phase Amplitude Coupling-Shunting Inhibition
The faster recruitment of type 2 oscillators did not persist for homogeneous networks with shunting inhibition (Fig. 8B). Instead, type 1 oscillators are recruited more quickly into nested theta gamma because shunting inhibition clamps neurons closer to the spiking threshold (see Fig. 6C1 vs. Fig. 4C1). Type 2 networks clearly lose their superiority, likely because of the tendency of type 2 oscillators with shunting inhibition to remain trapped at the fixed point inside the unstable limit cycle ( Figure 6B1 and C2). Figure 9A and B show examples of Type 1 and 2 networks with shunting inhibition. Just as switching from hyperpolarizing to shunting inhibition nullified the advantage of type network in robustness to suppression during steady state synchrony shown in Figures 3 and 5, changing the inhibition from hyperpolarizing to shunting nullifies the advantage of type 2 for phase amplitude coupling as predicted by Fig. 8. In fact, Fig. 9D shows that type 1 networks have better phase amplitude coupling for deep modulations (blue indicates a type 2 -type 1 difference is less than zero). Although type 2 appears to perform better at low modulation amplitudes, the vector strength is so small (blue at the bottom of Fig. 9C1 and C2) as to render this regime unimportant. The tendency of the unstable limit cycle in Fig. 6B1 to trap trajectories clearly reduces the speed of synchronization of type 2, and is likely responsible for its poorer performance with shunting inhibition. Supplementary figures 3 and 4 shows that the PAC tendencies for type 1 versus type 2 are preserved under the second type of heterogeneity utilized in supplementary figures 1 and 2.
Summary of Results
We present two major results. The first is that networks of heterogeneous neural oscillators with type 2 excitability are more resistant to suppression of the slower oscillators than those with type 1 excitability when they are coupled with hyperpolarizing inhibition but not with shunting inhibition. The second result is that theta phase to gamma amplitude coupling is more strongly recruited in the type 2 networks, again for hyperpolarizing but not shunting inhibition. Moreover, we provide mechanistic explanations for these phenomena, which frame conditions for the generality of these results.
The tendency of slower type 1 neurons to be more suppressed than type 2 neurons for hyperpolarizing inhibition relies on two principles. The first is the creation by inhibition of a slow channel near the SNIC bifurcation for type 1. The channel serves to line the fixed points of the heterogeneous oscillators up along the slow nullcline and along the direction of motion of the fast subsystem. This arrangement clearly favors the faster oscillators, and is less prominent for shunting versus hyperpolarizing inhibition.
The second principle is the circular motion about the subcritical Hopf bifurcation for type 2 neurons and the existence of a boundary in the phase plane between spiking and skipping neurons allows for broader participation in steady state network oscillation for hyperpolarizing but not shunting inhibition. The critical assumption is that the stable fixed point just prior to the Hopf bifurcation lies near the shunt reversal potential, but well above the hyperpolarizing reversal potential. This assumption seems reasonable based on the definitions of hyperpolarizing and shunting inhibition combined with the location of the Hopf at the point that positive feedback from the sodium channel destabilizes the rest potential. Thus hyperpolarizing inhibition pushes all neurons leftward (in the fast direction) and the circular motion brings them downward and into the fast vector fields that give neurons a much more equal chance to fire than type 1. However, shunting inhibition, rather than freeing trajectories from the unstable limit cycle surround the stable fixed point just prior to the Hopf, traps them within that limit cycle, favoring suppression.
These principles apply to the transient synchronization in theta-nested gamma as follows. Note that the steady state synchronization and participation are similar between type 1 and type 2.
However, type 2 neurons are more quickly recruited into nested theta gamma for hyperpolarizing inhibition because their intrinsic dynamics do not require accumulation of synaptic inhibition to split the populations into spiking and skipping groups. Faster recruitment also allows synchronization, which requires a minimum number of active neurons, to persist longer. The phase plane portrait also accounts for the slower recruitment of type 2 neurons with shunting inhibition into nested theta gamma and the shorter persistence that accompanies late recruitment. The mechanism is again the tendency of shunting inhibition to clamp the type 2 trajectories inside the unstable limit cycle, which slows recruitment into theta nested gamma.
Generality
A heterogeneous, multicompartmental neuron can be considered as a chain of diffusively coupled oscillators. Although the SNIC bifurcation is generic in single compartment models, studies of heterogeneous oscillators, each exhibiting a SNIC in isolation [39], exhibit a complicated solution structure in a diffusively coupled system. It is not clear that a heterogeneous, multicompartmental neuron can exhibit a SNIC precisely, so it was important to show that the principles described in the preceding section for type 1 neurons still hold for a network in which individual neurons do not have a SNIC, but rather a SN not on an invariant limit cycle. The slow channel still forms.
Another important way in which we checked the generality of the results was to introduce a second type of heterogeneity. We varied the slope of the f/I curve for type 1 neurons and both the cutoff frequency and slope for the type2 neurons by rescaling the temporal dynamics. The cutoff frequency for fast spiking PV+ basket cells [40] had an estimated standard deviation of 40%.
Thus, the 10-60 Hz range in cutoff frequencies explored should map onto biologically plausible distributions of cutoff frequencies. Qualitatively, our major results were preserved under these manipulations ( Supplementary Figs. 1-4).
Previous comparisons of type 1 versus type 2 excitability
Previously Rinzel and Ermentrout [41] contrasted the phase portraits and bifurcation structure for type 1 versus type 2 excitability using the Morris Lecar [42] model in the two regimes. Izhikevich [16] also showed in the phase plane that various minimal conductance-based models could exhibit saddle node and Hopf bifurcations. Neither study explored the implications of the bifurcation type for synchronization.
In the weak coupling regime, Ermentrout and colleages [43,44] found that type 2 neurons receiving noisy common input synchronize better than type 1. Borgers and Walker [45] compared the synchronization tendencies of type 1 [20] and type 2 [46] interneurons embedded in excitatory/in-hibitory networks with type 1 excitatory neurons. They found that when gap junctions were incorporated into their model, there was a sharper transition from pyramidal interneuronal network gamma to interneuronal network gamma in the type 2 networks compared to type 1 networks.
Rich et al [47] found that excitatory/inhibitory networks with type 2 interneurons were more robust to changes in network connectivity compared to type 1. None of the type 1 versus type 2 model comparisons cited above were calibrated to match spike shape, shape of the F/I curve, time constant and input resistance across excitability types as were the models in our study. Thus, we extend these previous comparisons so that any difference in network activity is due to the bifurcation type.
Cycle skipping mechanisms
Cycle skipping to enforce population synchrony within the inhibitory neuron population has been observed in several studies in networks with both type 1 [48], and type 2 interneurons [17,49].
However, the bifurcation structure for the four cases (combinations of type 1 and type 2 excitability with either hyperpolarizing or shunting inhibition) has not been examined in detail previously. It is clear that interneurons do not fire on every gamma cycle in vivo. For example, CA3 FS cells fired at 21 Hz during 45 Hz gamma at 45 Hz, thus they fired roughly on every other cycle [50].
Cycle skipping provides a mechanism by which coupled oscillator models can produce tightly synchronized firing with sparse firing in individual neurons. This mechanism is distinct from a clustering mechanism that was advanced to introduce robustness of population synchrony. In that mechanism, interneurons did not participate on every gamma cycle because synchronized subclusters fired in sequence [51]. Previously [17], we found that cycle skipping more prominent in Type 2 networks, but it was not possible to do a fair comparision across the parameter space with the unmatched models we used in that study. The appropriate conclusion is that cycle skipping is intrinsic to the dynamics of type 2 neurons connected by hyperpolarizing inhibition, but in all other cases synaptic mechanisms predominate over intrinsic mechanisms. Cycle skipping allows robustness to heterogeneity in excitatory drive and connectivity for both type 1 and type 2 networks, overcoming a perceived weakness of coupled oscillator models of network gamma since they were introduced (see next section).
Previous studies on suppression
Computational studies have often shown that fast oscillations based on reciprocal inhibition are exquisitely sensitive to heterogeneities in the network [1,20,23,36]. Inhibitory neural networks generally lose synchrony as heterogeneity is increased in one of two ways, phase dispersion or suppression, depending on the ratio of the time constant for decay of inhibition to the population frequency [36,52] . Here, we looked at fast synapses, which tend to favor the suppression regime over phase dispersion. Early work [20,36,52] on robustness of inhibitory interneuronal synchrony focused of type 1 excitability only. These studies did not include conduction delays, and concluded that hyperpolarizing inhibition and non-physiological low levels of heterogeneity in excitatory drive were required for synchrony. Subsequent studies [25] used the same type 1 Wang and Buzsaki model added a ring topography with conduction delays ranged from 0.7 to 10.5 ms; they found that networks with shunting inhibition were more robust to heterogeneity than with hyperpolarizing inhibition. Their raster plots clearly show more suppression for type 1 for hyperpolarizing compared to shunting inhibition, in agreement with our findings. They repeated their simulation with the type 2 model of Erisir et. al. [46], and obtained similar synchronization results, as in our study.
However, they did not show raster plots or report on suppression in the type 2 networks. We also used a distribution of delays (from 0.7 to 3.5) to prevent the formation of two cluster solutions [35] and stabilize global synchrony by moving the operating point away from the destabilizing discontinuity in the phase resetting curve at 0 and 1.
Therapeutic directions for PAC and cognition
The hippocampal theta rhythm has been shown to be necessary for spatial learning by rats in a water maze [53]. Phase amplitude coupling between theta and gamma likely plays an important role in cognition [54], and we have shown here that type 2 excitability in interneurons strengthens phase amplitude coupling between theta and gamma rhythms for hyperpolarizing but weakens it for shunting inhibition. This suggests the possibility that interneurons could modulate their excitability type according to whether inhibition is hyperpolarizing or shunting, or vice versa. Moreover, this suggests a therapeutic strategy to manipulate excitability type, and thereby theta nested gamma phase amplitude coupling, by targeting currents active in fast spiking interneurons in the subthreshold regime to tip the balance toward outward currents.
The balance of inward and outward currents in the voltage range traversed between action potentials during the interspike interval (ISI) determines whether a neuron exhibits type 1 or type 2 excitability, and is easily modifiable by altering this balance [55,56]. For type 2, outward currents predominate at steady state in this region of subthreshold membrane potential, but are activated more slowly than the inward currents. Therefore, the dwell time in this region of membrane must be brief so that the outward current does not equilibrate and stop the depolarization caused by the inward currents between spikes. There is a maximum ISI for which repetitive spiking can be supported, hence a minimum frequency below which the neuron cannot sustain repetitive firing.
In contrast, inward currents predominate at steady state in the subthreshold range on membrane potentials spanned by the interspike interval, therefore arbitrarily low firing rates can be sustained.
Decreasing inward or increasing outward currents that have a steady state component during the ISI favors type 2 excitability, and the opposite manipulations favor type 1. These manipulations, depending upon whether inhibition is hyperpolarizing or shunting, could increase theta/gamma PAC. Increased theta/gamma phase amplitude coupling may in turn improve some aspects of cognition.
Network
For all simulations we used 300 neurons of the same excitability type, connected by bi-exponential inhibitory synapses. In the network, the input current for each neuron is given by the following equations: were uniformly randomly distributed between 0.7 and 3.5 ms.
Numerical simulations and bifurcation analysis
The bifurcation analysis was performed in XPPAUTO [57].
The network models were implemented as Python 2.7 script for the simulation package NEURON [58]. The integration time step was constant at 0.01 ms. Synaptic activation initially was set to zero for all simulations. The membrane potential of each neuron was initialized randomly from a normally distribution with mean -50mV and standard deviation of 20 mV and the slow n variable was initialized at the steady-state value for the membrane potential. The data presented on vector strength, participation, CV of participation and total suppression for networks biased in the oscillatory regime were averaged over ten trials at each parameter setting for runs of duration 2.5 s with the initial 500 ms ignored to minimize the effects of transients. Each trial had its own random connectivity pattern, random initialization of the state variables, random distribution of bias currents, random delay distribution and random noise sources. For the sinusoidal drive simulations the PAC was averaged over 20 periods of sinusoidal drive, again averaged over 10 trials as described above, but no transients were deleted. Simulations were run on Louisiana Optical Network Infrastructure, QB2-cluster. The source code is available at [URL]-1 (will be publicly accessible upon publication; code for reviewers ZhD7rUh64oD1roKUHc7D).
The phase of each spike within a cycle used was calculated using the length of that particular cycle. Cycle lengths are variable and were computed using the peaks in the population rate as described in Tikidji-Hamburyan et al. [17]. The phase was used to construct the vectors for the vector strength calculation.
In the supplementary figures, we applied a multiplicative scale factor ( i F ) to the rates of change of both variables. Since the Gaussian noise term simulates Brownian motion in the membrane potential, in which distance is proportional to the square root of time, the noise term as then divided by the square root of the scale factor. Taken together, these manipulations simply scale the intrinsic frequency since all intrinsic (but not synaptic) processes are sped up or slowed down equally.
Measure of phase amplitude coupling
Since we apply the theta drive, there is no uncertainty with respect to the phase of the theta oscillation, in contrast to the uncertainty in experimental data such as local field potentials or the EEG. This simplified our analysis. Moreover, since we apply the exact same amplitude of theta modulation to different networks, we were interested not only in the tightness of the coupling of the theta drive and the evoked nested gamma oscillation, but also in the magnitude of the gamma oscillations. Therefore in order to quantify the coupling between theta phase and gamma amplitude, we choose the mean vector length (MVL) [59,60], but without normalization of the amplitude. Table 1, attributes in Table 2) were adjusted to make the comparison as fair as possible. A. Frequency/ current (f/I) curves for type 1 (black dots) and type 2 (red crosses) overlap for the range of bias currents (horizontal bar) used in our heterogeneous networks. The vertical bar shows the range of frequencies exhibited at the heterogeneous bias current levels. The f/I curves were measured using current steps of sufficient duration to allow any transients to die out and establish a steady frequency. Arbitrarily slow frequencies can be obtained for type 1 (not shown) near the bifurcation, but type 2 has a minimum cutoff frequency below which it cannot fire. Between current steps, the membrane potential was returned to its resting value. The bistable range for type 2 is evident from the current values at which a zero frequency quiescent solution coexists with a repetitively firing solution. This region was determined using XPPAUTO. B. The spike shapes and the interspike interval are very similar for type 1 (black curve) and type 2 (red dashed curve). B1. Minimal inhibition. In contrast to the absence of a fixed point in A1, there is a stable fixed point (technically fixed points, vertical arrow) at the intersection of the average V-nullcline and the n-nullcine. This fixed point is surrounded by an unstable limit cycle (red curve) that forms the boundary between spiking and quiescent trajectories. B2. Inhibition midpoint. As in A2 inhibition accumulates as neurons escape to the right and fire action potentials. The right branch of the Vnullcline is now the firing threshold (vertical arrow). B3. Inhibition maximum. As in A3, two groups are evident corresponding to those that fired and those that skipped. C. Summary. C1. This figure shows that the movement of the fixed point as inhibition waxes and wanes is (leftward) in the fast, horizontal direction for type 1. C2. In contrast, the downward and leftward movement of the fixed point for type 2 has a component in the slow direction, which helps equalize the opportunity of slow neurons to fire as compared to fast neurons. In contrast to the absence of a fixed point in A1, there is a stable fixed point at the intersection of the average V-nullcline and the n-nullcine. This fixed point is again surrounded by an unstable limit cycle (red curve) that forms the boundary between spiking and quiescent trajectories. B2. Inhibition midpoint. As in A2 inhibition accumulates as neurons escape to the right and fire action potentials. However in this case the ghost of the unstable limit cycle unfurls into a quasithreshold (red curve) and separates spiking and skipping trajectories (vertical arrow labeling the red curve as the firing threshold). B3. Inhibition maximum. As in A3, two groups are evident corresponding to those that spiked and those that skipped (diagonal arrow) . C. Summary. C1. This figure shows that there is less leftward movement of the fixed point as inhibition waxes and wanes in the fast, horizontal direction for type 1 compared to in Fig. 4C1. C2. The fixed point for type 2 tends to trap trajectories inside the unstable limit cycle (red curve, see blowup in inset), which enforces skipping. Top: Representative sparsely firing single neuron type 2 traces with subthreshold oscillations due to network activity. Middle: Raster plots for 300 neurons with faster neurons (based on I app) shown at the top. Bottom: Simulated LFP consisting of summated inhibitory currents throughout the network. C. Heatmaps for 2D parameter space of modulation depth (given in terms of the peak of the sinusoidal conductance waveform) and frequency. Asterisks show parameter values from A and B. C1. Vector strengths for type 1 networks. C2. Vector strengths for type 2 networks. C3. Difference of vector strengths for type 2 and type 1 networks. Hot colors indicate a difference greater than zero. The difference is well above zero in the lower lefthand cornerplot meaning that the vector strength for type 2 networks is higher. Figure 8. Speed of Recruitment of Theta-Nest Gamma in Homogeneous Inhibitory Networks with Hyperpolarizing Versus Shunting Inhibition. A. Homogeneous inhibitory network with hyperpolarizing inhibition showing that type 2 neurons are recruited more quickly into thetanested gamma (leftmost vertical gray bar) and persist longer (rightmost gray bar). A1. Type 1. A2. Type 2. B. Homogeneous inhibitory network with hyperpolarizing inhibition showing that type 1 neurons are recruited more quickly into theta-nested gamma (leftmost vertical gray bar) and persist longer (rightmost gray bar). B1. Type 1. B2. Type 2. Figure 9. Theta Phase Gamma Amplitude Modulation of Heterogeneous Inhibitory Interneuronal Networks with Shunting Inhibition. One example of sinusoidal modulation of these networks, with a 5 Hz sinusoidal modulation in inhibitory conductance shown at the top for the same synaptic conductance and noise standard deviation as in Fig.7. A. Type 1 networks modulation. Top: Representative sparsely firing single neuron type 1 traces with subthreshold oscillations due to network activity. Middle: Raster plots for 300 neurons with faster neurons (based on I app) shown at the top. Bottom: Simulated LFP consisting of summated inhibitory currents throughout the network. B. Type 2 networks. Top: Representative sparsely firing single neuron type 2 traces with subthreshold oscillations due to network activity. Middle:. Raster plots for 300 neurons with faster neurons (based on I app) shown at the top. Bottom: Simulated LFP consisting of summated inhibitory currents throughout the network. C. Heatmaps for 2D parameter space of modulation depth (given in terms of the peak of the sinusoidal conductance waveform)and frequency. Asterisk shows the parameter values from A and B. C1. Vector strengths for type 1 networks. C2. Vector strengths for type 2 networks. C3. Difference of vector strengths for type 2 versus type 1 networks. Cool colors indicate a difference less than zero. The difference is below zero in this plot, meaning that the vector strength for type 2 networks is higher, except at the bottom where the vector strength is relatively small. Ratio of vector strengths for type 2 versus type 1 networks. Cool colors indicate a ratio less than one. The ratio is always above one in this plot meaning that the vector strength for type 2 networks is higher, as in Figure 7. The additional heterogeneity is the same as in Supplementary Figure 1.
Supplementary Figure 4. Theta Phase Gamma Amplitude Modulation of Heterogeneous Inhibitory Interneuronal Networks with Hyperpolarizing Inhibition with Additional Heterogeneity in Intrinsic Dynamics. Heatmaps for 2D parameter space of modulation depth and frequency. A. Vector strengths for type 1 networks. B. Vector strengths for type 2 networks. C. Ratio of vector strengths for type 2 versus type 1 networks. Cool colors indicate a ratio less than one. The ratio is always below one in this plot as in Figure 9, meaning that the vector strength for type 2 networks is higher, except at the bottom where the vector strength is so small that there is negligible PAC. The additional heterogeneity is the same as in Supplementary Figure 1.
==
Domain: Physics Biology
|
[{"added": "2019-10-17T08:55:33.770Z", "created": "2019-10-11T00:00:00.000", "id": "208590069", "metadata": {"extfieldsofstudy": ["Biology", "Physics"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2019/10/11/802330.full.pdf", "pdf_hash": "d5f126e1fae349a014fb650f23ea7aeefbc3c166", "pdf_src": "BioRxiv", "provenance": "peS2o-0075.json.gz:35483", "s2fieldsofstudy": ["Biology"], "sha1": "934adc8492c32445efa31d08c4b28081f19326cf", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
How a Hemicarcerand Incarcerates Guests at Room Temperature Decoded with Modular Simulations
Hemicarcerands are host molecules created to study constrictive binding with guest molecules for insights into the rules of molecular complexation. However, the molecular dynamics simulations that facilitate such studies have been limited because three-dimensional models of hemicarcerands are tedious to build and their atomic charges are complicated to derive. There have been no molecular dynamics simulations of the reported water-soluble hemicarcerand (Octacid4) that explain how it uniquely encapsulates its guests at 298 K and keeps them encapsulated at 298 K in NMR experiments. Herein we report a modular approach to hemicarcerand simulations that simplifies the model building and charge derivation in a manner reminiscent of the approach to protein simulations with truncated amino acids as building blocks. We also report that apo Octacid4 in water adopts two clusters of conformations, one of which has an equatorial portal open thus allowing guests to enter the cavity of Octacid4, in microsecond molecular dynamics simulations performed using the modular approach at 298 K. Under the same simulation conditions, the guest-bound Octacid4 adopts one cluster of conformations with all equatorial portals closed thus keeping the guests incarcerated. These results explain the unique constrictive binding of Octacid4 and suggest that the guest-induced host conformational change that impedes decomplexation is a previously unrecognized conformational characteristic that promotes strong molecular complexation.
Introduction
Hemicarcerands comprise two identical bowl-shaped resorcinarene fragments tethered with four linkers (Fig. 1). These molecules, containing four equatorial portals in the linker region and two axial portals in the resorcinarene region, are developed as hosts to encapsulate small-molecule guests in a typical way that the guests enter and exit the host cavity at high temperatures and remain in the cavity at low temperatures 1,2 . The binding characteristics and solubility of hemicarcerands are governed mostly by the linker structures. Octaacid 4 (Octacid4; tethered with a 4,6-dimethylisophthalic acid fragment as shown in Fig. 1) is reportedly a unique hemicarcerand that is water soluble and can form host•guest complexes (hemicarceplexes) with small molecules in water without the need to raise temperature 3 . Therefore, the Octacid4-like hemicarceplexes are useful model systems for the study of constrictive binding, which is a type of molecular complexation that is affected by the activation energy required for a guest to enter the host cavity through a size-restricting portal [4][5][6] , to obtain insights into the rules of molecular complexation.
To date, only a few molecular dynamics (MD) simulations of hemicarcerands or hemicarceplexes have been reported for the facilitation of constructive bindings studies [7][8][9][10] . This scarcity is partly because hemicarcerands have more than 200 atoms, making their three-dimensional models tedious to build. The scarcity is also due to the complexity of hemicarcerands, all of which have multiple conformations 8,11 . As shown in Fig. 2, variations of the torsions along the four linkers of Octacid4 can result in many distinct conformations, which complicates the derivation of the conformationdependent atomic charges of the hemicarcerand. These technical complexities, as detailed below, may explain why there have hitherto been no aqueous MD simulations of Octacid4 to understand how it uniquely encapsulates various small-molecule guests at 298 K and keeps them encapsulated at the same temperature such that the bound guests can be differentiated in their NMR spectra from those in the bulk phase 3 .
Herein we report a modular method that simplifies the model building and charge derivation for MD simulations of Octacid4 and its hemicarceplexes. We also report the characterization of the host -4 -cavity and motions of the encapsulated guests that explains the unique constrictive binding of Octacid4 and offer new mechanistic insight into molecular complexation.
Results
The modular approach. In addition to the general technical complexities in hemicarcerand simulations noted above, the charge derivation for water-soluble hemicarcerands such as Octacid4 is particularly complicated. Because of the need to balance the atomic charges of the water-soluble hemicarcerand with the charges of the aqueous solvent and the charges of the small-molecule guest when using an AMBER forcefield such as FF12MClm 12 , the hemicarcerand charges need to be derived from ab initio calculations using the HF/6-31G* basis set that uniformly overestimates the polarity of the molecule. This is because (1) aqueous solvent models (such as the widely used TIP3P empirical water model) include polarization effects due to the empirical calibration to reproduce the density and enthalpy of liquid vaporization 13 , and (2) the small-molecule guest bears the restrained electrostatic potential (RESP) charges that are derived from ab initio calculations using the HF/6-31G* basis set [14][15][16] . Further, to obtain the atomic charges of a water-soluble hemicarcerand without any bias toward one particular conformation of the molecule, the hemicarcerand charges need to be obtained from ab initio calculations at the HF/6-31G*//HF/6-31G* level with multiple conformations of the molecule followed by using the Lagrange multiplier to force identical charges on equivalent atoms in these conformations 16 . These ab initio calculations and the Lagrange multiplier are computation-demanding and labor-intensive, respectively. For example, the ab initio calculation of one Octacid4 conformation at the HF/6-31G*//HF/6-31G* level took ~168 CPU hours using the computers at the University of Illinois Urbana-Champaign National Center for Supercomputing Applications.
During the course of our Octacid4 simulations, we recognized that the technical complexities described above differ little from those of MD simulations of proteins with large numbers of atoms -5 -and conformations. One known approach to circumventing the complexities of protein MD simulations is to perform the simulations using the modular concept underlying the secondgeneration AMBER forcefield 17 . This forcefield builds a protein model with truncated amino acids as building blocks and derives the atomic charges of each building block from its multiple conformations using the Lagrange multiplier to force identical charges on equivalent atoms in these conformations 16 . Given the effectiveness of the modular concept as demonstrated by the reported autonomous miniprotein folding MD simulations that achieved agreements between simulated and experimental folding times within factors of 0.6-1.4 18 , we used to the modular concept to simplify the model building and charge derivation for hemicarcerands and devised a modular method for hemicarcerand simulations. We describe below two key attributes of our method for performing the MD simulations of Octacid4.
The building block of Octacid4. Because Octacid4, similar to most hemicarcerands, has C4 symmetry, we divided it into four identical building blocks (termed HC1; Fig. 3). In this building block, C1 and C20 are designated as the head and tail atoms of the residue, respectively, similar to the designation of head and tail atoms in a truncated amino acid residue for protein simulations.
First, we assembled the four residues into a linear molecule. This assembly followed the same scheme as that for assembling protein residues-namely, constructing a covalent bond between the tail atom of the preceding residue and the head atom of the following residue. Second, we converted the linear molecule to a cyclic molecule by specifying a cross-link (viz., constructing a covalent bond) between the head atom of the first HC1 and the tail atom of the last HC1. Third, we converted the cyclic molecule to Octacid4 by specifying (i) four cross-links between C4 of the preceding HC1 and C18a of the following HC1, (ii) four cross-links between O5 of the preceding HC1 and C17a of the following HC1, and (iii) four cross-links between O16a of the preceding HC1 and C6b of the following HC1.
The Octacid4 sequence and all cross-links can be specified all together, thus making the hemicarcerand model building conceptually simple, algorithmic, and generalizable. Notably the HC1 building block can, in the same manner, be further divided into three sub-building blocks for hemicarcerands without C4 symmetry.
Atomic charges of HC1. Given the HC1 residue defined above, we obtained the atomic charges of Octacid4 via the RESP charge derivation for HC1 with the following specific conditions, in addition to the general conditions that are the same as those for protein RESP charges (such as the two-stage fitting for the methyl and methylene groups and forcing identical charges on equivalent intramolecular atoms) 16,17 .
First, we converted HC1 to HCD1 by attaching blocking groups to the junction atoms of HC1 ( Fig. 3) as these groups are needed to mimic the polar and aromatic groups abutting the junction atoms in Octacid4. We then performed the ab initio calculation of HCD1 to derive the HCD1 charges by using the Lagrange multiplier to force (1) HCD1 to have a net charge of -2 and (2) the total charge of all blocking groups to be zero, similar to the protein charge derivation from ab initio calculations of the acetyl-and N-methyl-blocked amino acids whose blocking group are used to mimic adjacent residues of the central amino acid 16 . Second, to balance the hemicarcerand charges with those of the solvent and the guest, we obtained the HCD1 RESP charges from ab initio calculations of HCD1 at the HF/6-31G*//HF/6-31G* level. Third, to avoid any bias toward one particular conformation of Octacid4, we obtained the HCD1 RESP charges from ab initio calculations using (1) two HCD1 proteins that uses the alpha helical and beta strand side-chain conformations 16 . Last, we extracted the HC1 charges from the HCD1 charges.
In contrast to the computation-demanding and labor-intensive charge derivation for the intact Octacid4 molecule, the HC1 charge derivation can be readily performed with HCD1. For example, the ab initio calculation of each of the two HCD1 conformations at the HF/6-31G*//HF/6-31G* level took 31-40 CPU hours, considerably less than the ~168 CPU hours for Octacid4 noted above.
Characterization of the Octacid4 cavity. Using the modular method described above, we performed 16 sets of 20 316-ns distinct, independent, unrestricted, unbiased, and classical isobaric-isothermal MD simulations of Octacid4 and its hemicarceplexes with seven small-molecule guests at 298 K and (Table 1) from the 14 reported guests that formed hemicarceplexes with Octacid4 3 .
As apparent in Fig. 4, the Octacid4 cavity is confined by a set of atoms shown with the sphere model. Therefore, we used the variation in the radius of gyration (Rg) of these atoms to estimate the change in cavity volume. Overall, the standard errors for the average Rgs of the cavity for the 16 sets of MD simulations were either 0.01 Å or 0.02 Å (Table 1) Table 1).
The average and standard error of the cavity Rg for the 20 simulations at 298 K were 5.44 Å and 0.02 Å, respectively (Table 1). For the simulations at 340 K, the Rg also remained at ~5.4 Å and spiked to >5.7 Å, but the frequency of the spikes was much higher at 340 K than that at 298 K ( Fig. 5B and Fig. S1), and the average and standard error of the Rg at 340 K were 5.45 Å and 0.02 Å, respectively (Table 1). For the simulations of the DEA, p-xylene, and naphthalene hemicarceplexes, we found a cavity Rg of ~5.6 Å with a few or no spikes of >5.7 Å at 298 K and 340 K ( Fig. 5 and Fig. S1); the respective average and standard error of the cavity Rg of the simulations at 298 K were 5.64 Å and 0.02 Å for DEA, 5.56 Å and 0.02 Å for p-xylene, and 5.65 Å and 0.02 Å for naphthalene. The corresponding -9 -values at 340 K were 5.65 Å and 0.02 Å for DEA, 5.59 Å and 0.02 Å for p-xylene, and 5.66 Å and 0.02 Å for naphthalene (Table 1). These time series of Rg and their average values indicate that (1) the hemicarceplexes involving relatively bulky guests in water all adopted one cluster of conformations with their cavities larger than that of the closed conformation of apo Octacid4 in vacuo and smaller than that of the open conformation of apo Octacid4 in vacuo, and (2) these hemicarceplexes mostly kept their cavities contracted with all equatorial portals closed.
Characterization of guest motion in Octacid4.
We also performed conformational cluster analyses on the simulations at 340 K for apo Octacid4 and the seven hemicarceplexes. All hemicarceplex simulations were performed using an initial conformation in which the host adopted the open conformation of apo Octacid4 in vacuo ( Fig. 2A) and the guest was manually docked into the host cavity with its maximal dimension perpendicular to the axial axis (viz., the axis passing the two axial portals). Interestingly, as detailed below, in the aqueous MD simulations, all seven guests adopted a new orientation with their maximal dimensions parallel to the axial axis (Fig. 6). Similar results were obtained (Fig. S2) when the analyses were performed using the simulations at 298 K. For the DMSO hemicarceplex (Fig. 6A), the most populated conformation had the guest oxygen atom pointing to the equatorial portals and the two guest methyl groups pointing to the axial portals; in the average conformation of the largest conformation cluster, the guest was shrunk to a ball with the oxygen atom on one side and the two overlapping carbon atoms on the other, indicating that the guest had rotated around multiple axes.
-10 -For the EtOAc hemicarceplex (Fig. 6B), the most populated guest conformation (population: 94.7%) was not fully extended (with C1-C2-O1-C3 and C2-O1-C3-C4 being -114° and 81°, respectively), and the population of the fully extended guest conformation with corresponding torsions of -180° and 180° was 0.1%; the most populated guest conformation had its two oxygen atoms pointing to the equatorial portals and its two methyl groups pointing to the axial portals; in the average conformation of the largest conformation cluster, the guest was shrunk to a short rod with the oxygen atoms in the middle and the two carbon atoms on both ends of the rod, indicating that the guest had rotated frequently around the axial axis and less frequently around the equatorial axis (the axis passing the two opposing equatorial portals). portals. In the average conformation of each cluster, the guest was shrunk to a long rod whose length was the same as that of the guest, indicating that these guests all had rotated exclusively around the axial axis. reportedly kept its guests encapsulated at the same temperature 3 . Perhaps the guests were incarcerated by strong intrinsic binding (viz., the complexation governed by strong nonbonded intermolecular interactions). However, the involvement of strong intrinsic binding is debatable given the weak interaction of the cavity with DMSO or 1,4-dioxane as indicated by their free spins as described above (Table 1 and Figs. 6A and 6D). An alternative explanation is therefore needed.
Discussion
Unexpectedly, our characterization studies show that upon complexation, Octacid4 adopted only Octacid4 in water adopts one cluster of conformations with closed equatorial portals. These conformations satisfactorily explain how Octacid4 can incarcerate a range of guests at 298 K, which is the temperature at which the guests enter the cavity.
-13 -Implication. Collectively, our studies suggest that Octacid4 can open one of its equatorial portals without elevating temperature before complexation and close all of its equatorial portals without lowering temperature after complexation. These interesting and unexpected capabilities enable the unique constrictive binding of Octacid4 with a range of small-molecule guests and suggest further that the guest-induced host conformational change that impedes decomplexation is a previously unrecognized conformational characteristic that is conducive to strong molecular complexation. This characteristic could broaden the theoretical dissection of the experimentally observed complexation affinity and the design of new complex systems for materials technology, data storage and processing, catalysis, drug design and delivery, and medicine by accounting for not only intrinsic binding (which is limited because, confined by the cost of chemical synthesis, only a small number of functional groups can be introduced to improve nonbonded intermolecular interactions) but also constrictive binding (which requires one or a few functional groups to trigger the formation of a host conformation that hinders decomplexation as demonstrated by the Octacid4 hemicarceplexes).
Molecular dynamics simulations.
Hemicarcerand Octacid4 or its hemicarceplex neutralized with sodium ions was solvated with the TIP3P water 13 ("solvatebox molecule TIP3BOX 8.2") using tLEaP of the AmberTools 16 package (University of California, San Francisco) and then energy-minimized for 100 cycles of steepest-descent minimization followed by 900 cycles of conjugate-gradient minimization to remove close van der Waals contacts using SANDER of the AMBER 11 package (University of California, San Francisco), forcefield FF12MClm 12 , and a cutoff of 8.0 Å for nonbonded interactions. The resulting system was heated from 5 K to 298 or 340 K at a rate of 10 K/ps under constant temperature and constant volume, and then equilibrated for 10 6 timesteps under constant temperature of 298 K or 340 K and constant pressure of 1 atm employing isotropic molecule-based scaling. Finally, a set of 20 distinct, independent, unrestricted, unbiased, isobaric-isothermal, and 316--14 -ns MD simulations of the equilibrated system was performed for Octacid4 or its hemicarceplex using PMEMD of the AMBER 16 package (University of California, San Francisco), forcefield FF12MClm 12 , and a periodic boundary condition at 298 K or 340 K and 1 atm. The 20 unique seed numbers for initial velocities of the 20 simulations were taken from Ref. 22 Root mean square deviation and radius of gyration. Carbon root mean square deviations were calculated manually using ProFit V2.6 ( [URL]/). Radius of gyration was calculated using CPPTRAJ.
Data availability
All data generated for this study are included in this Article and its Supplementary Information. K. G. M. performed the literature search; prepared the z-matrices of the HC1 and HCD1 residues; performed energy minimization of HC1, HCD1, and Octacid4; analyzed the MD simulation result of p-xylene•Octacid4; contributed to revisions of the manuscript. Y.-P. P. conceived the modular method; designed the HC1 and HCD1 residues and all protocols for MD simulation and analysis; performed the remaining computational study; analyzed the data; wrote the manuscript.
Competing interests
The authors declare no competing interests. MoGIO: The motion of the guest incarcerated in Octacid4. Axial spin: rotation around the axial\===
Domain: Physics Biology Chemistry. The above document has 2 sentences that start with 'We also report', 2 sentences that start with 'For example, the ab initio', 2 sentences that start with 'For the simulations', 2 sentences that end with 'at 298 K', 2 sentences that end with 'temperature 3 ', 2 sentences that end with 'in these conformations 16 ', 2 sentences that end with '0.02 Å, respectively (Table 1)', 2 sentences that end with 'K ( Fig'. It has approximately 2971 words, 113 sentences, and 30 paragraph(s).
|
[{"added": "2020-11-26T09:06:43.996Z", "created": "2020-11-20T00:00:00.000", "id": "227172673", "metadata": {"extfieldsofstudy": ["Biology", "Physics"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2020/11/20/2020.11.20.390328.full.pdf", "pdf_hash": "497f56dce58d8deaa75ac082be60d1ef03717826", "pdf_src": "BioRxiv", "provenance": "peS2o-0075.json.gz:38219", "s2fieldsofstudy": ["Chemistry"], "sha1": "36eee9b9b0b124f482445822e349341f39bf848a", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
A neuronal basis for fear discrimination in the lateral amygdala
In the presence of new stimuli, it is crucial for survival to react with defensive responses in the presence of stimuli that resemble threats but also to not react with defensive behavior in response to new harmless stimuli. Here, we show that in the presence of new uncertain stimuli with sensory features that produce an ambiguous interpretation, discriminative processes engage a subset of excitatory and inhibitory neurons within the lateral amygdala (LA) that are partially different from those engaged by fear processes. Inducing the pharmacogenetic deletion of this neuronal ensemble caused fear generalization but left anxiety-like response, fear memory and extinction processes intact. These data reveal that two opposite neuronal processes account for fear discrimination and generalization within the LA and suggest a potential pathophysiological mechanism for the impaired discrimination that characterizes fear-related disorders. When perceiving new stimuli, organisms need to distinguish between threats versus harmless stimuli. Here, the authors find a set of cells in the lateral amygdala that is required to discriminate or generalize new auditory stimuli based on similarity to previously fear-associate sounds.
E ncountering a new stimulus or situation represents one of the major challenges for organisms; i.e., which sensory stimuli should be approached and which should be avoided? For survival, animals, including humans, cannot spend time evaluating all possible consequences of a new situation. They must be able to predict them while the situation is on-going. To do this, organisms compare information about the context in which the new experience is occurring with stored knowledge about past emotional experiences. When new stimuli are perceptually similar to those associated with danger in the past, organisms respond with defensive responses, such as avoidance or immobility. Conversely, greater dissimilarity yields different behaviors (e.g., curiosity and exploration) [1][2][3] .
As a site essential for detecting threats, the amygdala is also important for the generalization of fear responses to new stimuli that resemble threats [4][5][6][7][8][9] , and the inappropriate activation of amygdala neurons may cause inappropriate fear reactions to harmless stimuli 4,[10][11][12][13][14] . In this framework, little is known regarding whether and how the amygdala participates in the evaluation of new stimuli that are different from threats. It has been proposed that increasing dissimilarity between new and threat stimuli decreases overall activity in the amygdala, thereby preventing inappropriate fear reactions 6,7,9 . However, recent studies have shown that the amygdala may also take part in learning safety [15][16][17] and extinguishing previously learned threat stimuli 4,18,19 . Therefore, the present study is aimed at investigating the neuronal mechanisms that are activated within the amygdala during the evaluation of new stimuli that may be perceptually similar to or different from previously learned threats. Here we show that a specific subset of neurons within lateral amygdala (LA) are activated when a new stimulus is evaluated as "harmless". These cells differ in part from those engaged by fearful stimuli, and the pharmacogenentic blockade of these neurons caused fear generalization.
Results
Different LA activation by uncertain stimuli presentation. Rats were trained to associate a pure tone of a specific frequency (conditioned stimulus, CS, 1 kHz) with a painful unconditioned stimulus (a mild electric foot shock, US). We chose this conditioning procedure to a single type of auditory stimulus because it mimics real-life threatening experiences that occur without fine and prolonged discrimination 8,20 . This allows the investigation of the neural processes that occur when subjects are facing totally new stimuli that may or may not resemble the threatening event. One week after training, we monitored the rats' behavior when presented with the CS (Group 1) or with new tones of increasing different frequencies (3 kHz, Group 2; 7 kHz, Group 3; or 15 kHz, Group 4). Twenty min later, all groups were presented with the CS to test their fear memory retention (Fig. 1a). As control group we used naive rats. Half of naive animals were exposed to the 1 and 7 kHz tones only during the test trial, while the others were exposed to 1 kHz tone unaccompanied by any US during training and, 1 week later, to either the 1 and the 7 kHz tones during test. Since we did not detect any differences between the two groups ( Supplementary Fig. 1), they were collected altogether ("naive").
Rats exhibited marked defensive behavior (i.e., a freezing response) to the CS (1 kHz, Group 1) and to a new tone with a closer frequency (3 kHz, Group 2). Conversely, Group 3 displayed an intermediate level of freezing to the 7 kHz tone, whereas Group 4 showed less freezing to the 15 kHz tone (Fig. 1b). The behavior of the rats within each group was similar except for Group 3, in which the perceptual features of the tone were interpreted in two opposite directions: half of the rats (5/9) showed a low level of freezing ("discriminators", D), while the others (4/9) exhibited higher freezing response ("generalizers", G) (Fig. 1c). The threshold (~43%) for assigning each animal to the "discriminator" or "generalizer" group was estimated through an expectation-maximization (EM) algorithm which yielded the maximum likelihood estimates for fitting a Gaussian mixture model (GMM) (see Methods and Supplementary Fig. 2). All conditioned groups displayed a higher and comparable level of freezing response to the CS with respect to naive animals ( Supplementary Fig. 3). In all animals, to analyze neuronal activation by new tones and the CS in the LA, we performed a cellular compartment analysis of temporal activity using fluorescent in situ hybridization (catFISH), a technique that allows the detection of neurons activated by two different events [21][22][23] . We analyzed the RNA expression of two activity-dependent genes, Arc/Arg 3.1 (Arc) and Homer 1a (H1a) in the same animal (Fig. 1d, e). Arc mRNA is detected in the nucleus 5-8 min after a salient event, while H1a mRNA is visualized in the nucleus 25-30 min afterwards 22,23 (Supplementary Fig. 4). In animals exposed to two behavioral epochs separated by 20 min, epoch 1 drives nuclear H1a, while epoch 2 induces nuclear Arc expression (Fig. 1d). Single Arc or H1a expression therefore reflects selectivity for one of the two events, while double-labeling demonstrates that the same neuron is engaged by both epochs [21][22][23] .
In Group 1, during the presentation of the two stimuli (CS-CS) there were many neurons activated during both events (Fig. 1e, f). Similar results were observed in Group 2 (3 kHz-CS) and in generalizer animals in Group 3 (7 kHz-CS) (Fig. 1e, f). Hence, fear generalization to a new tone recruited the same neurons that were activated by the CS. Conversely, in the discriminator animals in Group 3 (7 kHz-CS), the presentation of the new tone and the CS induced a significant reduction in the percentage of neurons expressing both Arc and H1a nuclear mRNA (i.e., neurons that responded to both the CS and the new tone) and increased the percentage of cells that responded separately to the CS and, strikingly, to the new tone (Fig. 1e, f). These data reveal that the LA is activated during the presentation of both harmful or new stimuli evaluated as not dangerous. In particular, in the presence of new stimuli evaluated as "harmless", LA activity relies on neurons that are partially different from those activated by the CS and that are silent during threatening experiences. The two different subpopulations of neurons that were activated by threatening vs harmless stimuli were intermingled within LA (Fig. 1e).
We then examined Group 4 (15 kHz-CS) to determine whether a similar phenomenon also occurs in the case of tones markedly different from the CS. However, in this group, the presentation of the new tone recruited a very low percentage of LA neurons (Fig. 1e, f). Notably, the discriminator animals in Groups 3 and 4 displayed similar behaviors during either CS or harmless stimuli, but only discriminator animals in Group 3 showed enhanced neuronal activity to either CS or harmless stimuli. These data allowed us to exclude that the neuronal activity detected only in Group 3 was the mere consequence of the motor or emotional behaviors. Altogether, these data demonstrate that, when stimulus features are markedly different from the CS, LA activity is lower than in the other conditioned groups, as previously proposed 6,7 . Conversely, when stimulus features may lead to opposite interpretations (i.e., safe or dangerous), in the LA there are neurons that are activated if the stimulus is evaluated as harmless. In the latter case, our data also suggest that the global activity of the LA is high and comparable to that displayed during threat events (Fig. 1f). This hypothesis was also corroborated by the fact that Discriminator animals in the Group 3 had higher number of activated neurons during new tone presentations than naive animals (Fig. 1f) Naive n = 9, 15 kHz, n = 5; 7 kHz, n = 9; 3 kHz, n = 5; and 1 kHz, n = 5. CS conditioned stimulus. b The percentage of freezing following the new tone presentation progressively increased in similarity to the CS (F (4, 28) = 36.32, P < 0.001). The freezing of naive and 15 kHz animals was similar (P > 0.05) and, in the meantime, it was lower than freezing of other groups (P < 0.01). Freezing of 7 kHz animals was different also from 3 kHz and 1 kHz (CS) group (P < 0.001). c Freezing in "discriminator" animals (D, n = 5) was lower than that observed in "generalizer" (G, n = 4) animals during the 7 kHz tone delivery (P < 0.001). d Time course of catFISH experiments. e Representative images showing neurons expressing single nuclear H1a (green arrows) and Arc (red) mRNA expression and double-labeled cells (yellow) in the naive, 15 kHz, 7 kHz (discriminators and generalizers), 3 kHz, and 1 kHz groups. Scale bar, 20 µm. f Dot plots showing the percentage of cells activated following new tone presentation (expressing only H1a), CS presentation (only Arc), and during both events (expressing both Arc and H1a). These results revealed an increase in H1a (new tone)-or Arc (CS)-expressing neurons and a decrease in doublelabeled cells in the "discriminator" group (F (5, 27) = 13.68 (left), P < 0.001; F (5, 27) = 12.68, P < 0.001 (middle); F (5, 27) = 30.28, P < 0.001 (right)). Raw data were expressed as a number of neurons labeled for Arc, H1a or both mRNA divided for the all counted neuronal nuclei analyzed. For each animal, we then calculated the mean of these raw data. g The total rate of H1a was lower in both the naive and the 15 kHz groups than in the other groups (F (5,27) = 24.67, P < 0.001). h The total rate of Arc was lower in the naive rats (F (5, 27) = 28.39, P < 0.001). i Venn diagrams showing the percentage of H1a-(green), Arc-(red), and double-(yellow) labeled neurons in LA in the different experimental conditions. Diagrams' size was scaled on the basis of H1a or Arc total ratios, and percentages were calculated by dividing the number of H1a-, Arc-, and double-labeled neurons for the total number of cells activated in at least one of the two events. *P < 0.05, **P < 0.01, ***P < 0.001. All data are mean and SEM. One-way ANOVA with Newman-Keuls test (b, f, i, j) To better investigate the differential activation of LA neurons, we calculated the reactivation ratio by dividing the amount of doubly labeled cells for the number of H1a-positive cells. Then, we compared this index with the reactivation ratio predicted by chance. The observed ratios were significantly higher with respect to predicted chance levels in each behavioral group (Supplementary Table 1) 24 . To depict the overall activity occurring in all groups, we calculated the percentage of neurons that were active during the first (Fig. 1g) or the second (Fig. 1h) event. During the first event, Group 7 kHz G and 7 kHz D did not differ each other and neither from Group 1 and 2, whereas all of these groups differed from Group 4 and naive animals (Fig. 1g, i). During the second event, all groups differed from naive animals (Fig. 1h, i).
To address whether our results represent a general feature of LA participation to fear discrimination irrespective of the tone frequencies employed, we performed similar experiments but by counterbalancing the tone paired to the US. Rats were conditioned to associate a 15 kHz tone to the US. One week after training, rats were presented with the CS (15 kHz) (Group 1) or with new tones of increasing different frequencies (7 kHz, Group 2; 1 kHz, Group 3). Twenty min later, all groups were presented with the CS to test their fear memory retention ( Supplementary Fig. 5a). catFISH analysis revealed results similar to those detected in the above experiment ( Supplementary Fig. 5). Finally, we found similar neuronal processes in two other structures that are involved in fear discrimination, namely the basal amygdala (BA) (Fig. 2a-f) and the prelimbic area (PrL) (Fig. 2g-l) of the medial prefrontal cortex. These data suggest that the recruitment of different active neuronal subpopulations may be a common neuronal mechanism that occurs in several brain sites during discrimination processes.
Characterization of the neurons activated by harmless cues. We next addressed the question of which specific cell types are involved in discriminator animals using triple fluorescent in situ hybridization. We combined catFISH with the detection of a third riboprobe for a specific neuronal marker. By analyzing colocalization with CaMKIIa, a marker of excitatory neurons, we found that within the LA, the majority of the cells that responded to the CS were excitatory neurons (70.89 ± 2.45%). Strikingly, a large percentage of neurons (50.54 ± 5.64%) activated by the new tone also overlapped with the CaMKIIa-positive population (Fig. 3a, b and Supplementary Fig. 6). Thus, within LA, there are distinct subpopulations of pyramidal neurons that are engaged by threat CS or by new harmless stimuli. We then analyzed whether the excitatory neurons activated during discrimination processes belonged to a specific excitatory subpopulation. Previous studies showed that in the BA (but not in the LA), excitatory pyramidal neurons are engaged during the extinction processes to mediate fear inhibition (i.e., "fear off" neurons) 18,25 . Some of these neurons are identified by the expression of Thy1, which distinguishes a specific subpopulation of pyramidal neurons 25,26 . We therefore sought to determine whether the excitatory neurons activated during the discrimination process might express Thy1. Triple-FISH analysis revealed that during the presentation of the new tone, CS or both, only a minimal fraction of excitatory neurons expressed Thy-1 (Fig. 3c, d and Supplementary Fig. 6). These results are consistent with those of previous studies 25 demonstrating that Thy-1 is expressed within the BA but not in the LA.
A previous study showed that a gene, gastrin-related peptide (Grp), is expressed within the pyramidal neurons of the LA 27 . These excitatory neurons release GRP peptide, which activates inhibitory neurons to increase the GABAergic inhibition of principal neurons 27 . Remarkably, GRPR-deficient mice showed enhanced fear memory 27 . We therefore sought to determine whether some of the pyramidal neurons activated during discriminative processes express Grp. Triple-FISH revealed that during the presentation of the new tone, CS or both, a large majority of excitatory neurons also express Grp (Fig. 3e, f and Supplementary Fig. 6).
We next investigated which inhibitory interneurons could be recruited by these processes by analyzing the colocalization of Homer 1a and Arc with either parvalbumin (PV), somatostastin (SOM), or calretinin (CR) mRNA ( Fig. 3g-l and Supplementary Fig. 6). Some PV, SOM, and CR interneurons were activated only during CS presentation, whereas a different portion of each neuronal subtype was recruited during only discriminative processes. Moreover, a portion of Arc + H1a expressing neurons also belonged to these different interneurons ( Fig. 3g-l). These data reveal that threatening and harmless stimuli activated different subtypes of PV, SOM, and CR interneurons in the LA. Because a large variety of interneurons subtypes have been described within the amygdala 28,29 , the observed differences may reflect this heterogeneity, and future studies will identify the specific cellular markers that characterize distinct inhibitory subtypes. These findings also showed that Arc was expressed in inhibitory neurons of LA, as previously reported in hippocampal, somatosensory, and striatal inhibitory neurons 30 . To further address this issue, we analyzed the colocalization between Arc and Gad1 (a neuronal marker for inhibitory neurons) mRNA in our samples. catFISH analysis revealed that the expression of Gad1 occurred in 23.35 ± 2.28% of LA neurons expressing Arc ( Supplementary Fig. 7).
Taken together, our data suggest that discrimination processes within the LA engaged an intricate network of both pyramidal cells and inhibitory interneurons that are partially different from those activated by the CS (Fig. 3m).
Blockade of LA neurons induced fear generalization. Do neurons activated during discrimination processes play a causal role in detecting harmless stimuli? To answer this question, selectively manipulated neurons activated during a specific experience without affecting neighboring cells. We accomplished this using a pharmacogenetic approach in which c-fos-lacZ transgenic rats carried a transgene where c-fos promoter drives lacZ gene transcription, leading to β-galactosidase (β-gal) expression 21,31,32 . Induction occurs only in strongly activated neurons in which βgal and Fos are coexpressed and not in neighboring non-activated or weakly activated cells 21,31,32 . These neurons can be inactivated 90 min after rats have performed a behavioral task by administering the prodrug Daun02: β-gal converts Daun02 into Representative images showing single-labeled H1a (green arrows)-and Arc (red arrows)-expressing cells, and double-labeled cells (yellow arrows) in the naive (n = 5), 7 kHz D (discriminators, n = 5), 3 kHz (n = 5) and 1 kHz (n = 5) groups. Scale bar, 20 µm. c In BA, following new tone and CS presentation, the percentage of H1a-or Arc-expressing cells was higher in the 7 kHz D group than in other groups (F (3,16) = 19.57, P < 0.001 (left); F (3, 16) = 16.41, P < 0.001 (middle); F (3, 16) = 23.67, P < 0.001 (right)). d Total rates of H1a (F (3,16) = 19.44, P < 0.001) or e Arc (F (3,16) = 20.78, P < 0.001) were lower in the naive than in the other groups. f Scaled Venn diagrams showing the percentage of H1a (green), Arc (red), and H1a + Arc (yellow) expressing neurons in BA in the different behavioral groups. The neuronal populations activated during both new tone or CS presentation in 7 kHz D group were less overlapping with respect to other groups. g catFISH was performed in the layers II-III of the prelimbic cortex (PrL). The section diagram was drawn on the basis of our DAPIstained sections. h Images showing H1a and Arc nuclear expression in the naive, 7 kHz D and 1 kHz groups. i In PrL, the percentages of cells single-labeled for H1a or Arc were significantly higher in the 7 kHz D group than in the naive and 1 kHz groups (F (2,9) = 44.82 (left), P < 0.001; F (2, 9) = 9.91, P < 0. 01 (middle)). Conversely, the percentage of double-labeled cells was lower in both the 7 kHz and the naive group than in the 1 kHz group (F (2,9) = 13.55, P < 0. 01 (right)). j The total rates of both H1a (F (2,9) = 9.76, P < 0. 01) and k Arc (F (2,9) = 10.93, P < 0. 01) were lower in the naive group than in the 7 kHz D and 1 kHz groups. l In PrL, scaled Venn diagrams indicated that in 7 kHz D group the neuronal populations activated during new tone or CS presentation were less overlapped with respect to other groups. *P < 0.05, **P < 0.01, ***P < 0.001. All data are mean and SEM. daunorubicin, which induces apoptotic cell death in 3 days after injection. First, we verified whether Arc mRNA activation colocalized with cFos expressing cells in our experimental protocol. catFISH analysis showed that 87.23% ± 1.25 of Arc-labeled neurons expressed also cFos mRNA ( Supplementary Fig. 8), in line with previous studies showing that c-fos-expressing cells exhibit also the activation of other activity-dependent genes such as Arc, H1a, and zif268 33,34 .
Transgenic rats were exposed to the association between a tone (1 kHz) and US, as described in the above experiments. One week later, they were presented with a new tone of 7 kHz, and 90 min later, the discriminator animals were bilaterally infused with Vehicle or Daun02 in the LA (Fig. 4a). As a further control group, Arc we added a "Daun02-delayed" group, in which Daun02 was injected one week after the new tone presentation, i.e., when the level of β-galactosidase expression induced by new tone was returned to baseline (Fig. 4a). On test day, the Daun02-injected animals displayed a significant increase in freezing in response to the 7 kHz tone with respect to controls ("Vehicle" and "Daun02delayed" groups) ( Fig. 4e and Supplementary Fig. 9), suggesting that the pharmacogenetic deletion of LA neurons activated during discrimination processes caused a shift from discrimination to fear generalization. A post-mortem cytochemical analysis revealed that the presentation of the new stimulus induced βgal in 83.04 ± 2.25% of LA neurons expressing Fos (Fig. 4b) and that the Daun02-injected rats had markedly fewer Fos-expressing neurons (Fig. 4c, d).
It is unlikely that these results were due to nonspecific neurotoxicity in the LA or to interference with the overall functions of the LA because both conditions should have decreased (and not enhanced) defensive responses 35 . To further address this point, we tested fear memory expression by presenting the CS. No differences were detected among groups (Fig. 4e). In addition to confirming that our pharmacogenetic procedure did not affect the overall functions of the LA, the latter finding also indicated that neurons that are necessary for fear discrimination are conversely not necessary for CS memory expression. Intriguingly, catFISH analysis showed that some neurons were activated by both new stimuli and CS presentation. On the other hand, applying a pharmacogenetic approach after discrimination processes showed that these commonly activated cells are not necessary for fear memory expression. Thus, these neurons might be useful for comparing new stimuli with threatening ones, but they may be dispensable for fear memories. Future studies will help to define the precise role of these cells.
In the Vehicle-and Daun02-injected animals, we did not detect any difference in spontaneous behaviors displayed before CS presentation (Fig. 4e). To better define the effects of the pharmacogenetic deletion of specific LA neurons on spontaneous behavior, Vehicle-and Daun02-injected animals were submitted to the open field and elevated plus maze tests. In both paradigms, we did not detect any differences between groups (Fig. 4f, g). These data show that enhanced fear behavior was not the result of a general effect on anxiety-like or unconditioned fear responses.
We next investigated whether neurons activated during discrimination processes are specifically necessary for discriminative processes or whether they may serve to inhibit fear more generally and/or to switch from a high to a low fear state. Previous studies showed that neurons in the BA are necessary for this transition during fear extinction 18,25,26 . Therefore, in another group of animals, Daun02 or Vehicle was injected after the presentation of the new tone, and the animals subsequently underwent extinction training (Fig. 4h). No differences were found between groups during the extinction training. These data reveal that neurons activated during discrimination between fear and harmless cues do not serve to inhibit fear nor to switch between exploratory and defensive behaviors. These findings are in line with the results of the catFISH analysis showing that neurons activated by discrimination belong to an excitatory subpopulation that is different from the "fear off" subpopulation 18,25,26 . Altogether, these data support the view that, in the LA, there are cells that are specifically involved in discriminating between harmless and threatening stimuli and that these cells are not required for extinguishing threatening stimuli.
To additionally test the specificity of our approach, in another group of transgenic rats, we administered Daun02 after CS presentation (Fig. 4i) and we found that it did not affect discrimination processes during the presentation of a new tone, whereas it impaired subsequent fear memory expression (Fig. 4i). These results confirm that different neuronal processes are engaged in the LA during discrimination process and fear memory expression and that the pharmacogenetic disruption of these different cell types produces opposite behavioral outcomes.
Our previous catFISH analysis also showed that neurons in the LA were specifically engaged in response to a 7 kHz tone but not to a 15 kHz tone. Therefore, another group of transgenic rats received Vehicle or Daun02 following the presentation of a new 15 kHz tone. On test day, the two groups showed similar responses to the 15 kHz tone and the CS (Fig. 4j). These data demonstrate that the LA is necessary for discrimination only in cases of uncertain stimuli to which an unambiguous interpretation is impossible, while it is dispensable if new stimuli are totally different from the CS.
Discussion
Our findings shed new light on the neuronal processes that are engaged during the presentation of new stimuli. In keeping with previous studies 6,7 , we found that LA activity is higher when new stimuli resemble threatening ones and lower when new stimuli largely differ from threats. On the other hand, we provide evidence showing that when the perceptual features of a new stimulus are neither close to nor very different from a threatening stimulus, this ambiguity may be resolved by the activation of two opposite neuronal subpopulations within the LA. In some animals, the same neurons that were activated by threats were also activated in the presence of the new stimulus, and this led to generalized fear. Conversely, in other rats, the same new stimulus led to the activation of neurons that are partially different from those activated during fear-related experiences. The recruitment of this neuronal ensemble allows discriminative processes to occur and produces an opposite behavioral outcome. Deleting the latter neuronal subpopulation caused fear generalization but did not affect innate fear, fear memory expression or extinction processes. These findings support the idea that these cells are specifically activated to recognize and react appropriately to new stimuli that are evaluated as harmless in uncertain situations rather than serving to switch fear off, as previously reported for some BA neurons 18,25,26 . They may therefore take part in resolving ambiguous situations. Our data were obtained by testing rats one week after fear training. This allowed us to exclude that LA activity, when present, was due to any non-specific effects related to painful stimulation. Conversely, a previous study showed that Arc may be activated 15 min after painful stimulation irrespective of whether a fear memory was formed or not 36 , thus suggesting that shortly after training LA activity may be increased either by memory processes or by painful stimuli. A similar effect of painful stimulation on short-term neuronal activity was also reported in the cerebellar vermis 37 .
Since Pavlov's studies 1 , it has been known that behavioral responses towards stimuli that are increasingly dissimilar to one paired to an aversive event decay smoothly. The perceptual model of fear generalization proposes that neuronal activity arises in brain sites as a consequence of the perceptual similarity between a new stimuli and the CS 38 Fig. 4 LA neurons activated by a new tone presentation are necessary for fear discrimination processes. a Experimental design for the Daun02 inactivation technique, see Methods for details. Daun02 (n = 9),Vehicle (n = 9), "Daun02-delayed" group (n = 4). b New tone presentation induced Fos (red nuclei) and β-galactosidase (green nuclei) expression in LA neurons. The merged panel shows nuclei that were double-labeled for β-galactosidase and Fos (yellow). Scale bar, 100 µm. c Photomicrographs of Fos-stained LA neurons in the Vehicle-(left) and Daun02-injected (right) rats at 90 min after 7 kHz sound presentation. Scale bar, 50 µm. d There were more Fos-positive cells in the Vehicle-than in the Daun02-injected animals. e Freezing response of the three groups ("Vehicle", "Daun02-delayed", and "Daun02") during new tone (7 kHz) presentation was similar among groups. After Daun02 injection, freezing to the 7 kHz tone was higher than in the two control groups (3 × 2 mixed ANOVA, main effect of group: F (2, 19) = 7.25, P < 0.01, main effect of trial: F (1, 19) = 9.23, P < 0.01, group × trial interaction: F (2,19) = 20.07, P < 0.001, simple main effect of group in post-injection trial: F (2, 19) = 24.13, P < 0.001). Freezing to the CS was similar between groups (one-way ANOVA: P > 0.05). f, g In open field (OF) and elevated plus maze (EPM) tests, there were no differences between Vehicle and Daun02 groups. h No differences were detected during extinction sessions (n = 6 rats in each group). i In another experimental group, Daun02 (n = 5 rats), Daun02-delayed (n = 5), or Vehicle (n = 6 rats) was injected after CS presentation. Three days later, both a new tone and the CS were presented. Freezing responses to CS in the Daun02-injected animals were decreased (3 × 2 mixed ANOVA, main effect of group: F (2, 13) = 15.36, P < 0.001, main effect of trial: F (1, 13) = 0.80, P > 0.05, group × trial interaction: F (2, 13) = 14.11, P = 0.001, simple main effect of group in postinjection trial: F (2, 13) = 20.57, P < 0.001), but there were no differences between groups during new tone presentation (one-way ANOVA: P > 0.05). j Daun02 administration following the presentation of a new 15 kHz tone did not modify freezing to either this tone or to CS (n = 9 rats in each group). **P < 0.01, ***P < 0.001. All data are mean and SEM. Unpaired t test (d, j); one-way ANOVA (e, i); 3 × 2 mixed ANOVA (e, i); 2 × 2 mixed ANOVA (f, g, j); 2 × 7 mixed ANOVA (h). See Supplementary Note 1 for a more detailed description of statistical results of this figure are totally different. Conversely, when a stimulus has intermediate features, there is no association between the number of recruited LA neurons and freezing levels. In this condition, the activity tuning within the amygdala may deviate from both perceptual features and the overt behavioral fear tuning. These data also suggest that imaging of the amygdala as a whole may not provide an accurate index of discriminative processes, at least in the LA.
It is likely that neurons activated within the LA may be part of a more complex network that encompasses several other brain areas, such as the medial prefrontal cortex 2,40,41 and the auditory cortex 20,[42][43][44][45] , both of which are involved in fear memory and discrimination processes.
Our pharmacogenetic experiments also demonstrate that inappropriate fear can be caused by disrupting the neurons that are engaged when animals encounter new harmless stimuli. This finding suggests a potential pathophysiological mechanism for the impaired discrimination that characterizes fear-related disorders, such as phobias and post-traumatic stress disorders. Stress and traumatic events might affect the correct functionality of these neurons, leading to fear overgeneralization. If confirmed, this finding suggests that appropriate treatments for fear-related disorders should not be aimed at decreasing amygdala global activity but should instead be directed toward strengthening the specific activation of these neurons. Fear conditioning. Rats were trained to associate a conditioned auditory stimulus (CS) with a painful unconditioned stimulus (US) as in our previous studies 21,43,44 . The floor of the conditioning cage was made of stainless steel rods connected to a shock generator set to deliver 1 mA current. The chamber was fitted with a loudspeaker connected to a tone generator set to deliver an 80 dB, 1000 Hz pure tone (CS); the loudspeaker was located 20 cm above the floor. One animal at a time was placed inside the chamber and left undisturbed for 2 min. Then, it was exposed to a series of seven consecutive auditory CSs, each lasting 8 s and paired, during the last 1 s, with an electric foot shock (1 mA; 1 s); the seven sensory stimuli were separated by intervals of 22 s. "Familiar tone" group underwent the same experimental procedure but without any painful stimulation. Naïve animals were presented with the different tone only during the test trial.
New auditory stimuli and fear memory retention test. The presentation of novel auditory stimuli and the retention of fear memory was tested 1 week after the conditioning. Rats were handled for two consecutive days (5 min per day), habituated to an apparatus different from those used for conditioning and in a different room, in order to avoid conditioned fear behavior to contextual cues 43 . The cage consists in a transparent plastic cage enclosed within a sound-attenuating box equipped with an exhaust fan, which eliminated odorized air from the enclosure and provided background noise of 60 dB.
On the third day, we performed the behavioral test, divided in two different events separated each other by a time interval of 20 min in order to allow catFISH analysis. During the first event, after 1 min of free exploration, we presented a new auditory stimulus never presented before, repeated for four times (8 s, with a 22 s interval). In the separate behavioral groups, we delivered a tone of a different frequency (15 kHz, 70 dB; 7 kHz, 71 dB; 3 kHz, 73 dB). The session lasted 3 min after which the rat was returned to its home cage for 19 min. Then, animals were placed back in the same environment and exposed to CS (1 kHz, 72 dB) four times (8 s, with a 22 s interval). This sound was administered in a similar manner to that used during conditioning, but without the foot shock. A further behavioral group, was presented with 1 kHz sound during both events.
For the experiments in which we counterbalanced the frequencies of tones we repeated the same procedures except for the frequency of the CS (15 kHz) and of the new tones (7 kHz and 1 kHz).
The rats' behavior was recorded by a digital videocamera and the videos were reviewed to determine the duration of defensive responses. Freezing response was employed as an index of defensive behavior. Freezing was expressed as the percentage of time during which there was complete absence of somatic mobility, except for respiratory movements. The assessment of freezing was done by one person blinded to the animal's assignment to an experimental group.
In order to evaluate the rats belonging to the 7 kHz Group as discriminators "D" or generalizers "G", we employed an EM algorithm which yielded the maximum likelihood estimates for fitting a Gaussian mixture model (GMM) 46 .
By applying this iterative method to the vector of freezing responses to the New 7 kHz Tone, under the assumption that data points were generated from a twocomponent mixture of Gaussian distributions, we obtained an approximation of the probability density that most likely generated the data. The EM-GMM estimated a threshold of~43%. Therefore, animals with a freezing response lower than 43% to the New Tone were classified as discriminators, whereas animals showing a freezing higher than 43% to the New Tone were classified as generalizers (see Supplementary Fig. 2).
Behavioral paradigms for Fos-LacZ experiments. Fos-LacZ rats underwent fear conditioning as previously described (see Fear conditioning paragraph). One week after training, animals were presented with a new sound (7 kHz pure tone, 10 s). Each animal was placed inside the chamber and left undisturbed for 1 min. Then, it was exposed to the new tone, repeated for three times (10 s, with a 40 s of interval). Ninety minutes later, when β-galactosidase was near maximal levels 31,47 Daun02 or vehicle was bilaterally infused into the LA. Rats were returned to their home cages for 3 days in order to produce cell-specific inactivation 47 . On test day, both vehicle and Daun02-injected rats were returned in the cage and, after 1 min of exploration, 7 kHz tones were delivered, as previously described. Fear memory retention was tested 3 h later by delivering the CS (1 kHz pure tone) previously paired with the foot shock. "Daun02-delayed" rats underwent to the same experimental procedure but Daun02 injection was performed 1 week after new tone presentation.
In the second experiment, one week after fear conditioning, Daun02 was injected following the presentation of the CS (1 kHz pure tone, 8 s, interval of 22 s) previously paired with foot shock. The test was performed as in the first experiment.
The third experiment was performed as the first one (7 kHz experiment), but 15 kHz pips, lasting 1 s and delivered at 1 Hz for 15 s (inter-trial interval, 45 s), were used as new tone.
Fear extinction protocol. During the fear extinction procedure, Fos-LacZ rats were placed in the same environment as that in which they were presented with 7 kHz tone and were exposed to CS (1 kHz) 36 times (8 s, with a 32 s interval). This sound was administered in a manner identical to that used during conditioning, but without the foot shock. This paradigm was administered for two consecutive days, one session per day.
Open field paradigm. The open field apparatus consisted of a plastic opaque box (50 × 80 × 40 cm). Rats were placed in the center of the apparatus and their behavior was recorded for 10 min. The analyses were conducted using the Smart 3.0 software (Panlab, Cornella, Spain).
Elevated plus maze paradigm. The apparatus consisted of four arms (two open without walls and two enclosed by 30 cm high walls) 50 cm long and 10 cm wide. Each arm of the maze is attached to plastic legs, such that it is elevated 53 cm off a base that it is on. Rats were placed in the center of the apparatus and their behavior was recorded for 10 min. The analyses were conducted using the Smart 3.0 software (Panlab, Cornella, Spain). For the treatment of c-fos-lacZ transgenic rats, Daun02 (Hycultec, catalog no. HY-13061) was dissolved to a final concentration of 5 mg/ml in a solution of 10% DMSO, 6% Tween-80, and 84% phosphate-buffered saline. A volume of 1 µl of both Daun02 or vehicle was used per injection site (v = 0.3 µl/min), and the needle was left in place for additional 3 min. The incision was then closed with stainless steel wound clips, and the animal was given a subcutaneous injection of the analgesic/anti-inflammatory ketoprofen (2 mg/kg body weight); it was kept warm and under observation until recovery from anesthesia. Needle track placement was verified in Nissl stained sections. The sections were histologically verified under a microscope magnified at ×2 and ×4.
Tissue preparation and histological procedures. Immediately after testing, rats were anesthetized and then rapidly decapitated with a guillotine. Brains were quickly removed and frozen in isopentane that had been supercooled on dry ice (approximately −80°C). Frozen brains were stored at −80°C. For sectioning, brains were warmed to −20°C and fixed to the platform of a cryostat with Tissue-TeK O. C. T. Compound (VWR). Sections (20 μm thick) were mounted on slides (Superfrost Plus, VWR), which were then sealed in boxes and stored at −80°C until use.
Fluorescent in situ hybridization (catFISH). catFISH analysis was used to examine the expression of Arc/Arg 3.1 (Arc) and Homer 1a (H1a) genes. Briefly, an Arc antisense riboprobe was directed to the region from exon I to III, while an H1a probe was directed to the 3′ UTR. The vectors were linearized with EcoR1, purified and used for in vitro transcription with the DIG RNA Labeling kit (SP6/T7) (Roche, 11175025910), in the presence of fluorescein-UTP (incorporated into the H1a probe) or digoxigenin-UTP (incorporated into the Arc probe). The yield and integrity of riboprobes was confirmed by gel electrophoresis. At the end of this process, probes were purified by spin chromatography.
For characterization experiment, we add a third probe in order to detect the neuronal identity of Arc and H1a expressing neurons. To this aim, we analyzed the In order to analyze a possible colocalization between Arc and Gad1 or between Arc and cFos, the primers used to make riboprobes were: for Gad1: (FW) ACCAGATGTGTGCAGGCTAC, (RV) ACAGATCTTGACCCAACCTCTC; for cFos:(FW) TGTCAGGGAAGAGTAGGGGTC, (RV) CCAGACACAGGTGGAGCAAG. These two riboprobes were incorporated with digoxigenin-UTP labeling mix (Roche), incubated with anti-digoxigenin-POD (Roche) and, then, with a Cy3 substrate kit (1:50, PerkinElmer). For this experimentArc was incorporated in fluorescein-UTP labeling mix (Roche), incubated with anti-fluorescein-POD (Roche) and, then, with a Fluorescein substrate kit (1:50, PerkinElmer).
Slides were imaged using a Leica SP5 confocal microscope using four lasers (488, 520, 570, and 633 nm) corresponding to peaks in the emission spectra of DAPI (cell nuclei), fluorescein (H1a mRNA), Cy3 (Arc mRNA), and Cy5 (neuronal marker's mRNA), respectively. The objective lens was set at ×63 magnification. Data were acquired using a z-stack (1 μm thickness per section in a stack), the height of which was determined by the penetration of one detectable probe per sample (usually 8 μm thickness per stack). The pinhole, photomultiplier tube gain and contrast settings were constant for all image stacks acquired from a slide. Cells were considered for analysis if the nucleus was present in at least four sections of the z-stack. Only putative neurons were included in the analysis, and glial cells, identified from their small size (~5 μm diameter) and bright, uniform nuclear counterstaining, were excluded. Cells that were positive for both DAPI and Cy3 were considered Arc-positive, cells with both DAPI and fluorescein were considered H1a-positive, and cells with DAPI, Cy3, and fluorescein were positive for both mRNA. In triple catFISH analysis, we defined the number of cells expressing H1a, Arc, and H1a + Arc as described above. Then, we calculated the number of cells that were also Cy5-positive (neuronal marker's mRNA) in order to define the percentage of colocalization between H1a-, Arc-or double-labeled cells and each neuronal marker analyzed (CamKIIa, Grp, Thy-1, PV, Som, and CR). In double catFISH experiment performed BY using Arc and Gad1 or Arc and cFos riboprobes, Arc was visualized in the emission spectra of fluorescein, and Gad1 and cFos in the Cy3 emission spectra.
Cells counts were performed manually; to prevent bias, the experimenter was blinded to the relationship between the images and the behavioral conditions they represented. Raw data were expressed as a percentage of the total neuronal nuclei analyzed per stack. Typically, for PrL, 8 confocal z-stacks (175 × 175 µm square; zoom fraction, 1.4) were taken from cortical layer II-III of each animal: images were collected from two bilateral slides at 2.8 anteroposterior coordinate 48 . Thus, in this region we collected an average of 521 ± 30 DAPI-labeled cells per animal.
For LA and BA, four confocal z-stacks (189 × 189 µm square; zoom fraction, 1.3) in LA and eight confocal z-stacks in BA were taken from each animal: images were collected from two bilateral slides at anteroposterior coordinates ranging from −2.6 to −3.0 48 . In LA, we collected an average of 174 ± 7 DAPI-labeled cells per animal, and in BA an average of 312 ± 15 DAPI-labeled cells per animal.
In order to define the percentage of colocalization between Arc and Gad1 and Arc and cFos, the percentage of colocalization was obtained by counting the number of cells expressing Arc and Gad1 or Arc and cFos.
In catFISH analysis, the percentages of stained cells for different groups were presented as mean and SEM.
Beta-galactosidase and Fos immunohistochemistry. On the injection day (Vehicle/Daun02), 90 min after testing, a group of cfos-lacZ rats was deeply anaesthetized and perfused intracardially with 4% paraformaldehyde in order to examine the colocalization of Fos and β-galactosidase protein expression. The brains were dissected, stored overnight at 4°C, and transferred to 30% sucrose. Coronal sections (50 μm) were cut on a vibratome and collected in phosphatebuffered saline (PBS). Free-floating sections were incubated in a blocking solution (4 % bovine serum albumin (BSA), 10% normal goat serum, and 0.5% Triton X-100) for 1 h at room temperature. Then, they were incubated in rabbit antibody to c-Fos (1:500 dilution, Santa Cruz Biotechnology, sc-52) and sheep antibody to βgal (1:1000, Aves Labs, BGL-1010) in the blocking solution overnight at 4°C. Subsequently, sections were washed with PBS and incubated for 1 h at room temperature with AlexaFluor 488-labeled goat anti-sheep IgG (1:400 dilution, Life Technologies, A11039) and AlexaFluor-568-labeled goat anti-rabbit IgG (1:400 dilution, Life Rechnologies, A11036) diluted in PBS, for 1 h on a shaker at room temperature. Sections were washed in PBS, mounted with mounting media containing DAPI (Vector) and cover-slipped.
Fos immunohistochemistry. Three days after injection and 90 min after new tone (7 kHz) presentation, both vehicle and Daun02-injected rats were deeply anaesthetized and perfused intracardially with 4% PAF. Brain sections were processed for Fos immunohistochemistry. We used the primary antibody to c-Fos (1:500, Santa Cruz Biotechnology, sc-52) and the sections were developed with AlexaFluor-568-labeled goat anti-rabbit antibody (1:400 dilution, Lifetechnologies, A11036) as described in the previous section Immunohistochemistry analysis. Tissues were imaged using three lasers (488, 520, and 570 nm), each corresponding this time to the peak emission spectrum for DAPI (Nissl stain for cell nuclei), Fluorescein (β-gal) and CY3 (Fos), respectively. Images were acquired using a z-stack (1 μm thickness per section in stack), the height of which was 8 μm. Cells were counted for analysis if the nucleus was present on at least 4 sections of the z-stack. The objective lens was set at ×63 magnification. Cells which were positive for both DAPI and CY3 were considered Fos-positive, cells with both DAPI and fluorescein were consideredβ-gal-positive, and cells with DAPI, CY3, and Fluorescein were considered double-labeled for both proteins. The results were expressed as a percentage of the total neuronal nuclei analyzed per stack.
Typically, four confocal z-stacks (189 × 189 µm square area; zoom fraction = 1.3) were taken in LA from each animal: images were collected from four slides at antero-posterior distance of −2.8 mm from the Bregma 48 .
Statistical analyses. Since all data passed Kolmogorov-Smirnov's test and Brown-Forsythe test, parametric statistics were employed through all the experiments. F test was employed to test equality of variances where unpaired t test were used.
Concerning catFISH analysis, raw data were expressed as a number of neurons expressing Arc, H1a or both RNA divided for the total neuronal nuclei analyzed per stack. In order to determine the overall activity within LA during new tone or CS presentation we calculated the total rate of H1a, defined as the percentage of H1a-positive nuclei (expressing single H1a and both H1a + Arc) out of the total number of labeled nuclei. Arc total rate was calculated as percentage of Arc-positive nuclei (expressing single Arc and both H1a + Arc) out of labeled nuclei. The percentages of stained cells for different groups were presented as mean and SEM. In catFISH experiment, behavioral data and cell counts were analyzed by performing Student's two-tailed unpaired t test or one-way ANOVA followed by Newman-Keulspost hoc multiple comparison tests.
In order to test the main effect of group, the main effect of trial and the group × trial interaction effect in Fos-LacZ experiments, we employed a 3 × 2 and a 2 × 2 mixed-design ANOVA with group (Vehicle, Daun02-delayed, Daun02) as between-subjects variable and trial (pre-injection, post-injection) as within-subjects variable. Where the interaction was significant, we performed a simple main effects analysis (i.e., effect of group in pre-and post-injection trials separately).
In order to test the main effects of group and trial, and the interaction effect in the extinction paradigm, we employed three 2 × 7 mixed-design ANOVAs with group (Vehicle, Daun02) as between-subjects variable and trial (tones 1-7 in the Day-1 session, tones 30-36 in the Day-1 session, or tones 30-36 in the Day-2 session) as within-subjects variable. For each mixed ANOVA model we assessed the Sphericity assumption through Mauchly's Test of Sphericity. Where it was violated, we applied the Greenhouse-Geisser correction accordingly.
In order to compare the observed reactivation ratio and the reactivation ratio predicted by chance, we applied a calculation procedure 24 . We considered the amount of green [G: H1a + (H1a + Arc)] and red [R: Arc + (H1a + Arc)] cells, and the total amount of neuronal nuclei in the same regions [D: DAPI]. We used Y/G to calculate the observed reactivation ratio, as in our previous work 21 . The chance for each neuron to be yellow [Y: H1a + Arc] is R/D*G/D, and the predicted Y = (R/D*G/D)*D = R/D*G. Thus, the predicted chance level of reactivation ratio is Y/ G = (R/D*G)/G = R/D. The predicted ratio (R/D) and the observed ratio (Y/G) of each group were compared using a Student's two-tailed paired t test (Supplementary Table 1). In order to test the difference between two groups, we used a Student's two-tailed unpaired t test.
All statistical analyses were performed using Graphpad Prism 6 and SPSS Statistics 22 (IBM). The Gaussian mixture model was implemented with XLStat 19.4 (Addinsoft).
catFISH experiment was replicated two times while Fos-LacZ experiments were replicated three times.
Data availability. The data that support the findings of this study are available from the corresponding author upon reasonable request.
==
Domain: Psychology Biology
|
[{"added": "2023-02-08T14:38:03.999Z", "created": "2018-03-23T00:00:00.000", "id": "256643273", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/s41467-018-03682-2.pdf", "pdf_hash": "d7921fa627ddd294939f69ff08e5a4be5c67098a", "pdf_src": "SpringerNature", "provenance": "peS2o-0075.json.gz:30780", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "d7921fa627ddd294939f69ff08e5a4be5c67098a", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Allele-specific expression in a family quartet with autism reveals mono-to-biallelic switch and novel transcriptional processes of autism susceptibility genes
Autism spectrum disorder (ASD) is a highly prevalent neurodevelopmental disorder, and the exact causal mechanism is unknown. Dysregulated allele-specific expression (ASE) has been identified in persons with ASD; however, a comprehensive analysis of ASE has not been conducted in a family quartet with ASD. To fill this gap, we analyzed ASE using genomic DNA from parent and offspring and RNA from offspring’s postmortem prefrontal cortex (PFC); one of the two offspring had been diagnosed with ASD. DNA- and RNA-sequencing revealed distinct ASE patterns from the PFC of both offspring. However, only the PFC of the offspring with ASD exhibited a mono-to-biallelic switch for LRP2BP and ZNF407. We also identified a novel site of RNA-editing in KMT2C in addition to new monoallelically-expressed genes and miRNAs. Our results demonstrate the prevalence of ASE in human PFC and ASE abnormalities in the PFC of a person with ASD. Taken together, these findings may provide mechanistic insights into the pathogenesis of ASD.
Results
Quality and quantity of DNA and RNA of a parent-child quartet with ASD met the requirements for deep sequencing. We analyzed ASE on a genome-wide scale and investigated whether dysregulated ASE occurs in persons with ASD using samples from a family quartet with ASD consisting of genomic DNA from the parent and offspring and RNA extracted from postmortem PFC of the offspring (Fig. 1a). Sequencing of parental genomic DNA provided information regarding single nucleotide polymorphisms (SNPs), which is essential for determining the parental source of offspring's transcripts. In this family quartet, both offspring were female, and one had been diagnosed with ASD (Fig. 1a). Supplementary Fig. 1 provides a more detailed pedigree of this family, which shows epilepsy and deafness co-occurred in both the affected and unaffected offspring in addition to other familial health conditions. Next, we performed DNA sequencing of parents and offspring; RNA-sequencing and follow-up transcriptomic and ASE analysis was performed on the offspring. The detailed demographic and deep sequencing information is presented in Supplementary Table 1. The quantity and quality of genomic DNA from parents and offspring met the requirements for DNA-sequencing (Fig. S2a); in addition, the quantity and quality of the offspring's RNA met the requirements for RNA-sequencing (Fig. S2b). Total read number and mappability of the offspring's RNA-Seq data met the requirements for further statistical analysis (Supplementary Table 1). Taken together, the DNA and RNA from the offspring qualified for deep sequencing.
Gene and miRNA expression are altered in the PFC of the offspring with ASD. To determine whether dysregulated gene expression was present in the postmortem PFC of offspring with ASD, we compared gene expression patterns of the offspring with and without ASD on a genome-wide scale. We focused on genes with expression levels of more than 0.3 FPKM in the offspring without ASD and a fold change between the offspring with and without ASD larger than 0.5 or less than −0.5, on a base-10 logarithmic scale ( Fig. 1b and Supplementary Table 2). Rather than use the typical two-fold change, we used a fold change of 3.16 (=10^(0.5)), which is a more stringent threshold. When compared to the offspring without ASD, we detected 2293 up-regulated and 5980 down-regulated gene isoforms in the PFC of the offspring with ASD. We then performed gene ontology (GO) enrichment analysis on the affected gene isoforms to further investigate which functions, processes, and components were affected. Focusing on gene isoforms involved in brain-related functions, we observed that genes related to the development of dendrites, axons and the olfactory bulb were affected ( Fig. 1c and Supplementary Table 3). None of the GO terms survived after false discovery rate (FDR) correction. The functions of the genes related to olfactory bulb development (ID2, AGTPBP1 and SEMA3A) are related to overall brain development [35][36][37] . If we applied more stringent criteria (log 2 (ASD/non-ASD > 2 or −2 and q value < 0.05), we only detected 67 genes with altered expression in the offspring with ASD (Supplementary Table 4). Due to the small size of the genes, GO analysis could not be further pursued. To validate the accuracy of our heatmap results, we confirmed the expression levels of the top genes that were up-regulated (such as RNF220, SEMA3E, and SEPT5) and down-regulated (such as HLA-A, NPAS4, and TNFRSF8) with qPCR (Fig. 1d). These genes were chosen based on the availability of reliable primers, high expression levels, and a difference in expression level between ASD and non-ASD samples. To examine if the altered genes could be observed in other ASD cohorts, we compared our altered genes with RNA-Seq data from Dr. Weinberger's group 7 for the dorsolateral prefrontal cortex from three persons with ASD and three matched controls. We observed that expression of 2406 out of 8273 genes was also altered in this ASD cohort ( Fig. S3a and RNA-Seq data, we used the same criteria of altered gene expression as we used for our ASD cohort. To investigate if the altered genes could be observed in a larger ASD cohort, we compared our altered genes with RNA-Seq data from a study by Dr. Geschwind's group 10 of the cortex from persons with ASD and matched controls. We found that expression of 387 out of 1087 genes was also altered in this ASD cohort (Supplementary Table 6).
Small RNAs, such as miRNAs, affect gene expression, but whether dysregulated miRNAs contribute to the etiology of ASD has not been well studied. To address this question, we compared the expression levels of miRNAs from the PFC of the two offspring on a genome-wide scale. We focused on miRNAs with a fold change between the offspring with and without ASD larger than 0.5 or less than −0.5, on a base-10 logarithmic scale, which resulted in a threshold of fold change of 3.16 (=10^(0.5)). When PFC miRNAs from the offspring with ASD were compared with those of the offspring without ASD, we found 105 up-regulated and 125 down-regulated miRNAs in the offspring diagnosed with ASD ( Fig. 1e and Supplementary Table 7). We then performed GO analysis on the affected miRNAs ( Fig. 1f and Supplementary Table 8) and examined genes targeted by altered miRNAs (Supplementary Table 9). None of GO terms survived after FDR correction. To validate the accuracy of our heatmap results, we performed qPCR to confirm the expression levels of the top up-regulated miRNAs (such as miR365a-3p, miR-2277-3p, and miR-184-3p) and down-regulated miRNAs (such as miR-21-5p, miR-136-3p, and miR-127-3p) (Fig. 1g). These miRNAs were chosen based on the availability of reliable primers, high expression levels, and a difference in expression level between ASD and non-ASD samples. To determine whether those altered miRNAs could be detected in other ASD cohorts, we compared our altered miRNAs with RNA-Seq data from Dr. Weinberger's group 7 for the dorsolateral prefrontal cortex of three persons with ASD and the three matched controls, using same criteria for analysis as we used in our ASD cohort. We observed that expression of 12 out of 210 miRNAs was also altered in this ASD cohort (Fig. S3b and Supplementary Table 5). To investigate if the altered miRNAs could be detected in a larger ASD cohort, we compared our altered miRNAs with RNA-Seq data from a study by Dr. Geschwind's group 16 of the cortex from persons with ASD and matched controls. We found that expression of 1 out of 58 miRNAs was also altered in this ASD cohort (shaded genes, Supplementary Table 6). Taken together, our results showed the PFC of the offspring with ASD contained genes and miRNAs with altered expression. Autism susceptibility genes were altered in the PFC of the offspring with ASD. To investigate whether autism susceptibility genes were preferentially altered, we compared expression levels of autism susceptibility genes in the postmortem PFC of the two offspring. The source of autism susceptibility genes was from the Simons Foundation Autism Research Initiative (SFARI, September 2016; Supplementary Table 10). We focused on genes with expression levels of more than 0.3 FPKM in the offspring without ASD and a fold change between the offspring with and without ASD larger than 0.5 or less than −0.5, on a base-10 logarithmic scale and as a result, the threshold for fold change used was 3.16 (=10^(0.5)). Compared with the offspring without ASD, the PFC of the offspring with ASD exhibited 142 up-regulated and 312 down-regulated autism susceptibility gene isoforms ( Fig. 2a and Supplementary Table 11). We performed a hypergeometric test to assess whether the above overlaps were more than one would expect by chance and found significant overlaps (p < 0.001). To validate the heatmap results regarding differences in regulation of these gene isoforms in the offspring with ASD, we compared the offspring's top up-regulated autism susceptibility genes (such as AGAP1, EFR3A and KAT6A, Fig. 2b) and down-regulated autism susceptibility genes (such as NRXN2, SERPINE1, and BBS4, Fig. 2c). These genes were chosen based on the availability of reliable primers, high expression levels, and a difference in expression level between ASD and non-ASD samples. To determine whether the altered ASD susceptibility genes could be observed in other ASD cohorts, we compared our altered ASD susceptibility genes with RNA-Seq data from Dr. Weinberger's group 7 for the dorsolateral prefrontal cortex of three persons with ASD and three matched controls. We observed altered expression in 51 out of 455 genes in this ASD cohort ( Fig. S3c and Supplementary Table 5). For analysis of the data in the cohort from Dr. Weinberger's data, we used the same criteria for the ASD susceptibility genes we used in our ASD cohort. To investigate if the altered genes could be observed in a larger ASD cohort, we compared our altered genes with RNA-Seq data from a study by Dr. Geschwind's group 10 of the cortex from persons with ASD and matched controls. We found that expression of 29 out of 1087 genes was also altered in this ASD cohort (shaded genes, Supplementary Table 6).
When compared with the offspring without ASD, the following autism susceptibility miRNAs 16,38 were altered in the offspring with ASD: miR-619-5p, miR-23a-3p, miR-103a-3p, miR-106b-5p, miR-143a-3p, miR-146a-5p, and miR-204-3p ( Fig. 2d and Supplementary Table 12). A hypergeometric test showed the above overlaps were not significant (p = 0.92). Since miRNAs regulate gene expression, we also examined genes targeted by the affected miRNAs (Supplementary Table 12). We then performed gene ontology (GO) enrichment analysis on the affected miRNA target genes to further investigate which functions and pathways were affected. However, we did not observe any significant results. Our data showed altered expression of autism susceptibility genes from SFARI in the PFC from the offspring with ASD.
Allele-specific gene expression was altered in the postmortem PFC of the offspring with ASD.
To examine allele-specific gene expression and to determine whether dysregulated allele-specific gene expression occurred in the offspring with ASD, the parents' and offspring's genomic DNA was analyzed with DNA-Seq, and the offspring's postmortem PFC RNA was analyzed with RNA-Seq followed by ASE analysis. First, we observed a distinct allele-specific gene expression pattern, which contained a diagonal line with one paternally-dominant cohort and one maternally-dominant cohort in both offspring ( Fig. 3a and Supplementary Table 13). Genes within the diagonal line indicate both of their alleles were expressed equally, which represented the majority of genes; genes above or below the diagonal line indicate genes that were expressed predominantly from either maternal or paternal allele. This pattern has been observed consistently in different brain regions related to the mouse visual system 39 as well as in cell types of the mouse visual cortex 40 . In contrast, when we compared the ASE patterns of all of SFARI's autism susceptibility genes, several genes differed between the offspring with and without ASD ( Fig. 3b and Supplementary Table 14). A hypergeometric test demonstrated these differences in genes between siblings were not by chance (p < 0.001). We validated allele-specific expression for our candidate genes with Sanger sequencing and determined LRP2BP and ZNF407 were both mono-allelically expressed in the offspring without ASD, but bi-allelically expressed in the offspring with ASD ( Fig. 3c,d, top). LRP2BP was also monoallelically expressed in other non-ASD brain samples (Fig. S4a). This mono-to-biallelic switch reflects their expression levels ( Fig. 3c,d, bottom). Because there is an overlap between autism susceptibility genes from SFARI (Supplementary Table 10, September 2016) and known human imprinted genes from the Geneimprint website (Supplementary Table 15, September 2016) (Fig. 3e), we validated seven of the 19 overlapped genes with Sanger sequencing (Fig. 3f,g). Due to the lack of availability of SNPs and low gene expression levels, the remaining overlapped genes could not be validated. For genes that could be validated, Sanger sequencing showed ATP10A, CTNNA3, DLGAP2, GABRB3, and HTR2A were not imprinted in either of the offspring (Fig. 3f), whereas MAGEL2 and SNRPN were imprinted in both (Fig. 3g). The imprinting status of ATP10A, DLGAP2, and HTR2A was further confirmed in other brain samples (Fig. S4b). Taken together, our data suggest that allele-specific gene expression occurs in human PFC and dysregulated allele-specific gene expression occurred in the PFC of the offspring with ASD.
Allele-specific miRNA expression was altered in the PFC of the offspring with ASD. To determine whether allele-specific expression of miRNAs occurs in the PFC of humans and whether the expression is dysregulated in persons with ASD, we profiled allele-specific miRNA expression in the PFC of the offspring with and without ASD. We observed a pattern of ASE for miRNAs, which showed a diagonal line with a paternally-dominant cohort and a maternally-dominantly cohort (Fig. 4a), similar to that seen for genes (Fig. 3a). In addition, this pattern has been observed consistently in different brain regions related to the mouse visual system 39 . Heatmap clustering compared miRNAs from the two offspring whose fold change on a base-10 logarithmic scale was larger than 0.5 or less than −0.5 ( Fig. 4b and Supplementary Table 16), and as a result, the threshold for fold change used was the 3.16 (=10^(0.5)). We validated the allele-specific miRNA expression with Sanger sequencing (Fig. 4c,d). There was no difference in allele-specific miRNA expression between offspring without and with ASD. However, we identified maternally-expressed miRNAs (miR-299 and miR-654) in both offspring (Fig. 4d). In summary, our data show distinct patterns of allele-specific miRNA expression in the PFC of both offspring. Table 17). We mapped the confirmed imprinted genes and miRNAs into 23 human chromosomes (Fig. 5). This mapping identifies the parent-of-origin-specific genes and miRNAs expressed on a genome-wide scale in the human PFC. Canonical genomic imprinting involves silencing of the maternal and paternal allele. In contrast, noncanonical genomic imprinting involves maternal or paternal allele expression biases 42 . We identified a noncanonical imprinted gene, NOS1 (Fig. 6a) in the PFC of both offspring.
Novel transcriptional processes of autism susceptibility genes were identified in human PFC. Validation of allele-specific expression in the PFC of both offspring with Sanger sequencing identified three novel transcriptional processes of autism susceptibility genes. First, we identified a novel human-specific site of RNA editing in KMT2C (Fig. 6b). When we analyzed RNA editing in human adult and fetal PFC (Fig. 6c), we observed T-to-C RNA editing of KMT2C sense transcript, which was in contrast to A-to-G RNA editing for KMT2C antisense, which could cause a phenylalanine-to-leucine change at the protein level (p. F291L). This pattern of RNA editing was not seen when we analyzed PFC from postnatal day 28 (P28) and embryonic day 15.5 (E15.5) mouse (Fig. 6d,e, respectively), suggesting this RNA editing in KMT2C is human-specific. These differences in RNA editing patterns between human and mouse tissue were also seen when we examined human and mouse blood (Fig. 6c,e). Second, we identified a development stage-and brain-specific maternally-expressed gene, DUSP22 (Fig. 7). We found that DUSP22 was maternally expressed in adult PFC but bi-allelically expressed in fetal PFC and adult blood (Fig. 7b). The imprinting status of DUSP22 has been validated in other fetal tissue (Fig. S4b). Because we have been unable to identify the exonic SNPs in mouse Dusp22, we have not examined the ASE pattern of Dusp22 in mouse PFC and blood. Finally, we identified a development stage-specific paternally-expressed miRNA, miR-335 (Fig. 8), which was paternally-expressed in the adult PFC but paternally-biased in the fetal PFC. Taken together, our results add new information about the dynamic transcriptomic processes in the brain.
Discussion
Our genome-wide analysis provides new information regarding expression of allele-specific genes and miRNAs in human PFC and persons with ASD. Analysis of the PFC revealed a distinct allele-specific expression pattern, which contained a diagonal line with a maternally dominant cohort and a paternally dominant cohort. Moreover, we identified novel allele-specific genes such as DUSP22 and miRNAs such as miR-335 in both the offspring with . miR-335 is paternally expressed in the PFC of the offspring without and with ASD. Sanger sequencing was performed to analyze allele-specific miR-335 expression in the PFC of the offspring without and with ASD, fetal PFC, and blood from the parents of the fetus. "P" stands for paternal expression. "B" stands for biallelic expression. "P*" stands for paternally-biased expression. SNP information is shown as (paternal allele/maternal allele). and without ASD, as well as a mono-to-biallelic switch for LRP2BP and ZNF407 in the offspring diagnosed with ASD. We also identified a novel human-specific site of RNA editing in KMT2C. Importantly, our study results indicate that a genome-wide ASE map could provide a powerful model for understanding neuropsychiatric disorders through the study of key features of dynamic allele-specific gene and miRNA expression in human PFC, supplemented with the study of the roles of dysregulated allele-specific genes and miRNAs in ASD.
Dual specificity phosphatase 22 (DUSP22) is an enzyme, which activates the JNK signaling pathway 43 . JNK activation has been shown to play an essential role in organogenesis during mouse development by regulating cell survival, apoptosis, and proliferation 44 . The physiological role of DUSP22 in the brain is unclear, and a rare DUSP22 deletion was found in a patient with autism and mild intellectual disability 45 . DUSP22 in the PFC from the offspring without and with ASD consistently showed maternal expression. Our finding further extends a previous finding showing DUSP22 is a monoallelically-expressed gene 46 . Interestingly, DUSP22 from fetal PFC showed bi-allelical expression, therefore it would be of interest to investigate how a biallelic-to-monoallelic switch is regulated for DUSP22 during prefrontal development. Moreover, LRP2BP and ZNF407 in the PFC from the offspring with ASD consistently showed a mono-to-biallelic switch. Importantly, dysregulation of LRP2BP 47 and ZNF407 48 has been identified in persons with ASD. However, the roles of LRP2BP and ZNF407 in the brain have not been identified and the function of LRP2BP protein is still unknown. In contrast, the function of ZNF407 protein has been shown to regulate glucose homeostasis 49 . Therefore, because glucose homeostasis is critical for normal brain function, dysregulation of ZNF407 could affect brain development and function. It would be of interest not only to investigate how a mono-to-biallelic switch is regulated for LRP2BP and ZNF407, which could provide insight into their roles in ASD specifically, but also to study the roles of LRP2BP and ZNF407 more generally during neurodevelopment. The PFC from the offspring without and with ASD consistently showed paternal expression of miR-335, which further extends a previous finding showing mouse miR-335 is paternally-expressed 50 .
Lysine methyltransferase 2 C (KMT2C) has histone methylation activity and is a transcriptional coactivator. In our studies, we found that T-to-C RNA editing occurred in the KMT2C sense transcript. This editing causes a missense mutation, which converts phenylalanine to leucine at amino acid 291 of KMT2C. Phenylalanine and leucine are both neutral and non-polar amino acids, therefore, this change should have a minor effect on the protein structure of KMT2C. Indeed, when we used SIFT ( [URL]/) to predict whether such amino acid substitution affects protein function, the result predicted that this change is a tolerated substitution. It would be of interest to investigate the physiological significance of the T-to-C editing event in the sense transcript and an A-to-G editing event in the antisense transcript.
It has been reported that parent-of-origin-specific expression is also brain-region specific 51 . Our results demonstrate expression of parent-of-origin-specific genes and miRNAs in human PFC, which has not previously been reported. Our transcriptomic and ASE analysis of the PFC of the offspring with and without ASD of a family quartet identified dysregulated gene and miRNA expression, and ASE in the offspring with ASD, which could identify a novel set of autism susceptibility genes and miRNAs. It will be important to identify the regulatory mechanisms for dysregulated transcriptomic and allele-specific expression of these genes and miR-NAs identified in the offspring with ASD. To this end, it will be essential to determine if stringent identification of allele-specific genes together with the systematic screening of allele-specific chromatin and DNA modifications could unravel markers for further mechanistic validation. It would also be interesting to further investigate whether cis-transcribed non-coding RNAs, intra-nuclear allelic positioning, and chromosomal interactions are associated with the allele-specific expression.
Several methods are available for seeking the allele-specific expressed genetic locus on a genome-wide scale, which include in silico prediction pipelines 52 , SNP genotyping arrays 53 , gene expression arrays 54 and transcriptome sequencing approach 55 . The transcriptomic approach is based on detecting allelic expression with RNA-Seq reads that map heterozygous SNPs, where the identity of the base is used to distinguish allelic origin and a reciprocal cross is used to discriminate parent-of-origin specificity from strain-specific or random biases. The transcriptomic approach for allele-specific gene expression is a paradigm shift in comparison to previous methods. However, recent literature has indicated a high FDR could explain the majority of novel imprinted genes in an RNA-Seq approach resulting from the contribution of several factors. First, systematic errors in technical and biological replicates include priming, fragmentation, and PCR biases that arise during library construction and sequencing. One can adopt a mock reciprocal cross as a negative control to set a false discovery cutoff for systemic errors. Second, the strain-specific effect could be due to cis-eQTL and tissue-specific effect could be due to trans-eQTL. Third, in comparison to the inbred mouse, the human is an outbred species and the complexity of haplotype phasing is much higher, which might lead to wrong read alignment (mapping bias). Since we cannot acquire a large sample size of SNPs calling data from the same population as the sampled genome, the only way to perform haplotype calling is to make inference via an established reference haploid genome from the same population. If the reference haploid genome distinctly differs from the sampled genome, then FDR inevitably rises. Recently, a Bayesian approach for analysis of ASE using a personal diploid genome as a reference sequence has been established and shows less biased alignments and more consistent ASE 56 . Fourth, small sample size and confounding underlying diseases (e.g. epilepsy, congenital deafness, or intellectual disability) may contribute to false detection. Fifth, in our study, we only had tissue from the prefrontal cortex of the offspring. However, other brain regions, such as the cerebellum, are also reported to be related to the anatomical and neuropathological causes of ASD 57 and have different miRNA expression patterns 16 in comparison with the prefrontal area.
Bipolar disorder was also present in the offspring with ASD. Because the age of onset for autism (2 to 3 years of age) is much earlier than that of bipolar disorder (25 years) and the offspring with ASD died at age 29 years, the impact of autism should be higher than that of bipolar disorder. In addition, many psychiatric disorders share genetic roots 58 . For example, there is an overlap between rare genetic variations linked to bipolar disorder and those implicated in autism 59 at least one comorbid non-ASD developmental disorder 60 , therefore, it would be difficult to acquire cases of persons with ASD only. The results obtained here are more likely due to ASD rather than bipolar disorder.
Due to tissue availability, we sampled different regions of the prefrontal cortex (BA10 for the offspring without ASD and BA8 for the offspring with ASD) to conduct the analysis. The differential expression analysis may be confounded by the tissue specificity of gene expression in BA8 and BA10. To explore the impact of this potential confounding factor and verify the similarity of these two regions, we first compared the expression patterns of different regions of the central nervous system with principle component analysis (PCA) (Fig. S6a,b). PCA demonstrated similarities between BA8 and BA10 in comparison with the cerebellum and frontal cortex by PC2. It also showed a strong difference of BA8/10 from the amygdala, caudate nucleus, and spinal cord by PC1. Heatmap analysis again showed similarities between BA8 and BA10 in comparison to other regions of the central nervous system (Fig. S6c). To further confirm whether the differential gene expression in Figs 1b and 2a was due to brain region, we used heatmap and volcano analysis to examine those differentially expressed genes and observed that less than 7% of those genes were significantly affected (Fig. S6d to g and Supplementary Table 18). These findings show the two regions are similar, which supports our use of BA8 and BA10 for the comparison of gene expression.
In spite of these potential limitations and confounding factors, our results provide valuable clues for identifying biomarkers and biological signatures of ASD, and increase our understanding of potential underlying mechanisms that contribute to the pathogenesis of the disorder. Our findings not only advance our knowledge of allele-specific gene and miRNA expression in ASD, but also provide the first genomic map for allele-specific gene and miRNA expression in human PFC. These results could also provide clues to the evolutionary development of allele-specific expression. In addition, therapeutic targets and strategies for brain disorders such as ASD, as well as those involving ASE, could be determined. This will require development and validation of a plausible pipeline for ASE analysis in order to identify new genetic candidates for epigenetic mechanisms related to neuropathological characteristics of ASD, which could serve as targets for therapies of ASE-linked neurological disorders.
Methods
Subject material. We studied one family quartet comprised of parents and two offspring with and without ASD. In addition, one family trio, comprised of parents and one fetal offspring, was examined. Additional post-mortem samples included one fetal frontal cortex and one adult frontal cortex and cerebellum, shown in Supplementary Fig. 4. We assessed prefrontal cortex tissue from the dorsorostral pole of the frontal lobe corresponding to Brodmann's area (BA) 10 for the offspring without ASD, and BA 8 for the offspring with ASD. Areas were based on tissue availability. Supplementary Fig. 1 and Supplementary Table 1 provides a more detailed pedigree of the family quartet showing epilepsy and deafness co-occurred in both the affected and unaffected offspring in addition to other familial health conditions. Moreover, we assessed prefrontal cortex tissue from the fetus and blood from the parents of the fetus. The detailed information of this family trio is shown in Supplementary Table 1 Detection of variants. DNA from the parents and offspring was extracted and sequenced according to standard whole-genome sequencing protocol. The sequencing reads were trimmed with Trimmomatic to obtain the qualified reads. The reads were then aligned to the human reference genome GRCh38 using BWA and processed with SAMtools. Picard ( [URL]/) was implemented to mark the duplicate reads and exclude them from downstream analyses. The read alignments were further refined with GATK for local realignment of reads around known insertions and deletions (indels) and recalibration of base quality. GATK was also applied to call single-nucleotide variants (SNVs) and short indels. Data from SNVs were consequently used to construct the haplotype scaffolds as described below. The RNA-Seq and DNA-Seq data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number GSE98581.
Allele-specific expression analysis. The SNV data from the parents and offspring were processed with VCFtools. SHAPEIT was then used to phase the SNV data with the family pedigree for building phased haplotype scaffolds. To improve phasing accuracy, the information of recombination rates between SNPs was provided via a genetic map retrieved from The 1000 Genomes Project Phase 3. The reference panel of phased haplotypes belonging to Utah residents with Northern and Western European ancestry (CEU) from The 1000 Genomes Project Phase 3 was also applied to align SNPs between the dataset and the panel for assisting in reliable phasing. The information of phased haplotypes was subsequently analyzed with in-house scripts to create haploid genomic sequences for the parents and offspring based on the human reference genome GRCh38. For RNA-Seq analysis, the sequences generated were filtered to obtain qualified reads. ConDeTri was implemented to trim or remove the reads according to the quality score. Qualified reads after filtering low-quality data were analyzed using TopHat/ Cufflinks for gene expression estimation. The gene expression level was calculated as FPKM (Fragments Per Kilobase of transcript per Million mapped reads). A comprehensive analysis of the tissue using mammalian transcriptome data sets suggests that a lower cutoff of FPKM = 0.3 is often justifiable therefore we applied this cutoff for all analysis of mammalian transcriptomes. For differential expression analysis, CummeRbund was employed to perform statistical analyses of gene expression profiles. For allele-specific expression analysis, MMSEQ was then implemented to estimate allelic imbalance and deconvolve the alignment of reads to diploid transcripts derived from diploid genomic sequences and Ensembl gene annotation 74 following the mapping of RNA-Seq reads with Bowtie. SNPs for confirming imprinted miRNAs are within primary miRNA sequence. We defined the sequence for primary miRNAs as the genomic locus from 500 bp upstream to 500 bp downstream of the mature miRNA sequence. The FPKM value is much less than from regular read analysis because only the read with SNPs can be used for allele-specific expression analysis.
Reverse transcription quantitative PCR (RT-qPCR). Total RNA was extracted from the postmortem PFC of the offspring using a NucleoSpin ® miRNA kit (Macherey-Nagel, 740971). Total RNA (10 ng) was converted to cDNA and amplified by One Step SYBR ® PrimeScriptTMRT-PCR Kit II (Takara, PR086A). Quantitative real-time PCR was performed with a StepOnePlus Real Time PCR System (Applied Biosystems). Ct values were generated using StepOne Software version 2.2.2. The expression level of each gene was normalized to B2M. All primer sequences of candidate genes were designed by Primer3 software ( [URL]3-0.4.0/) and are shown in Supplementary Table 19. miRNA quantification. We extracted miRNA using a NucleoSpin ® miRNA kit (Macherey-Nagel, 740971).
Because the length of miRNA is too short to perform normal qRT-PCR, the miRNA was lengthened with a Poly(A) tail (Poly(A) Tailing Kit; Ambion, AM1350). The poly(A) tailed miRNA was reverse transcribed into cDNA with a poly(T) anchor adaptor. The miRNA was amplified and quantitated by qPCR using a specific miRNA forward primer and a universal adaptor primer. Information of primer sequence is shown in Supplementary Table 20.
Similarly, the qPCR product was too small for Sanger sequencing. Therefore, to determine the sequence, the amplified PCR product was inserted into a plasmid vector, and the vector was transformed into bacteria with the TOPO TA Cloning Kit (Invitrogen, 450071), cloned, and cultured according to the manufacturer's directions. Plasmid DNA containing the inserted qPCR product was extracted with the Presto Mini Plasmid kit (Geneaid, PHD100). Sanger sequencing of the purified plasmid used M13-tailed primers, which can yield sequences up to approximately 200 bp.
Graphic representation and statistical analysis. Heatmaps were generated with Pretty Heatmaps software (pheatmap package in R 3.3.2). The hierarchical clustering of heatmaps and the supplementary tables were measured in Euclidean distance. Micro-RNA target prediction was performed via DIANA microT-CDS web-based program. Pedegree was generated with Genial Pedigree Draw (www.pedigreedraw.com). Gene ontology (GO) analysis were performed via web-based Gorilla program and miEAA program for gene set enrichment analysis (GSEA) adapted for miRNA. Statistical analysis and graphic illustrations were performed using R 3.3.2 and Sigmaplot 13.0. All values with technical triplicates are expressed as the mean ± standard error of mean (s.e.m.). Data availability statement. All data are available in this manuscript.
==
Domain: Psychology Biology
|
[{"added": "2023-02-18T14:45:39.614Z", "created": "2018-03-09T00:00:00.000", "id": "256956892", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/s41598-018-22753-4.pdf", "pdf_hash": "599eb2d775ecb578bfa76863514eaec7235e39f5", "pdf_src": "SpringerNature", "provenance": "peS2o-0075.json.gz:32454", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "599eb2d775ecb578bfa76863514eaec7235e39f5", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Pro-Inflammatory Priming of the Brain: The Underlying Cause of Parkinson’s Disease
Parkinson’s disease (PD) is a multifactorial neurodegenerative pathology characterized by the progressive loss of dopaminergic neurons in the substantia nigra of the brain. Aging is considered the main risk factor for the development of idiopathic PD. However, immunity and inflammation play a crucial role in the pathogenesis of this disorder. In mice, we showed that pro-inflammatory priming of the brain sensitizes to severe PD development, regardless of animal age. Age-related sub-acute inflammation, as well as the activation of the immune response upon exposure to harmful stimuli, enhances PD manifestations. The severity of PD is influenced by the engagement of host resistance mechanisms against infection based on the removal of iron (Fe) from the circulation. The sequestration of Fe by immune cells prevents pathogens from proliferating. However, it leads to the formation of a Fe-loaded circulating compartment. When entering the brain through a compromised blood-brain barrier, Fe-loaded immune cells contribute to enhancing neuroinflammation and brain Fe overload. Thus, pro-inflammatory priming of the brain exacerbates neuronal damage and represents a risk factor for the development of severe PD symptoms. Further investigations are now required to better understand whether therapeutic interventions inhibiting this phenomenon might protect against PD.
Introduction
Parkinson's disease (PD) is a progressive and multifactorial neurodegenerative disorder. PD etiology is still poorly understood. Aging is considered the main risk factor. However, genetic and environmental influences were also correlated to PD onset [1]. The lack of direct evidence causing PD hindered the development of specific therapies. There is no cure for this pathology. Available drugs might alleviate PD symptoms and retard its progression but do not prevent disease manifestations [2,3].
PD is caused by the loss of dopaminergic neurons (DNs) in the substantia nigra of the brain [4,5]. Its prevalence doubled since 2000. In 2019, approximately 8.5 million people were diagnosed with PD. This represents 1% of the entire population over 60 years and 5% of individuals over 85 years of age [6]. PD is associated with the development of motor and non-motor symptoms. Clinical signs range from tremors at rest, bradykinesia, dystonia and gait disturbances to mood and sleep disorders [7]. Cognitive impairment affects 30-40% of PD patients, who present an increased risk of developing severe dementia [8]. PD is also associated with the development of gastrointestinal (GI) manifestations, frequently reported before the appearance of motor symptoms [9][10][11]. The direct communication between the gut and the brain is provided by the autonomous nervous system [12,13]. Truncal vagotomy was shown to diminish the risk for PD development [14]. Hence, the gut started to be investigated as a potential target for therapeutic intervention against PD. This was also encouraged by the existence of a retrograde transport of α-synuclein (α-syn) to developed by young animals in response to MPTP (Figure 1b). However, we confirmed that aging increases PD severity. This finding was also corroborated by the pronounced neuroinflammatory phenotype, as shown by the higher activation of microglia in old mice that were exposed to MPTP (Figure 1c-e). The quantification of pro-inflammatory markers, like MHCII and CCR2, confirmed the results obtained (Supplementary Figure S1a,b). Neuroinflammation is possibly enhanced by the infiltration of peripheral immune cells into the brain of old mice exposed to MPTP. An increase in CD4 Helper and CD8 Cytotoxic T cells was detected in the brain of PD-induced aged animals (Figure 1f). CD44 was used as a marker of activation and the expression of CD62L as a naïve cell phenotype, which is lost during cell differentiation. CD44 expression was higher in both immune cell populations in the same mice (Figure 1g,h). The correspondent plot quantification confirmed increased neuroinflammation developed by old mice in relation to young animals upon PD induction (Supplementary Figure S1c-f).
Overall, these data confirmed that aging is associated with the development of a more severe PD phenotype in response to MPTP. l. Sci. 2023, 24, x FOR PEER REVIEW 3 of 21 note that these results also validate the use of MPTP as a pharmacologic PD model, in agreement with the scientific literature [47]. No significant differences in DNs loss were observed in young mice, induced or not with PD. This result is possibly justified by milder symptoms developed by young animals in response to MPTP (Figure 1b). However, we confirmed that aging increases PD severity. This finding was also corroborated by the pronounced neuroinflammatory phenotype, as shown by the higher activation of microglia in old mice that were exposed to MPTP (Figure 1c-e). The quantification of pro-inflammatory markers, like MHCII and CCR2, confirmed the results obtained ( Supplementary Figure S1a,b). Neuroinflammation is possibly enhanced by the infiltration of peripheral immune cells into the brain of old mice exposed to MPTP. An increase in CD4 Helper and CD8 Cytotoxic T cells was detected in the brain of PD-induced aged animals ( Figure 1f). CD44 was used as a marker of activation and the espression of CD62L as a naïve cell phenotype, which is lost during cell differention. CD44 expression was higher in both immune cell populations in the same mice (Figure 1g,h). The correspondent plot quantification confirmed increased neuroinflammation developed by old mice in relation to young animals upon PD induction (Supplementary Figure S1c-f). . Non-treated 52-60 weeks (old) mice were used as controls. Motor dysfunction was evaluated by assessing the time of performance on a pole test. Mice were monitored for 30 days after MPTP administration. (b) Tyrosine Hydroxylase (TH) mRNA expression, quantified by qRT-PCR upon isolation of the substantia nigra of the brain of mice as in (a). The results were normalized to GADPH, used as a housekeeping gene, and expressed as mean ± SD (n = 10 mice per group
Aging-Associated Inflammation Promotes the Formation of Fe-Loaded Inflammatory Cells
Comparative analyses were carried out to better understand the causes underlying an increased sensitivity of old mice to PD when compared to young animals. We recently demonstrated that aged mice develop a neuroinflammatory phenotype, a physiological process also occurring during healthy aging [31]. Increased permeability of the BBB was shown to contribute to this phenomenon. In the absence of PD, a higher expression of adhesion molecules, ICAM and VCAM, was observed by flow cytometry in the brain of old mice when compared to young animals ( Figure 2a). The correspondent plot quantification confirmed the data obtained (Supplementary Figure S2a,b). An impaired BBB integrity was found in aged animals. BBB permeability was measured by the extravasation of Evans Blue dye into the brain when administered intravenously (Figure 2b). A compromised BBB favored the infiltration of immune cells into the brain. We performed in vitro experiments to test the ability of peripheral cells to switch microglia towards a pro-inflammatory phenotype. Isolated primary microglia were stimulated with sera collected from young and old animals, and the release of pro-inflammatory cytokines was measured by ELISA. Higher levels of TNF and IL-6 were detected in response to sera from old mice (Figure 2c,d). To understand the reason for this phenomenon, we assessed its correlation with the generation of Fe-loaded cells [31]. Calcein quenching is an indirect method to measure Fe overload. However, it indicates the intracellular accumulation of Fe in circulating immune cells. Differences in calcein quenching were observed when comparing young and old mice ( Figure 2e). The notion that peripheral immune cells were Fe-loaded was confirmed by quantifying the expression of FtH by qRT-PCR. An increased FtH was found in circulating immune cells when isolated from aged mice (Figure 2f). To mimic the damage induced by Fe-loaded cells to neurons, we exposed primary DNs in vitro to sera collected from young and old mice. An increased neuronal death was observed when neurons were treated with sera collected from aged animals. Pre-treatment with Fe chelator, deferiprone, was found to prevent DNs loss (Figure 2g). Similar toxicity was observed when DNs were stimulated with supernatant collected from microglia, which were previously exposed to sera collected from old mice (Figure 2h). The treatment turned microglia pro-inflammatory and capable of triggering neuronal damage.
Figure 2. Aging-associated inflammation promotes the generation of inflammatory Fe-loaded cells.
All mice were of a C57BL/6 background (a) Gating strategy for the detection of adhesion molecules, VCAM and ICAM, from blood-brain barrier endothelial cells (CD31+), freshly isolated from the brain of non-treated 8-12 weeks old (young) and 52-60 weeks (old) mice. (b) BBB permeability, measured by the extravasation of Evans Blue dye into the brain and referred to as Brain Edema, in young vs. old mice, as in (a). (c) TNF and (d) IL-6 levels, measured by ELISA, in the supernatant collected from microglia, upon stimulation with sera collected from mice as in (a). (e) Calcein median fluorescence intensity, measured by flow cytometry, in immune cells isolated from sera collected from mice as in (a). (f) mRNA expression of FtH, measured by qRT-PCR in cells as in (e) collected from mice as in (a), which results were normalized to ArbP0 (g) Neuronal viability, assessed by Crystal Violet staining, upon stimulation with sera collected from young and old mice, pre-treated or not with deferiprone (DFP). (h) Neuronal viability was assessed by Crystal Violet staining in response to stimulation with supernatants collected from microglia previously exposed to sera from young and old mice. The results were expressed as mean ± SD (n = 7-10 mice per group). The Student's t-test was applied to define statistical differences.
Overall, these data indicate that PD results from the accumulation of peripheral events, which critically contribute to the pathogenesis of this disease. All mice were of a C57BL/6 background (a) Gating strategy for the detection of adhesion molecules, VCAM and ICAM, from blood-brain barrier endothelial cells (CD31+), freshly isolated from the brain of non-treated 8-12 weeks old (young) and 52-60 weeks (old) mice. (b) BBB permeability, measured by the extravasation of Evans Blue dye into the brain and referred to as Brain Edema, in young vs. old mice, as in (a). (c) TNF and (d) IL-6 levels, measured by ELISA, in the supernatant collected from microglia, upon stimulation with sera collected from mice as in (a). (e) Calcein median fluorescence intensity, measured by flow cytometry, in immune cells isolated from sera collected from mice as in (a). (f) mRNA expression of FtH, measured by qRT-PCR in cells as in (e) collected from mice as in (a), which results were normalized to ArbP0 (g) Neuronal viability, assessed by Crystal Violet staining, upon stimulation with sera collected from young and old mice, pre-treated or not with deferiprone (DFP). (h) Neuronal viability was assessed by Crystal Violet staining in response to stimulation with supernatants collected from microglia previously exposed to sera from young and old mice. The results were expressed as mean ± SD (n = 7-10 mice per group). The Student's t-test was applied to define statistical differences.
Overall, these data indicate that PD results from the accumulation of peripheral events, which critically contribute to the pathogenesis of this disease.
Brain Fe Accumulation and Inflammation Sensitize to an Enhanced PD Severity
Aging is associated with an increased level of brain Fe (Figure 3a). Microglia are considered the reservoir of Fe in the brain. Their ability to accumulate Fe was confirmed by calcein quenching, evaluated in microglia isolated from young and old animals ( Figure 3b). The pro-oxidant nature of Fe is capable of switching microglia towards a pro-inflammatory phenotype [48]. This was also demonstrated by the release of pro-inflammatory cytokines when these cells were exposed to Fe. An increased level of TNF was assessed in the supernatant of primary microglia treated with Fe ( Figure 3c). Increased cell death was also observed when DNs neurons were exposed to Fe (Figure 3d). While the neuroinflammation, developed by old mice, is associated with an increased brain Fe accumulation, no differences were found in young and old animals upon PD induction (Figure 3e). This indicated that Fe accumulation in the brain occurs before PD onset and is not induced by disease progression. Hence, we showed that the activation of the immune system primes the brain to Fe overload and neurodegeneration. The contribution of the immune system to PD development was confirmed by using mice deprived of adaptive immunity. Rag-2-deficient, TCRβ-deficient or JHT-deficient mice, depleted of T and/or mature B cells, were found to be more resistant to PD induction in relation to aged-matched wild-type animals ( Figure 3f).
Brain Fe Accumulation and Inflammation Sensitize to an Enhanced PD Severity
Aging is associated with an increased level of brain Fe (Figure 3a). Microglia are considered the reservoir of Fe in the brain. Their ability to accumulate Fe was confirmed by calcein quenching, evaluated in microglia isolated from young and old animals ( Figure 3b). The pro-oxidant nature of Fe is capable of switching microglia towards a pro-inflammatory phenotype [48]. This was also demonstrated by the release of pro-inflammatory cytokines when these cells were exposed to Fe. An increased level of TNF was assessed in the supernatant of primary microglia treated with Fe ( Figure 3c). Increased cell death was also observed when DNs neurons were exposed to Fe (Figure 3d). While the neuroinflammation, developed by old mice, is associated with an increased brain Fe accumulation, no differences were found in young and old animals upon PD induction (Figure 3e). This indicated that Fe accumulation in the brain occurs before PD onset and is not induced by disease progression. Hence, we showed that the activation of the immune system primes the brain to Fe overload and neurodegeneration. The contribution of the immune system to PD development was confirmed by using mice deprived of adaptive immunity. Rag-2deficient, TCRβ-deficient or JHT-deficient mice, depleted of T and/or mature B cells, were found to be more resistant to PD induction in relation to aged-matched wild-type animals ( Figure 3f). (e) Brain Fe content was measured in young mice, exposed or not to MPTP. All results presented so far were expressed as mean ± SD (n = 6-10 mice per group). (f) Locomotor dysfunction of young wild-type, Rag-2-deficient, TCRβ-deficient or JHT-deficient mice, aged between 8-10 weeks and in C57BL/6 background, intoxicated with high doses of MPTP (20 mg/kg, i.p., 4 injections 2 h apart). Motor dysfunction was evaluated by assessing the time of performance on a pole test. Mice were monitored for 30 days after MPTP administration. The results are expressed as mean ± SD (n = 10 mice per group). Statistical analysis was performed by applying the one-way ANOVA test.
Overall, these data indicate that the infiltration of peripheral immune cells into the brain is key to enhancing PD severity.
Infection-Driven Pro-Inflammatory Priming to the Brain Promotes Neuroinflammation
Considering the contribution of peripheral immune cells in enhancing PD severity, we assessed whether the activation of the immune system, in response to the infection, could mimic the phenotype observed in the brain of aged mice. Young animals were subjected to mild cecum ligation and puncture (CLP), and the results obtained were compared to mice exposed to sham surgery (S). The generation of Fe-loaded immune cells was observed in CLP-induced mice, not in S mice. The accumulation of Fe was assessed by calcein quenching (Figure 4a), and the expression of FtH in immune cells was measured by qRT-PCR ( Figure 4b). We previously demonstrated that the inflammatory status associated with aging correlates to increased permeability of the BBB. Its integrity was disrupted in mice subjected to CLP (Figure 4c). The increased expression of VCAM in CLP-induced mice was quantified by qRT-PCR and was shown to confirm previous findings (Supplementary Figure S3a Figure S3f-h). A time course was performed to assess whether the infiltration of immune cells into the brain upon CLP was an early symptom. The results obtained demonstrated that this is the case. Peripheral immune cells were detected in the brain a few days after CLP, as well as the activation of microglia (Supplementary Figure S4). This finding further confirms that the immune response elicited upon infection is a risk factor for the development of a neuroinflammatory phenotype.
Figure 4. Infection-driven pro-inflammatory priming to the brain promotes neuroinflammation.
All mice were of a C57BL/6 background (a) Calcein median fluorescence intensity, measured by flow cytometry, in leukocytes isolated from sera collected from mice subjected to sham surgery (S) or cecum ligation and puncture (CLP). (b) mRNA expression of FtH, measured by qRT-PCR in cells collected from mice as in (a), which results were normalized to ArbP0. (c) BBB permeability, measured by the extravasation of Evans Blue dye into the brain and referred to as Brain Edema, in young vs. old mice, as in (a). So far, the results were expressed as mean ± SD (n = 10 mice per group). The Student's t-test was applied to define statistical differences. Similar data were obtained when mice were exposed to LPS (Figure 5a-h; Supplementary Figure S5). The contribution of peripheral immune cells to microglia activation and neurodegeneration was also demonstrated in vitro. When primary microglia were exposed to sera collected from LPS-induced mice, the release of TNF in the supernatant increased (Figure 5i). When this supernatant was used to stimulate isolated DNs, we observed a reduction in cell viability (Figure 5j). Similar data were obtained when mice were exposed to LPS (Figure 5a-h; Supplementary Figure S5). The contribution of peripheral immune cells to microglia activation and neurodegeneration was also demonstrated in vitro. When primary microglia were exposed to sera collected from LPS-induced mice, the release of TNF in the supernatant increased (Figure 5i). When this supernatant was used to stimulate isolated DNs, we observed a reduction in cell viability (Figure 5j). Figure 5. Infection-driven pro-inflammatory priming to the brain promotes neuroinflammation. All mice were of a C57BL/6 background (a) Calcein median fluorescence intensity, measured by flow cytometry, in leukocytes isolated from sera collected from mice exposed or not to LPS (b) BBB permeability, measured by the extravasation of Evans Blue dye into the brain and referred as Brain Edema, in young vs. old mice, as in (a). So far, the results were expressed as mean ± SD (n = 10 mice per group). The Student's t-test was applied to define statistical differences. (c) Gating strategy, detailed in the legend of Figure 1 Figure 5. Infection-driven pro-inflammatory priming to the brain promotes neuroinflammation.
All mice were of a C57BL/6 background (a) Calcein median fluorescence intensity, measured by flow cytometry, in leukocytes isolated from sera collected from mice exposed or not to LPS (b) BBB permeability, measured by the extravasation of Evans Blue dye into the brain and referred as Brain Edema, in young vs. old mice, as in (a). So far, the results were expressed as mean ± SD (n = 10 mice per group). The Student's t-test was applied to define statistical differences. (c) Gating strategy, detailed in the legend of Figure 1, of freshly isolated infiltrated Helper (CD4+) and Cytotoxic (CD8+) T cells from mice as in (a), and their activation, assessed in (d) CD4 and (e) CD8 by the activation marker CD44. From the upper left quadrant, clockwise, plots identify naïve, memory and activated cells. (f) Gating strategy to obtain freshly isolated microglia (CD45int), within contour plots, from mice as in (a). (g) Activated microglia (CD45int MHCII+) and (h) (CD45int CCR2+), within contour plots, in mice as in (a). (i) TNF levels, measured by ELISA, in the supernatant collected from microglia, upon stimulation with sera collected from mice as in (a). (j) Neuronal viability, assessed by Crystal Violet staining, in response to stimulation with supernatants collected from microglia previously exposed to sera from mice as in (a).
Pro-Inflammatory Priming to the Brain Enhances PD Severity
As in previous experiments, we exposed microglia to sera collected from CLP-induced mice, and we assessed if those cells became pro-inflammatory. An increase in the release of TNF was measured in the supernatant of microglia stimulated with sera from CLPinduced mice (Figure 6a). In contact with DNs, the same sera induced neuronal death; cytotoxicity was prevented by pre-treating cells with deferiprone ( Figure 6b). Increased BBB permeability was shown by CLP-induced mice. In these animals, peripheral inflammation was associated with brain Fe accumulation (Figure 6c). PD severity was enhanced in mice that underwent CLP before MPTP induction. Locomotor dysfunction significantly increased when compared to S-induced mice. Mice subjected to CLP but not receiving MPTP were used as controls (Figure 6d). These findings revealed that inflammation primes the brain and sensitizes it to severe PD. Since this phenomenon is characterized by neuroinflammation and Fe accumulation, we assessed whether exposing mice to Fe chelation therapy upon infection conferred protection to the brain. Reduced infiltration of activated peripheral leukocytes was found in the brain of CLP-induced mice, which were therapeutically treated with deferiprone ( Figure 6e; Supplementary Figure S6a,b). A reduced microglia activation was also observed, as assessed by flow cytometry, quantifying the expression of the MHCII activation marker (Figure 6f; Supplementary Figure S6c). When mice received deferiprone, starting five days after CLP, a reduced PD severity was observed in relation to PD mice subjected to CLP but not receiving Fe chelation. CLP mice not exposed to MPTP were used as controls ( Figure 6g). As Fe is a pro-oxidant, we assessed whether the combination of an antioxidant and Fe chelation therapy prevents CLP mice from developing severe locomotor dysfunction in response to MPTP. Our findings showed that this was the case. The administration of deferiprone and N-acetylcysteine (NAC) significantly reduced PD severity in mice previously subjected to CLP. Beneficial effects were also observed by exposing those mice to NAC alone, although the locomotor dysfunction did not decrease to the same extent as when mice were receiving a combined therapy (Figure 6h).
Overall, these data demonstrated that while a pro-inflammatory priming of the brain enhanced PD severity, a combined treatment of Fe chelator and antioxidants protects mice against the development of severe PD symptoms. This indicates the existence of cross-talk between iron metabolism and inflammation, which ultimately affects the brain. Figure 6. Pro-inflammatory priming to the brain enhances PD severity. All mice were of a C57BL/6 background and aged 8-12 weeks (a) TNF levels, measured by ELISA, in the supernatant collected from microglia, upon stimulation with sera collected from mice subjected or not to CLP. (b) Neuronal viability, assessed by Crystal Violet staining, in response to stimulation as in a) receiving or not deferiprone (DFP) treatment. (c) Intracellular Fe content measured in the brain of Sham (S) or CLP mice. So far, the results were expressed as mean ± SD (n = 10 mice per group). The Student's t-test was applied to define statistical differences. (d) Locomotor dysfunction measured in C57BL/6 mice, induced or not with CLP and intoxicated with MPTP (15 mg/kg, i.p., 3 injections 2 h apart). Motor dysfunction was evaluated by assessing the time of performance on a pole test. Mice were monitored for 30 days after MPTP administration. (e) Gating strategy, detailed in the legend of Figure 1, of freshly isolated infiltrated Helper (CD4+) and Cytotoxic (CD8+) T cells from mice exposed to CLP and therapeutically treated or not with DFP. (f) Gating strategy to obtain freshly isolated activated microglia (CD45int MHCII+), within contour plots, in mice as in e). (g) Locomotor dysfunction in mice, induced or not with CLP, intoxicated with MPTP (15 mg/kg, i.p., 3 injections 2 h apart) and treated or not with deferiprone (DFP). (h) Locomotor dysfunction in mice, induced or not with CLP, intoxicated with MPTP (15 mg/kg, i.p., 3 injections 2 h apart) and treated or not with the combination of DFP and NAC. The results are expressed as mean ± SD (n = 10 mice per group). Statistical analysis was performed by applying the one-way ANOVA test.
Discussion
Aging is the main risk factor for the development and progression of idiopathic PD [49,50]. There is no animal model that fully resembles the pathology [51,52]. However, mice are widely used to mimic multiple aspects of human PD [53]. PD was induced by MPTP, a neurotoxin precursor capable of crossing the BBB due to its lipophilic nature. In the brain, MPTP is metabolized by glial cells into its active form, MPP+. This neurotoxic metabolite is highly toxic to DNs, causing mitochondrial dysfunction and cell death once taken up by dopaminergic transporters [54]. Its affinity for DNs leads MPP+ to reproduce the final phases of PD, i.e., DNs loss [55].
During aging, physiological changes sensitize the body to progressive tissue degeneration, which is often associated with organ failure and disease onset. In agreement, severe symptoms were observed when PD was induced in aged mice when compared to young animals. This effect occurred even upon the administration of sub-acute doses of MPTP, indicating that aging predisposes mice to develop a more pronounced phenotype. Impaired motor coordination was associated with DNs loss. Different factors contribute to neurodegeneration, among which is the activation of microglia. This is enhanced by the infiltration of peripheral immune cells, like Helper CD4 and Cytotoxic CD8 T cells, into the brain, through an impaired BBB. These events characterize age-related neuroinflammation and, as such, constitute a risk factor for PD development (Figure 1). No further increase in BBB permeability was observed upon PD induction, proving that this phenomenon physiologically occurs with advancing age. The contribution of a low-grade neuroinflammatory phenotype, developed during aging, to neuronal damage, was confirmed in vitro. Higher levels of pro-inflammatory cytokines were released when microglia were stimulated with sera collected from old mice. This effect was attributed to the ability of Fe to switch microglia phenotype towards a pro-inflammatory state upon their interaction with peripheral immune cells. Immune cells presented an increased level of intracellular Fe, as assessed by calcein quenching and FtH expression. Fe accumulation was not observed in the same immune cells isolated from young mice. Increased neuronal cytotoxicity was detected upon stimulation with the same sera, an effect prevented by prior exposure to deferiprone. When neurons were exposed to the supernatant collected from the aforementioned microglia, we observed an increase in neuronal loss (Figure 2). This indicated that the pro-inflammatory action of peripheral immune cells and microglia might damage neurons.
Higher levels of brain Fe are associated with aging and microglia dysfunction [26,56,57]. Senescent microglia become dystrophic, accumulate Fe [58] and increase ferritin expression. Dystrophic microglia were found in Lewy bodies in the substantia nigra, in post-mortem samples of PD patients [58,59]. Enhanced calcein quenching was observed in microglia isolated from old mice but not in those isolated from young animals. The ability of Fe to turn microglia pro-inflammatory was investigated in primary cells. In response to Fe, isolated microglia release pro-inflammatory cytokines, such as TNF. Neurons are also sensitive to Fe stimulation, as observed by the increased neuronal death measured upon treatment ( Figure 3). Hence, the disruption of Fe homeostasis in the brain strongly contributes to neuroinflammation and sensitizes disease onset, as previously reported [60]. Fe is highly abundant in the brain, where it is involved in many biological functions [32,33,41,42]. Its ability to undergo oxidation and reduction reactions, exchanging electrons with different substrates, turns it potentially toxic. By participating in the Fenton chemistry, Fe generates highly reactive oxygen species (ROS), which induce oxidative stress and tissue damage, as shown to occur also in the aging brain [60][61][62]. Our data found no increase in the accumulation of Fe in mice exposed or not to MPTP when comparing young vs. old animals ( Figure 3). This data confirmed that increased Fe levels occur throughout the lifetime [31,32,63] and are not detected prior to disease onset [31,32,[40][41][42]. Sequestered in peripheral immune cells, Fe circulates and enters the brain. The absence of adaptive immunity was shown to benefit PD, reducing its severity. When PD was induced by exposing TCRβ-deficient, Rag-2-deficient, or JHT-deficient mice, depleted of T and/or mature B cells, respectively, to high doses of MPTP, the resulting locomotor dysfunction was not as severe as in aged-matched wild-type animals. Deficient mice were more resistant to PD. This indicates that the immune system primes the brain to an increased disease severity (Figure 3), as recently demonstrated [31]. We showed that a sequence of events is necessary to induce neurodegeneration. To prove this statement, we exposed young mice to an infection-driven pro-inflammatory priming and assessed its contribution to the appearance of a neuroinflammatory phenotype. Our aim was to confirm that proinflammatory priming is a risk factor for PD, regardless of mice's age. During life, we are continuously exposed to different challenges, which cause inflammation and impact sensitive organs like the brain [35,64,65]. Upon infection, immune cells withhold Fe to restrict pathogen proliferation [36]. However, the accumulation of Fe within these cells, confirmed by calcein quenching and FtH expression, turns them pro-inflammatory [31,32]. We detected Fe-loaded immune cells in young mice exposed to mild CLP or sub-acute doses of LPS, which infiltrate the brain due to an increased BBB permeability. Consequently, microglia turn pro-inflammatory and enhance neuroinflammation (Figures 4 and 5). In vitro experiments were also conducted to assess the capacity of sera, collected from mice exposed to sub-acute doses of LPS, to activate microglia and induce the release of pro-inflammatory cytokines, like TNF. The supernatant of stimulated microglia was used to treat neurons, showing it causes neuronal death ( Figure 5). Similar data were observed by stimulating microglia or neurons with sera collected from CLP-induced mice, as assessed by the release of pro-inflammatory cytokines like TNF or DNs loss, respectively. This latter was inhibited by deferiprone, indicating the role of Fe in promoting neurodegeneration. To further demonstrate that the pro-inflammatory priming to the brain exacerbates PD severity, we exposed mice to CLP prior to MPTP intoxication and observed an increase in locomotor dysfunction. The protective effect of deferiprone against CLP-induced neuroinflammation further confirmed the involvement of Fe. The therapeutic administration of the Fe chelator administered a few days after CLP, prevented PD severity. Although Fe is necessary to elicit an immune response, its chelation is often required to prevent Fe-driven organ damage [66]. Since Fe is also a pro-oxidant, we assessed whether the combination of an antioxidant compound, NAC, and deferiprone protected against PD severity. The synergistic effect of this treatment confirmed that PD symptoms can be ameliorated by acting at a peripheral level. This notion is also in agreement with the administration routes of our therapy. Hence, our data clearly showed that, regardless of the mice's age, the pro-inflammatory priming to the brain sensitizes to neuroinflammation and Fe accumulation, both being key players in PD. The findings described in this study are consistent with the observation that certain infections induce neurological signs in patients, even if the brain is not directly affected [67][68][69]. All mice were maintained in cages (3-5 animals per cage), undisturbed, in an environmentally controlled room, in terms of temperature and humidity conditions, with a 12 h light/12 h dark cycle, and fed a standard diet and water at libitum. Animal care was taken to ensure that any mice exhibiting signs of suffering or distress were euthanized with CO 2 . This also included animals showing normal conditions of disruption and general ill-health signs, causing difficulty eating or drinking, moderate discomfort, pain or distress. Severe forelimb, hindlimb incoordination and locomotor disabilities constituted end-point criteria to sacrifice the animals used in our experiments. All symptoms were noted, registered, and discussed with the veterinarian of the animal facility.
Mice
The hypothermia caused by MPTP administration was prevented by using a heating pad, to which mice were allocated before the procedure. MPTP-induced mice were then transferred into pre-heated cages 2 h after the last injection. Animal cages were changed every 96 h after the procedure, and the mice's well-being was monitored until experiment day 30.
PD Induction and evaluation of locomotor dysfunction PD was pharmacologically induced, according to the protocol described in [45]. Locomotor dysfunction was assessed by placing mice upside-down on a vertical pole and measuring how many seconds they needed to descend [46]. A default time of performance of 120 s was set in case the severity of induced PD prevented mice from descending the pole.
Deferiprone treatment in mice Mice were injected with deferiprone (DFP, Sigma Aldrich, Ref. No. 379409; Darmstadt, Germany) at a dose of 10 mg/kg (1 injection per day, i.p.). Mice were injected for 15 days, starting 5 days after LPS or CLP injection.
LPS injection Ten µg of LPS were injected intraperitoneally (i.p.) once a day for 5 days to cause a mild inflammatory phenotype. No signs of stress were observed during or after the treatment, according to the health parameters measured daily (weight and temperature). After recovery, i.e., 3 weeks later, PD was induced, and disease progression was monitored for 30 days to then collect blood and organs for further analyses.
NAC injection NAC was dissolved in PBS, adjusted to pH 7.4, and administered to mice (15 mg/kg i.p.) starting 5 days after PD induction and every 12 h thereafter for 15 days.
Cecum ligation and puncture Mice were anesthetized with inhalable anesthetic 1-chloro-2,2,2-trifluoroethyl difluoromethyl ether (isofluorane) supplied at 500-1000 mL/min while maintained in an induction chamber; at 100-200 mL/min they were switched to nosecone to maintain that the animal anesthetized. The abdominal area of the animal was shaved, disinfected, and cleaned with 70% alcohol solution, followed by a betadine disinfectant solution. A small incision (1 cm) was made parallel to the midline. The cecum was exteriorized and ligated immediately distal to the ileocecal valve. In order to perform a low-grade CLP, the lumen was reduced by 50-60%, and the cecum was punctured once with a 23-gauge needle. Fecal content was extruded by applying pressure, and then the cecum was re-inserted into the abdominal cavity. The peritoneal wall was sutured with sterile 3-0 Dafilon sutures (Braun, Kronberg, Germany), and the skin was closed with a surgical staple (Autoclip 9 mm; Becton Dickinson, Franklin Lakes, NJ, USA). During the full procedure, the animal was kept in a temperature stabilizer (37 • C), the eyes were continuously moisturized to avoid blindness, and sterile isotonic fluids (1 mL 0.9% saline) were administrated subcutaneously (s.c.) at the end of the procedure to allow better recovery and fluid resuscitation. At the end of the surgery, the animal was allowed to recover on the temperature stabilizer (37 • C) for an extra 30 min, while antibiotics (Imipenem/Cilastine; Tienam; MSD; 0.5 mg/s.c./animal) were administered 2 h after the procedure and every 12 h during 72 h. Fully awake animals were transferred to standard cages and strictly monitored for health and welfare conditions (activity, behavior, temperature, and weight). The procedure caused little stress to the animal, and the outcome led to a sepsis condition, causing a mild inflammatory phenotype. No death was expected or observed. After recovery, i.e., 3 weeks later, PD was induced, and disease progression was monitored for 30 days to then collect blood and organs for further analyses.
Blood Brain Barrier permeability Mice were injected intravenously (i.v.) with 0.1 mL of 2% Evans Blue (Sigma-Aldrich, Ref. No 2129; Darmstadt, Germany), dissolved in saline solution and were killed 1 h later. Brain samples were harvested, weighed, placed in 2 ml of formamide and left for 48 h at 37 • C to extract the Evans Blue dye, as described in [70]. Absorbance was measured at λ = 620 nm (Bio Rad SmartSpec 3000). A standard curve with fixed concentrations of Evans blue was used to calculate dye extravasation into the brain. Data were expressed as mg of Evans Blue per g of brain tissue, as means ± standard deviation.
Isolation of brain-infiltrated immune cells Mice were sacrificed, perfused in toto (20 mL PBS) and brains were collected in HBSS, finely minced and digested in collagenase VIII (0.2 mg/mL, Sigma-Aldrich, Ref. No C2139; Darmstadt, Germany) for 30 min at 37 • C. Brains were homogenized by passing samples through a 100 µM strainer and collected in 20 mL HBSS. Next, samples were centrifuged for 10 min at 1500 rpm at 4 • C. Supernatants were discarded, and brains were washed in 20 mL HBSS and centrifuged again for 10 min at 1500 rpm at 4 • C. In order to separate the infiltrated leukocyte fraction, samples were resuspended in 10 mL of 30% Percoll gradient (GE Healthcare, Ref. No 17-0891-01; Darmstadt, Germany) and centrifuged for 20 min at 2500 rpm at room temperature, without break and acceleration. Myelin forming the upper layer was carefully removed, and brain-infiltrated leukocytes, precipitated at the bottom of the tube, were resuspended in 30 mL of PBS and centrifuged again for 10 min at 1500 rpm at 4 • C. The pellet obtained was then resuspended with 450 µL of PBS supplemented with 2% heat-inactivated FCS (Invitrogen) before being stained.
Flow cytometry staining The total number of infiltrated immune cells was measured by flow cytometry using a known concentration of reference 10µm latex beads suspension (Polysciences Europe GmbH, Ref. No CC10N-10; Hirschberg an der Bergstrasse, Germany), co-acquired with a pre-established volume of cellular suspension. Dead cells were excluded by Propidium Iodide. Singlets were gated among live cells based on size and granularity. Cells were stained with Fc block (anti-CD16/CD32; BD Pharmingen™-Ref. No 553141; Madrid, Spain) to prevent non-specific binding for 20 min at 4 • C. Cells were then washed in PBS supplemented with 2% heat-inactivated FCS and stained as follows. The presence of resident microglia (CD11b+CD45int) and CNS infiltrates (CD45hi), such as Helper and Cytotoxic T cells (TCRβ+CD4+ and TCRβ+CD8+), was determined. Their activation was assessed by the expression of surface markers, such as CCR2 and MHC II for microglia or CD44 and CD62L for T cells. Different cell populations infiltrating the brain were determined by using antibodies directly conjugated to PE, PE/Cy7, PercepCy5. 5 DNs isolation and culture DNs were isolated, following the protocol described in [71]. Briefly, mouse embryos were removed from the uterus of pregnant females at E12.5 and placed in ice-cold HBSS. The ventral midbrain was dissected at the microscope by separating the isthmus and mesencephalic-diencephalic boundary region with the aid of a curved scissor. The meninges were removed, and a cut in the mediodorsal midbrain was performed to flatten the tissue on a Petri dish. By using surgical blades, the wings of the butterfly shape, representing the substantia nigra, were dissected, and isolated regions were transferred into a falcon tube containing 15 mL of ice-cold HBSS. The samples were homogenized in 1 mL of pre-warmed solution of 0.05% trypsin-EDTA and incubated at 37 • C for 5 min. The enzyme activity of the trypsin-EDTA was stopped by adding 1 mL of DMEM/F12, supplemented with serum. The samples were centrifuged gently so as not to damage dissociated cells and washed in 1 mL of complete medium. This procedure was conducted twice, and with the aid of a fire-polished glass pipette, the tissue was triturated until obtaining a fine cell suspension. The samples were centrifuged at 400× g for 5 min at RT, and the medium was removed. Cells were then resuspended in 1 mL of complete medium, supplemented with N2. Once their viability was determined, through the Trypan Blue exclusion method, cells were seeded, on 24-well plates coated with Poly-L-ornithine/Laminin, in DMEM/F21, at a density of 150,000 cells per well. Plates were kept in a humidified tissue culture incubator and maintained at a 37 • C and 5% CO 2 . Experiments were performed 2 weeks after.
Microglia isolation and culture Microglia were isolated as follows.
Mouse pups, at post-natal days 0-3, were decapitated to expose the skull and to dissect brain hemispheres. Brain samples were transferred into a falcon tube containing 30 mL of cold HBSS. Meninges were removed, and the brains were dissected into small pieces. Those were placed in a falcon tube containing 5 mL of 0.05% trypsin-EDTA, to which 750 µL of DNase I (10 mg/mL) were added. The digestion was carried out by maintaining samples at 37 • C for 15 min. After trypsin inactivation, samples were centrifuged at 400× g, for 5 min, at 20 • C. Cells were resuspended in 5 mL of DMEM/F12 supplemented with 10% FBS and 1% penicillin/streptomycin (10,000 units; Invitrogen, ThermoFisher Scientific, Waltham, MA, USA). Microglia were then seeded in a T75 culture flask, previously coated with poly-l-lysine solution (0.1 mg/mL in sterile water), containing 10 mL of culture medium and maintained at 37 • C and 5% CO 2 . On day 2, the macrophage colony-stimulating factor (M-CSF; 100 ng/mL) was added to the medium, along with and granulocyte-macrophage colony-stimulating factor (GM-CSF; 100 ng/mL). Microglia were cultured for 3 weeks, and the medium was changed every week. At confluency, cells were detached with 0.05% trypsin-EDTA and seeded into 24 well plates for experiments.
Cytotoxicity Assay Cells were washed with PBS and exposed to the sera, collected from tested mice, i.e., aged animals or previously exposed to LPS or CLP, and pooled. Aliquots were prepared by diluting pools 1:10, and 50 µM of sera were added to the plate. All cells were covered with parafilm, allowing the sera to spread in the well and stimulate all the plates. In vitro, cytotoxicity was tested 24 h after. Cell death was also assessed in response to Fe treatment, provided as a ferric sulfate solution, Fe 2 (SO 4 ) 3 (100 µM), for 12 h. Cells were pre-treated with deferiprone (75 µM), 1 h before Fe 2 (SO 4 ) 3 and maintained thereafter. Control cells were washed with PBS and exposed to sera collected from control mice, i.e., young or non-treated animals. Viability was assessed by a crystal violet assay, as previously described [66,72,73].
Statistical analysis
Locomotor dysfunction or flow cytometry experiments were performed between 2 to 3 times, using between 2 and 4 mice per time. The results were pooled and expressed as mean ± standard deviation to assess statistical differences. The total number of mice per condition varied and is indicated in figure legends. In the graphs, each dot represents a mouse. ELISA and qRT-PCR experiments were conducted twice when the results reproduced the same trend. Representative graphs are shown and expressed as mean ± standard deviation. Statistically significant differences between the two groups were assessed using a two-tailed unpaired Mann-Whitney test or t-test, according to data distribution. Normal distributions were confirmed using the Kolmogorov-Smirnov test. Statistical differences between groups following a non-normal distribution were assessed by applying the Mann-Whitney. Comparisons between more than two groups were carried out by one-way ANOVA. No statistical method was used to predetermine the sample size. All statistical analyses were performed using GraphPad Prism 9 software. Differences were considered statistically significant at a p-value < 0.05. NS: Not significant.
Conclusions
In this study, we demonstrated that pro-inflammatory priming of the brain sensitizes the development of PD. The sub-chronic inflammation that occurs during aging or immune system activation in response to the infection might increase the risk of PD. Peripheral factors play a crucial role in symptom onset, and our study suggests the possibility that early therapeutic interventions can protect the brain from this multifactorial disorder. Our work might also highlight the reasons for the inefficacy of drug testing in PD trials, which include already diagnosed cohorts. A possible explanation could be that disease progression is influenced by multiple events. Neurodegeneration is the final step in a series of processes that originate in the periphery and subsequently affect the brain. Thus, new therapies should contemplate the fact that drugs targeting PD might not need to enter the brain, which so far is the limiting factor. Targeting disease etiology rather than clinical manifestations might protect patients from developing severe PD.
==
Domain: Psychology Biology
|
[{"added": "2023-04-30T15:22:38.027Z", "created": "2023-04-27T00:00:00.000", "id": "258405125", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.3390/ijms24097949", "pdf_hash": "28e6d2b2883d08e8f1b073988b0de55bfa71bd4c", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:33184", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "d4fd60ab59b79f805e2127e6a6786a69ca9ead07", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Sargassum horneri Extract Attenuates Depressive-like Behaviors in Mice Treated with Stress Hormone
Sargassum horneri, a brown seaweed, is known for its various health benefits; however, there are no reports on its effects on depression. This study aimed to investigate the antidepressant effects of S. horneri ethanol extract (SHE) in mice injected with corticosterone (CORT) and to elucidate the underlying molecular mechanisms. Behavioral tests were conducted, and corticotropin-releasing hormone (CRH), adrenocorticotropic hormone (ACTH), and CORT levels were measured. A fluorometric monoamine oxidase (MAO) enzyme inhibition assay was performed. Neurotransmitters like serotonin, dopamine, and norepinephrine levels were determined. Moreover, the ERK-CREB-BDNF signaling pathway in the prefrontal cortex and hippocampus was evaluated. Behavioral tests revealed that SHE has antidepressant effects by reducing immobility time and increasing time spent in open arms. Serum CRH, ACTH, and CORT levels decreased in the mice treated with SHE, as did the glucocorticoid-receptor expression in their brain tissues. SHE inhibited MAO-A and MAO-B activities. In addition, SHE increased levels of neurotransmitters. Furthermore, SHE activated the ERK-CREB-BDNF pathway in the prefrontal cortex and hippocampus. These findings suggest that SHE has antidepressant effects in CORT-injected mice, via the regulation of the hypothalamic-pituitary-adrenal axis and monoaminergic pathway, and through activation of the ERK-CREB-BDNF signaling pathway. Thus, our study suggests that SHE may act as a natural antidepressant.
Introduction
Depression is a common mental disorder characterized by feelings of profound sadness, deprivation of interest, and decreased motivation [1,2]. Its occurrence is increasing worldwide [3] and has more than doubled in numerous countries since the onset of the coronavirus disease-19 pandemic. The current treatments for depression include monoamine oxidase (MAO), serotonin reuptake and noradrenaline reuptake inhibitors, as well as tricyclic antidepressants [4]. However, these drugs often lead to side effects such as insomnia, anxiety, xerostomia, severe weight loss [5], bleeding, and cardiovascular and gastrointestinal complications; these side effects often create obstacles to alleviating and preventing depression [4,6]. Furthermore, synthetic medications may sometimes lead to maladaptive responses, and individuals with treatment-resistant depression face a higher risk of relapse [7]. On the other hand, natural functional materials tend to be cost-effective, with fewer side effects compared to traditional antidepressants, and they are well-received by individuals struggling with depression due to their high adaptability [8]. Therefore, there is
Sample Preparation
S. horneri collected from Jeju Island (South Korea) in May 2022 were washed and lyophilized. Dried S. horneri were ground using a blender and extracted with 70% ethanol/ water solution for 24 h at 25 • C. The extracts were subsequently filtered with filter paper (Whatman filter paper No. 2, Whatman, Maidstone, UK), concentrated by evaporation (Eyela, Tokyo, Japan), lyophilized, and stored at −70 • C until use. The yield of SHE was approximately 3.7 ± 0.1%.
HPLC Analysis
The samples were analyzed using the HITACHI CM-5000 HPLC system (Hitachi Seisakusho Co., Ltd., Tokyo, Japan) and equipped with a CM-5110 pump and auto-sampler. For separation, a Supelco Discovery ® HS C18 RP column (5.0 µm, 4.6 × 250 mm) was used. The separation was processed with a methanol-acetonitrile solvent (7:3, v/v) with a 30 min running time and an injection volume of 5 µL. The flow rate was 0.7 mL/min, and each sample was detected at 450 nm. Briefly, SHE was dissolved in the mobile phase and filtered with a 0.22 µm membrane filter, and then the filtered sample was subjected to HPLC analysis. The regression equation and correlation coefficient (R 2 ) of each standard curve were automatically determined using the HPLC system. The regression equation for fucoxanthin was y = 23927x + 6229.4 (R 2 , 0.9999). HPLC quantitative analysis was replicated four times. The concentration of fucoxanthin was 1.00 ± 0.02 (mean ± SD) mg/g extract, using the peak area in the standard chromatogram (Figure 1).
was replicated four times. The concentration of fucoxanthin was 1.00 ± 0.02 (mean ± SD) mg/g extract, using the peak area in the standard chromatogram (Figure 1).
Animals and Treatments
Three-week-old male ICR mice (24-30 g) were purchased from KOATECH Animal, Inc. (Pyeongtaek, Republic of Korea). Mice were provided with food and water under controlled temperature ( Mice were subjected to repeated intraperitoneal (i.p.) injections of CORT (40 mg/kg BW) for 3 weeks. The dosages were selected based on the relevant literature [24][25][26][27][28]. In this study, LT was used as the positive control. After 3 weeks, behavioral tests were performed,
Animals and Treatments
Three-week-old male ICR mice (24-30 g) were purchased from KOATECH Animal, Inc. (Pyeongtaek, Republic of Korea). Mice were provided with food and water under controlled temperature (21 ± 2 • C), humidity (55 ± 5%), and a 12 h light-dark cycle. All animal experiments were approved by the Institutional Animal Care and Use Committee of the Korea Food Research Institute (IACUC number, KFRI-M-22005), and all experiments were performed in accordance with the Arrive guidelines 2.0.
After a 1-week acclimatization period, the mice were divided into four groups (n = 10 per group): (1) SHAM group (injected with vehicle (VEH) + VEH, per oral (p.o.) administration), (2) CORT + VEH group (injected with CORT + VEH (p.o.)), (3) CORT + LT group (injected with CORT + L-theanine (LT) for positive control group at 50 mg/kg BW (p.o.)), and (4) CORT + SHE group (injected with CORT + SHE 500 mg/kg BW (p.o.)). Mice were subjected to repeated intraperitoneal (i.p.) injections of CORT (40 mg/kg BW) for 3 weeks. The dosages were selected based on the relevant literature [24][25][26][27][28]. In this study, LT was used as the positive control. After 3 weeks, behavioral tests were performed, and the mice were sacrificed for analysis. Brain tissues were isolated and stored at −80 • C. The experimental scheme is shown in Figure 2A.and the mice were sacrificed for analysis. Brain tissues were isolated and stored at −80 The experimental scheme is shown in Figure 2A. The tail-suspension test (TST) was conducted following established protocols [29]. Animals were suspended 15 cm above the table using an adhesive tape that was placed approximately 1 cm from the tip of the tail. The immobility time was recorded for 6 min using the TST apparatus (BioSeb, Chaville, France).
Forced-Swimming Test
The forced-swimming test (FST) was conducted following established protocols [30]. The mice were placed in the cylinders (height: 50 cm, diameter: 20 cm) and swam for 6 min. The immobility time was analyzed during the last 4 min of the test, excluding the initial 2 min. During 6 min, the immobility time was recorded and analyzed using SMART 3.0 software (Panlab SL, Barcelona, Spain).
Elevated-Plus-Maze
The elevated-plus-maze (EPM) test was conducted in accordance with established procedures [31]. The EPM apparatus was elevated 60 cm from the floor, with two open arms (30 cm × 5 cm × 0.5 cm) and two closed arms (30 cm × 5 cm × 16 cm) connected by a central platform (5 cm × 5 cm). Each mouse was located on the center platform and allowed to acclimate for 1 min before commencing the test. Mice were positioned at the center of the maze and allowed to move freely for 5 min. Behaviors were recorded and analyzed using SMART 3.0 software (Panlab SL).
Measurement of Serum CRH, ACTH, and CORT Levels
Serum CRH, ACTH, and CORT levels were measured using commercial ELISA kits following the manufacturers' instructions (CRH ELISA kit, MyBioSource, San Diego, CA, USA; ACTH and Corticosterone ELISA kits, Enzo Life Sciences, Farmingdale, NY, USA). Absorbance was measured at 450 nm using a microplate spectrophotometer (Jasco, Tokyo, Japan). All levels were calculated from a standard curve fitted to serial standards supplied by the manufacturers.
Measurement of MAO Activity
The fluorometric MAO enzyme inhibition assay was performed using a commercial kit (Thermo Fisher Scientific, Waltham, MA, USA). Briefly, human recombinant MAO-A or MAO-B enzymes (0.15 U/mL, Sigma-Aldrich, St. Louis, MO, USA), substrate (p-tyramine or benzylamine, 1 mM), horseradish peroxidase (1 U/mL), Amplex red reagent (200 µM) were mixed. SHE (1-200 µg/mL) was added to the mixture, and the mixture was incubated at 25 • C for 60 min while protected from light. Subsequently, fluorescence intensity was detected using a fluorescence microplate reader (Molecular Device, Sunnyvale, CA, USA) at excitation and emission wavelengths of 550 and 590 nm, respectively.
Measurement of Neurotransmitter Levels
Brain 5-HT, dopamine, and norepinephrine levels were determined following the manufacturers' instructions (5-HT and dopamine ELISA kits, Abcam, Cambridge, MA, USA; norepinephrine ELISA kit, Feiyuebio, Wuhan, China). The sample absorbance at 450 nm was determined using a microplate spectrophotometer (Jasco).
Statistical Analysis
The data are expressed as the mean ± the standard error of the mean (SEM) and were analyzed using one-way analysis of variance (ANOVA), followed by Dunnett testing using Prism 9 (GraphPad Software, Inc., San Diego, CA, USA). Statistical significance was set at p < 0.05.
SHE Improves Depressive-like Behaviors Caused by CORT Injection
Behavioral tests were conducted to analyze the antidepressant-like effects of SHE in mice induced by CORT injection (Figure 2B-D). In the TST and FST, the immobility time was significantly longer in the CORT + VEH group than in the SHAM group (TST: p < 0.05; FST: p < 0.001). However, these alterations were reversed by SHE administration to CORT-injected mice. In the EPM test, the time spent in open arms was reduced by approximately 3.3-fold in the CORT+VEH group, compared to that in the SHAM group (p < 0.01). However, SHE administration increased the duration of time spent in open arms (p < 0.01). Conversely, the time spent in closed arms was significantly increased by CORT injection in mice, but these effects were diminished by SHE administration, similar to that in the CORT + LT group. Our results indicated that SHE attenuated depressive-like behaviors in mice induced by CORT injection.
SHE Improves the Abnormal HPA Axis in Depressive Mice
To investigate whether SHE modulates HPA axis dysfunction in mice induced by CORT injection, we measured serum CRH, ACTH, and CORT levels. CRH and CORT levels in the serum of CORT-injected mice were significantly higher than those in the SHAM group (p < 0.001); however, SHE administration decreased these hormone levels to levels similar to those in the CORT + LT group. Although the ACTH levels increased in the CORT + VEH group (p < 0.05), significant changes were not observed following SHE administration (Figure 3A).
GR regulates HPA axis activity during stress response [32]. Therefore, we measured GR protein expression in the prefrontal cortex and hippocampus. GR expression was higher in the CORT + VEH group than in the SHAM group (prefrontal cortex: 1.8-fold, p < 0.001; hippocampus: 3.2-fold, p < 0.05); however, this increase was mitigated by SHE administration (prefrontal cortex: 1.6-fold, p < 0.01; hippocampus: 2.9-fold, p < 0.05). Our findings suggest that SHE improves the HPA axis dysfunction in mice induced by CORT injection (Figure 3B).
SHE Prevents the Abnormal Monoaminergic System in Depressive Mice
To assess whether SHE regulates the monoaminergic system in the CORT-injected depressive mice, we measured both MAO activity and its protein expression. First, we performed MAO activity assay in in vitro. The results showed that the MAO-A and MAO-B activities decreased by 1.7-fold and 2.0-fold at a concentration of 200 µg/mL SHE (MAO-A: 40.46% reduction; MAO-B: 49.58% reduction; Figure 4A).
Based on these results, we evaluated the levels of neurotransmitters, specifically 5-HT, dopamine, and norepinephrine, in the brains of CORT-injected mice. Although not significant statistically, a trend was identified whereby neurotransmitter levels decreased further in the CORT + VEH group than in the SHAM group. However, SHE administration led to increased levels of the neurotransmitters 5-HT, dopamine, and norepinephrine (Table 1).
SHE Prevents the Abnormal Monoaminergic System in Depressive Mice
To assess whether SHE regulates the monoaminergic system in the CORT-injected depressive mice, we measured both MAO activity and its protein expression. First, we performed MAO activity assay in in vitro. The results showed that the MAO-A and MAO-B activities decreased by 1.7-fold and 2.0-fold at a concentration of 200 µg/mL SHE (MAO-A: 40.46% reduction; MAO-B: 49.58% reduction; Figure 4A).
Based on these results, we evaluated the levels of neurotransmitters, specifically 5-HT, dopamine, and norepinephrine, in the brains of CORT-injected mice. Although not significant statistically, a trend was identified whereby neurotransmitter levels decreased further in the CORT + VEH group than in the SHAM group. However, SHE administration led to increased levels of the neurotransmitters 5-HT, dopamine, and norepinephrine (Table 1).
Furthermore, in the prefrontal cortex and hippocampus tissues of CORT-injected mice, MAO-A and MAO-B protein expressions were significantly higher in the CORT + VEH group Furthermore, in the prefrontal cortex and hippocampus tissues of CORT-injected mice, MAO-A and MAO-B protein expressions were significantly higher in the CORT + VEH group than in the SHAM group. However, SHE administration reduced the expression of MAO-A and MAO-B. These results suggest that SHE prevents a decrease in neurotransmitter levels in CORT-injected mice by modulating MAO expression (Figure 4B,C).
SHE Activates ERK-CREB-BDNF Signaling in the Prefrontal Cortex and Hippocampus of CORT-Injected Mice
Recently, many studies have implicated the ERK-CREB-BDNF signaling pathway in depression pathology [14]. Hence, we evaluated the ERK-CREB-BDNF signaling pathway in the prefrontal cortex and hippocampus using immunoblotting. In the CORT + VEH group, the phosphorylation of ERK and CREB and the expression of BDNF were significantly lower than those in the SHAM group. However, SHE administration significantly increased the phosphorylation of ERK and CREB, and the expression of BDNF in CORTinjected mice. These results suggest that SHE activates the ERK-CREB-BDNF signaling pathway in the prefrontal cortex and hippocampus of CORT-injected mice (Figure 5).
Discussion
In the present study, we demonstrated the antidepressant effects of SHE on CO induced depressive mice. CORT, an adrenal hormone associated with stress, lead
Discussion
In the present study, we demonstrated the antidepressant effects of SHE on CORTinduced depressive mice. CORT, an adrenal hormone associated with stress, leads to sustained depressive behaviors such as low motivation or anhedonia [33]. The animal model of depression utilized in this study, induced by CORT injection, is widely used to verify the effectiveness of antidepressant drugs and phytochemicals [33,34]. The chronic CORT-treated rodent model is suitable for depression, as it is often used to assess mechanisms involved in depression. Consistent with previous research, our study revealed that mice injected with CORT exhibited depressive behavior distinguished as increased immobility time and anxiety. However, SHE administration significantly improved depressive behaviors following CORT injection. Yegdaneh et al. [35] reported that brown algae reduced immobility time in the FST and elevated appetite in a novelty-suppressed feeding test in a mouse model of depression induced by Bacillus Calmette-Guérin. Kim et al. [13] reported that an ethanolic extract of Ishige foliacea alleviated behavioral patterns in the FST and TST compared with a depression group. The hexane extract of Sargassum plagyophylum alleviates depressive-like behaviors by reducing immobility time in the FST [36]. Similar to previous research, SHE reduced the immobility time in the TST and FST and increased the time spent in the open arms in EPM. Our results showed that SHE mitigates depressive-like behavior in a CORT-induced depressed-mouse model.
Under stress, the HPA axis is activated, resulting in the release of CORT [37]. CORT binds to the GR and subsequently translocates to the nucleus, activating glucocorticoidresponse elements to regulate the negative feedback process of the HPA axis [38,39]. However, normal regulation of the HPA axis is disrupted in depression, resulting in abnormal glucocorticoid overexpression, which further contributes to GR dysfunction [40]. Similarly, animal models of depression induced by CORT or restraint stress show high CRH and GR expression in chronically stressful situations [41,42]. Consistent with other studies, our results revealed that the serum's CRH, ACTH, and CORT levels significantly increased in CORT-injected mice. In addition, the expression of GR in the brain tissues was noteworthy and increased in the CORT + VEH group. Drugs such as fluoxetine, ketamine, and xiaoyaosan reduce serum's CORT concentrations and GR expression in animal models of chronic depression [43,44]. Similar to these antidepressants, SHE reduced GR expression and the concentrations of CORT and CRH, which were increased by CORT induction. Remarkably, unlike the concentrations of CRH and CORT, the ACTH concentration did not change after SHE administration. Previous studies measuring ACTH levels found no significant difference to the stress model at low concentrations of Camellia euphlebia and Lilium davidii extracts; however, a significant decrease was observed at high concentrations [45,46]. This suggests it is a dose-dependent response. We surmise that the singular SHE concentration in our experiments may not have reached the threshold to evoke discernible changes in ACTH levels. Given these findings, further experiments focusing on dose-response relationships are required. Nevertheless, because treatment with SHE reduced the levels of CRH and CORT, which were increased by CORT treatment, our results suggest that SHE may help alleviate HPA-axis dysfunction.
The absence of pleasant emotions and positive-affective states is a hallmark of depression, and these emotions are regulated by monoamine neurotransmitters [47]. Monoamine neurotransmitters, such as 5-HT, dopamine, and norepinephrine, are regulated by the MAO enzyme, which breaks them down. In mental disorders like depression and attentiondeficit hyperactivity disorder, the monoaminergic system undergoes disruption through the activation of MAO, and subsequently resulting in accelerated degradation of neurotransmitters and impaired synaptic transmission [48][49][50]. Numerous antidepressants have been developed and used to inhibit MAO activity and prevent neurotransmitter degradation [51]. Poltyrev et al. [52] reported that antidepressants, such as ladostigil, inhibit MAO-A and MAO-B by more than 60% and significantly reduce stress-induced depressive-like behaviors. Sargassum macrocarpum ethanolic extract and sargachromanol, which is isolated from Sargassum siliquastrum, inhibit MAO-A and MAO-B activities [53,54]. Similar to other Sargassum species, SHE also inhibited MAO activity. MAO expression in the CORT-injected mouse brain tissues was significantly reduced by SHE. In a mouse model of depression, 5-HT concentrations reduced by CORT administration were increased by the antidepressant fluoxetine [55]. In our study, the administration of SHE led to a significant increase in neurotransmitter levels than in the CORT + VEH group. This effect was similar to that of antidepressants, where depressed mice showed a downward trend compared to the SHAM group. These results suggest that SHE may elicit the inhibitory effects of the MAO enzyme and prevent neurotransmitter degradation.
BDNF is a brain-neurotrophic factor that maintains neuronal function by promoting neuronal survival and differentiation [56]. BDNF binds to tropomyosin receptor kinase B and contributes to the activation of the ERK-CREB pathway through the phosphorylation of ERK and CREB. Upon activation, the ERK-CREB pathway is involved in signal transduction, cell growth, and cell death, contributing to the improvement of depressive symptoms [14]. Previous studies have shown reduced expression of p-ERK, p-CREB, and BDNF in the hippocampus of animal models of chronic unpredictable mild stress (CUMS) [57,58]. The antidepressant activity of fluoxetine was observed by stimulating the ERK-CREB-BDNF pathway in the CUMS mouse model [59]. Similar to antidepressants, SHE significantly increased p-ERK, p-CREB, and BDNF protein expression in the prefrontal cortex and hippocampus of the CORT-induced mouse model. As these results were similar to those observed in other brown algae [13], our results indicate that SHE plays a role in activating the ERK-CREB-BDNF pathway, thus contributing to the antidepressant effects in CORTinduced mice.
Fucoxanthin is a natural carotenoid found abundantly in edible brown seaweeds [60]. It is known to possess physiological and biological characteristics, such as antioxidative, anti-obesity, anticancer, antihypertensive, anti-inflammatory, antidiabetic, and neuroprotective properties [61,62]. A recent study by Pangestutie et al. [63], highlighted fucoxanthine's potential to reverse oxidative stress and inflammation triggered by β-amyloid in BV2 microglia cells. This reversal was evidenced in the downregulated expressions of proinflammatory cytokines and a decrease in reactive-oxygen species formation. Furthermore, Jiang and colleagues [64] observed the efficacy of fucoxanthin in reducing the heightened immobility time of FST and TST after LPS treatment in mice. Specifically, a dosage of 200 mg/kg of fucoxanthin mitigated the overexpression of pro-inflammatory cytokines (IL-1β, IL-6, and TNF-α) and enzymes (iNOS and COX-2) in critical brain regions, implicating the modulation of the AMPK-NF-κB signaling pathway in its mechanism of action. In our studies, using HPLC analysis, we determined that the fucoxanthin content in SHE was 1.00 ± 0.02 mg/g extract. Given these findings, it is plausible to conclude that the fucoxanthin in the SHE likely contributed to its antidepressant effects.
However, this study had certain limitations. Firstly, because our experiments were conducted with a single concentration of SHE, it is necessary to verify the antidepressant effects of SHE at different concentrations in a CORT-injected mouse model. Secondly, although we observed the antidepressant efficacy of SHE in a CORT-induced depressive-mouse model, it is necessary to investigate other depressive models, such as social defeat stressor chronic mild stress-induced depressive models. Thirdly, our study suggests that SHE has an antidepressant effect. Therefore, it is crucial to investigate and identify other active compounds in SHE that contribute to the antidepressant effect, even though fucoxanthin has already been identified. Finally, clinical trials are required to fully demonstrate the potential of SHE as a treatment for depression. Although the present research has focused on studying and analyzing the mechanisms involved in alleviating depressive-like behavior and the antidepressant effects of SHE in vitro and ex vivo, human clinical trials are required to determine the safety and effectiveness of the treatment.
Conclusions
In the present study, our findings demonstrate that SHE exhibits antidepressant-like effects in CORT-induced depressive mice by regulating CRH and CORT levels. The HPA axis normalized by SHE reduces GR overexpression. In addition, SHE restores the monoaminergic system by inhibiting MAO activity, which degrades neurotransmitters. Furthermore, the ERK-CREB-BDNF signaling pathways are phosphorylated and upregulated by SHE.
Table 1 .
Effects of Sargassum horneri ethanol extract (SHE) on the neurotransmitter levels in the brain.
==
Domain: Psychology Biology
|
[{"added": "2023-10-14T15:59:37.946Z", "created": "2023-10-01T00:00:00.000", "id": "264074167", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "CLOSED", "oa_url": "https://www.mdpi.com/2076-3921/12/10/1841/pdf?version=1696940393", "pdf_hash": "8ec7ee71f74500ed47b3a427b7e6720be0550932", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:34443", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "eece5d24f3a97b2ea436ad5287c9510dad31d0a9", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
evALLution: making basic evolution concepts accessible to people with visual impairment through a multisensory tree of life
People with visual impairment have benefitted from recent developments of assistive technology that aim to decrease socio-economic inequality. However, access to post-secondary education is still extremelly challenging, especially for scientific areas. The under representation of people with visual impairment in the evolution research community is connected with the vision-based communication of evolutionary biology knowledge and the accompanying lack of multisensory alternatives for learning. Here, we describe the development of an inclusive outreach activity based on a multisensory phylogeny representing 20 taxonomic groups. We provide a tool kit of materials and ideas that allow both the replication of this activity and the adaptation of others, to include people with visual impairment. Furthermore, we provide activity evaluation data, a discussion of the lessons learned and an inclusive description of all figures and visual data presented. The presented baseline data show that people with visual impairment indeed have lack of access to education but are interested in and apt to understand evolutionary biology concepts and predict evolutionary change when education is inclusive. We show that, with creative investment, basic evolutionary knowledge is perfectly possible to be transmitted through multisensory activities, which everyone can benefit from. Ultimately, we hope this case study will provide a baseline for future initiatives and a more inclusive outreach community.
a physical barrier to the individual's learning experience due to the lack of multisensory alternatives to widespread learning activities (Salleh and Zainal 2010). Such lack of accessibility to knowledge is highly present in the fields of science, technology, engineering and mathematics (STEM) where instruction relies heavily on graphically conveyed information.
Efforts to develop assistive technology and promote the inclusive education of people with visual impairment are significant in some scientific areas (Cryer 2013), like chemistry (e.g. Fantin et al. 2016;Garrido-Escudero 2013;Supalo et al. 2008;Supalo and Kennedy 2014) and physics (e.g. Arcand et al. 2019;Ediyanto and Kawai 2019). However, the biological sciences seem to lag behind (but check Jones et al. 2006). More specifically, the teaching of evolution relies on visual media as its primary communication mechanism for conceptual understanding. Classical evolution case studies commonly used in formal education and outreach activities-such as beak morphology evolution in Darwin's finches (Grant and Grant 2002) or the industrial melanic peppered-moth selection (Cook et al. 2012)-are based on phenotype-environment associations and selective forces that demand a priori understanding of variability in shape, size and colour traits. All of these are visual characteristics of information, which are hard to grasp by people with reduced vision and inaccessible to people born blind.
The understanding of natural patterns is further compromised by the fact that the research of such case studies is communicated through 2-dimensional tables, plots and diagrams, available solely on screen or paper, all of which are particularly challenging to access for those with severe vision impairment (Karshmer and Bledsoe 2002;McCarthy and Shevlin 2017). Consequently, although teenagers with visual impairment show high interest in STEM areas, their motivation to pursue a carrier in such areas is reduced by the barriers felt while trying to learn (Bell and Silverman 2018). In fact, only one percent of STEM doctorate recipients has any sort of reported disability (data for the U. S. population; (Moon et al. 2012).
It is thus clear that there is an urgent need for improved accessibility to scientific knowledge in order to promote equity in education and a more diverse and inclusive scientific community. Outreach activities that transmit knowledge with a multisensory approach can be an important first step towards that end (Pérez-Montero 2019) and are known to benefit both the audience and the scientific community (Clark et al. 2016).
Our project aims at contributing to equity in accessibility to evolutionary biology knowledge by eliminating physical barriers to the understanding of the basic mechanisms of evolution and the resulting biodiversity pattern.
We here describe the development of a multisensory phylogeny, designed as an introduction to basic concepts in evolution for people with severe visual impairments.
We provide a tool kit that enables the repeatability of this activity together with guidelines that can be adapted and applied to several other outreach initiatives. We propose a two-step rationale to approach inclusive evolution teaching: The public needs to first (I) experience the pattern of biodiversity so that then we can (II) discuss the processes that led to such diversification. In addition to activity design, baseline data on the evaluation of the activity are presented.
Reproducibility framework
Touch as the main sense of communication raises challenges: natural history collections are usually too unique or fragile to be freely manipulated, live specimens pose animal welfare concerns, and commercially available models can be inaccurate and do not portray the real and detailed textures and patterns of biodiversity.
Our activity was developed during a whole year, which encompassed a great deal of communication with the blind community, psychologists, science communicators, museologists, evolutionary biologists and pedagogic institutions.
Based on the experience acquired from that process and the results of our activity, we here provide a theoretical framework to organize similar activities (Fig. 1).
There are three main steps to conveying evolution's patterns and processes to a public with visual disability: (1) make biodiversity accessible, (2) give an evolutionary and ecological context to the displayed biodiversity and (3) discuss the processes and evolutionary forces through which that biodiversity evolved.
1. First, we want to convey biodiversity as the resulting pattern of evolution. Sighted people can easily grasp the diversity of living beings and ecosystems through images; in order to include people with blindness we have to represent as much biodiversity as possible in an inclusive way. For this, communication with the blind community and creativity are essential. Sighted people tend to use auditory explanatory cues to convey information. However, this does not only create an immediate barrier for people with hearing disabilities, but our consultant from the blind community was quick to explain that touch had to be the main sense used. He made it clear to us with the following example: "If you tell me that a zebra is a horse with black and white stripes, and I was born blind, I will most likely not know what horses look like; stripes and colors might also be concepts that I do not understand". Thus, all collected materials should have the potential to be touched and then accompanied with auditory instruction by the teaching volunteers. To include people with hearing loss, the instruction should also be translated into sign language.
Collections such as Mollusks' shells and fossils are easily obtainable in great number and diversity-it was one of the resources more museums were willing to lend. The shape diversity within the phylum is accessible through touch, but the diversity of color patterns needs to be translated to haptic cues. When posed with such challenges we recurrently used hot glue to trace the pattern we wanted to be accessible. Hot glue is a great resource to turn 2D patterns into 3D (see butterfly wing models in Additional file 1: Fig. S1f; and mollusks table phylogeny in Fig. 1). Another easily accessible taxon is plants. Gardening centers in general have a great diversity of worldwide plants that can be easily sourced. We found that, despite it being the most familiar, participants responded very well to the plant branch, spending a lot of time there, and were excited by learning about plants' adaptations to specific environmental conditions.
Fig. 1
Basic framework to construct evolution outreach activities inclusive for people with blindness. Two types of materials are depicted: a collection of general easy access (Mollusk shells) and a classic evolution textbook example of adaptive speciation (the adaptive radiation of cichlids in the Tanganyika lake). Following this framework, all gathered materials are able to be adapted for inclusive outreach activities, independently of amount of branches represented and logistics. The pictures of the mollusk phylogeny on the round table are from a second activity, at an inclusive school, where the available room was significantly smaller than the original 125 m 2 space, thus more activities per floor-phylogeny branch (8 instead of 21) were included 2. Then, to approach biodiversity and evolution, the collected materials should be displayed in evolutionary context. For this, the specimens representative of the main branches can be displayed in several informative ways: (a) following the phylogeny of the species or orders depicted (e.g., mollusks organized as Monoplacophora, Bivalvia and Gastropoda; Fig. 1, left); (b) according to niche within an environment (e.g., Lake Tanganyika cichlids were arranged according to the depth (height) and substrate (sand, stones) they inhabit in the lake; Fig. 1, right); or (c) across different environments (e.g., plants were organized by climate regime including desert, tropical forest, Mediterranean forest and Taiga (Additional file 1: Fig. S1 t-w). These meaningful displays, together with oral pedagogic information, then allow us to become mindful of important concepts such as speciation, shared characters and evolutionary novelties, which in turn inform our understanding of phylogenetic patterns and common ancestry. Specific adaptations such as mouth position between benthic and limnetic fish, leaf shape and texture depending on climate are also a great way to introduce adaptation and natural selection.
In sum, any material gathered is capable of being used in a meaningful way, whether a classic textbook example or commonly found specimens. The activity is thus completely adaptable to available material and space, since all phylogenies can be simplified to have more or less branches, and activities can be designed for the participants to spend more or less time in contact with each branch (Fig. 1).
Multisensory tree of life toolbox
We provide a list of all materials used to represent taxa across the 20 branches (Additional file 2: Table S1) specifying which ones belonged to pedagogic collections from research institutes and education institutions such as museums or aquaria. Photographs of all branches and material display are also available (Additional file 1: Figure S1). All printable 3D models developed from scratch, by scanning real specimens, are available in their final form at MorphoSource.
The tree topology was based on reference phylogenies comprising the taxa of interest (Field et al. 2014;Hedges et al. 2015;Dos Reis et al. 2015) and on the interactive phylogeny OneZoom Tree of Life Explorer (Rosindell et al. n. d).
At each branch a volunteer educator provided information to the participants while assisting them in the exploration of the branch-specific materials. Prior to the activity, the educators were provided scripts containing information on what the branch-specific material illustrates, how to guide the people with visual disability to touch the materials and, for the branches where data collection was conducted, the branch-specific activity questions (see Additional file 4: Branch exercises). The scripted branch-specific questions that were given to participants for data collection on the predictability of evolution and basic evolution concepts-like adaptation and natural selection are provided (see methods sections below; Additional file 4: Branch exercises).
To allow for a general perception of the room display and guide the participants with visual impairment independently through the exhibition room, we designed individual haptic hand-maps. These consisted of a blueprint of the room drawn in hot glue on a thin wood plate with 3D information on the phylogenetic path on the floor together with the blueprint of the table display. While this resource was not useful in our implementation, since all participants with visual impairment preferred to be guided trough the activity by a staff member or by their accompanying sighted person; we think that in other contexts, such as museum exhibitions, this might be an inclusive resource that allows the visitor to independently explore the space.
Aspects important to take into account when building a multisensory phylogeny are the fact that partial specimens (like teeth or fur) should always be accompanied by a full model of the organism to make sure that people with visual disability can locate the specific material and make sense of it. Such models, like detailed toy animals, provide a general sense of scale that can help perceive certain biodiversity patterns-e.g. a lynx is bigger than a house cat, as are its teeth, skull and footprints despite the shapes of those structures being extremely similar.
Logistics of the room are also very important. In our case, there was not enough space for echinoderms and amphibians to reach the periphery of the phylogeny. These two branches were thus shorter ( Fig. 2) which is not optimal as it might inadvertently convey that these taxa are not extant species or that they are somehow 'less evolved' than those at the other branch tips. Another aspect to take into account is that, by placing mammals and especially hominids towards the exit and at the top of the room, we might involuntarily contribute to the wrong notion that evolution is a linear process towards humanization so often reinforced by images. However, it's worth noticing that, in our case, different participants followed completely different routes along the phylogeny, as the movement was mainly dictated by the available educators at the time of branch visit change.
Prior to the MSToL (multisensory tree of life) activity, the volunteer educators received information from the in-house psychologist at the Portuguese Association for the Teaching of the Blind (APEC-Associação Promotora do Ensino dos Cegos) on effective communication with people with visual impairment and the basics of assistance in orienting people with visual impairment.
Participant data collection
In order to evaluate our activity, data were collected from 25 participants with visual impairment (15 women and 10 men) and 23 sighted participants (17 women and 6 men; Additional file 3: Table S2). All data were collected in loco at the Portuguese Association for the Teaching of the Blind (APEC) on the day of the outreach activity (12 of March, 2019), before and after participation, following the questionnaire provided (Additional file 5: Questionnaire).
The questionnaire included personal data, an exercise of true or false and a word association exercise. Both exercises were scored to allow us to assess and compare the participants' knowledge on basic concepts of evolution, before (basal knowledge) and after the activity.
We recorded, for each participant, demographic dataage, gender-and also their education level, interest in biology/evolution and visual capacity. Education level of the participants was coded according to the Portuguese education system as follows: Primary School (PS) 1st to 4th grades; Middle School (MS) 5th to 9th grades; High School (HS) 10th to 12th grades; bachelor's degree (B) and master's degree (M). B and M are referred to as 'post-secondary education' throughout the manuscript. The level of visual impairment of the participants was assessed by asking each individual which one of the following 4 levels they identified with: 'No disability' , which includes people with glasses correcting for standard vision levels; 'Moderate vision loss', which refers to people with low vision with perception of shapes and colors (includes people with corrective lenses); 'Deep vision loss with residual light perception' , which refers to people with extremely low vision but that can still perceive some light variation; and 'Profound vision loss without any light perception' , which refers to people can not receive any visual cues.
Throughout the manuscript, when we refer to 'people/ participants with visual impairment' we are referring to the whole spectrum of visual impairment, specifically defined above for our panel of participants. When we refer to 'people/participants who are blind' we refer to people who identify with the fourth group describedprofound vision loss without any visual cues. We additionally recorded how long the person has lived with visual impairment.
Data were also collected, through an online questionnaire, from 15 out of 24 volunteer educators to assess Table S1 for a complete list of materials and Additional file 2: Fig. S1 for detailed branch photos) their emotional response to the activity and understand if the inclusive activity was mutually beneficial: for participants and educators.
The resulting data allowed us to (1) generally evaluate the effectiveness of the designed haptic activities, (2) establish baseline data about evolution knowledge and interest within a subset of the Portuguese community with visual impairment, (3) determine whether participants enjoyed the activity, and (4) begin to determine how such activities might improve the learning of evolution.
Data collection was approved by the University of Bath, Department of Psychology, Research Ethics Committee (code 17-273). All participants provided informed consent prior to participating, participated voluntarily and were informed of their right to withdraw participation at any point during data collection.
All data collected is presented anonymously in Additional file 3: Table S2.
Branch-specific exercises
In nine of the 21 branches, participants were asked branch-specific questions (Additional file 4: Branch exercises). These questions related to the haptic materials available on the table and were primarily exercises of prediction of phenotypic change, or adaptation scenarios about environmental change. In total, there were 29 questions focused on organisms' evolutionary responses to certain environmental changes, designed to assess the participants' understanding of adaptation, fitness, environment-phenotype associations, gradualism and natural selection, and their predictive ability of evolutionary change. Participants were not given possible answer options. They were asked the scripted questions (Additional file 4: Branch exercises) and the volunteer educator scored their answers in the record sheet according to the level of evolutionary thought: Answers in the 'maybe'/'I don't know' category were scored as 0; if the answer was not the known outcome of the evolutionary process but involved plausible evolutionary outcomes (extinction, mutation) it was scored as 1; if the answer took into account natural selection and adaptation it was scored as 2, with one extra point for the four questions in which gradualism or selection strength was considered. Finally, if the answer was in the 'nothing changes' category and did not consider any evolution outcome, it was scored as − 1.
Since not all participants answered all branch-specific questions, the standardized branch-score was calculated for 24 participants with visual impairment and 17 sighted participants as the sum of individual answer scores divided by the number of questions answered by the participant.
The 'prediction score' regards the subset of 18 questions, spanning 6 branches (three questions on the plant branch, three on corals, four on bony fish, three on molluscs, one on hominids and four on turtles) that require the participant to predict the outcome of an environmental change and was calculated and standardized as described above, for 23 participants with visual impairment and 17 sighted participants.
Word association exercise
We wanted to know if people were familiar with the scientific terms necessary to understand the basics of the theory of evolution and to what extent terms that are essential or might promote its misunderstanding were commonly associated with the concept of 'evolution' . For this we designed a word association exercise where participants were read a list of 33 words, one at a time, and upon hearing each one reported whether a given word was instinctively associated with evolution by responding 'True' or 'False' (Refer to Fig. 4 for the complete list of words and Additional file 5: Questionnaire). Words were scored as -1 if the word is usually associated with misunderstanding of the evolutionary process (e.g. perfecting); as 0 if the word is neutral and unnecessary to explain the theory of evolution (e.g. science); as 1 if the word is not necessary to explain the theory of evolution but it is related to it (e.g. Darwin) and as 2 if a word is necessary and fundamental to explain and understand evolution (e.g. natural selection). Individual scores were calculated by summing the word scores of a participant. Because we wanted to evaluate instinctive answers, a participant had on average 3 s to provide an answer and if hesitation was long, it was recorded as 'non-association' . When participants declared that they didn't know it was recorded as 'not applicable-NA' .
We recognize that the experimental design of this exercise has flaws: ideally, the number of incorrect terms should be similar to the one of correct and essential ones and the order in which the words were presented to each participant should have been randomized (all participants were read the same words in the same order). However, we believe that the report of the results might inform future outreach activities and thus present it as basal data and not as a proof of principle.
Data analysis
All data analyzed for comparisons (mean differences) and correlations were checked for deviations from assumptions using a Shapiro test (α = 0.05). If data distribution did not significantly differ from normal distribution, parametric versions of the relevant statistics were used, otherwise non-parametric statistics were used (Student's t-test Vs. Mann-Whitney U test for mean differences; and Parametric (bivariate) Vs. Non-parametric (Spearman) for correlations). Each result is accompanied by the specific test performed in the results section. Data were always standardized for the specific number of participants comprising the dataset.
We emphasize that our sample sizes are small which translates into data limited in its power and thus we provide these as explorative baseline data regarding our specific activity, which might inform future research questions on inclusive pedagogy.
All analysis and data visualization were performed in R version 3.6.0 (R Core Team 2019) and all raw data and analysis code are available to allow full analysis replication (Additional files 3 and 8, respectively).
A description of all figures and images presented below, accessible for people with visual impairment, is presented in Additional file 7.
The multisensory tree of life for all
The multisensory Tree of Life (referred to as 'MSToL' from here on) occupied 125 square meters and was composed of branches representing 21 extant taxa-plus eight fossil species. The majority of materials consisted of real biological samples ( Fig. 2; Additional file 1: Fig. S1 for detailed photos per branch; Additional file 2: Table S1 for a list of materials, source, sense stimulated and evolution concepts explored, per branch). All five basic senses-hearing, smelling, touching, tasting, seeing-were stimulated across the phylogeny, with touch being stimulated by all displayed material, becoming the main source of information acquisition for people with visual impairment.
Each consenting participant-23 sighted and 25 with visual impairment-contributed to data collection through a questionnaire applied before and after experiencing the MSToL room, and branch-specific quizzes. The resulting data allowed us to generally evaluate the success of the activity in terms of learning, and to establish baseline data on evolution knowledge for a subset of the Portuguese community with visual impairment.
Implementation: the hardest and easiest taxa to represent
Upon consulting with blind members of the community of the Portuguese association for the teaching of the blind (Associação Promotora do Ensino dos Cegos, APEC), it became clear that touch would be the most inclusive sense to explore the phylogeny. Thus, for the first step-allowing participants with visual impairment to assess biodiversity accurately-haptic communication was essential. Because we wanted all participants to experience real biodiversity patterns, we mainly acquired real specimens and biological samples that could be touched. This can, however, raise challenges for the representation of some taxa: museum collections with scientific value are usually unique and fragile, which hinders their free manipulation; live specimens pose animal welfare concerns; and commercially available models can be inaccurate and lack the detail needed to fully comprehend the range of biodiversity patterns. Several of these difficulties applied to the arthropods, which proved to be the most challenging taxa to translate into multisensory communication. We addressed this difficulty by incorporating 3D-prints of in-house μ-CT scanned specimens, insect sounds, edible insects, exuviae and structures built by arthropods, such as hives (Additional file 1: Fig. S1 e to g). Collaboration with education centers with available pedagogic collections was imperative to obtain material that accurately depicts biodiversity (Additional file 1: Fig. S1; Additional file 2: Table S1). The most effective branches in terms of degree of 'effort to find material' with respect to the magnitude of 'activities and information that can be extracted from it' were plants, birds and primates. A great variety of plants are commercially available and they are ideal to develop activities focused on plant-pollinator coevolution and phenotype-climate adaptations (Additional file 1: Fig. S1 t to w). For birds, songs, calls and full specimens of game species are easily obtainable and optimal for activities focused on phenotype-environment adaptations (Additional file 1: Fig. S1 n). The evolution of humans and related primates holds, generally, particular interest for the public (Pobiner 2012). Anthropological model collections can be expensive but are usually available at universities where anthropology is taught, which can be invited to lend this resource. A collection of hominid skulls is a great resource for discussion of human evolution and for understanding common ancestry while dismantling the myth that "Homo sapiens descends from monkeys".
While teeth and different types of fur can be very interesting resources to discuss adaptation, without complete spatial or morphological information they can be confusing for participants with visual impairment, especially for those born blind. To avoid this we made sure that models of the full organism were available for any type of partial specimen, which the participants reported to be extremely useful.
Public attendance
During the 12 h of activity, we received an estimated total of 100 participants, 60 of which had visual impairment. We did not restrict the amount of time to explore the MSToL, allowing the participants to do so at their own pace.
Participants, especially those with visual impairment, tended to remain in the room more than the predicted one hour, with some remaining for as long as four hours. This should thus be a full day activity at minimum, and ideally a multiple-day activity.
A subset of consenting participants responded to a standardized questionnaire which included general, as well as branch-specific, questions (Additional file 5: questionnaire), administered both before and after experiencing the MSToL room. The sample comprised 25 adults with visual impairment, with average age 62 (range: 18 to 82) and 23 sighted adults, with average age 58 (range: 24 to 90).
The first assessment of the data showed that the majority of participants with visual impairment had profound vision loss without perception of any visual cue, referred to as blindness throughout the manuscript (13/25) and only a minority (4/25) had moderate vision loss, with perception of shapes and some colors, and eight out of 25 had deep sight loss with residual light perception (see methods for details of the participants' self-assessment). The majority of participants with visual impairment had the impairment for more than half of their lifetime, with only seven living with it for less than that period.
Preferences: touch and birds
The sense preferences for experiencing the MSToL were consistent across participants, regardless of visual ability. Touch was listed as the most informative sense by 83.3% of participants with visual impairment and 50% of sighted participants (Additional file 3: Table S2). On the other end, gustation was listed as the least informative sense by 93.7% of participants with visual impairment and 53.3% of sighted participants (Additional file 3: Table S2), which was also the least stimulated sense across the phylogeny (Fig. 2b, Additional file 1: Table S1).
Despite an overall scattered preference across taxa, people with visual impairment showed slight predilection for the bird branch: five out of 19 people with visual impairment chose the birds as their favorite branch (Additional file 3: Table S2). This tendency might have been influenced by the communication skills of the educator responsible for the birds' branch, but it is worth noting that even for pragmatic reasons birds are a good taxonomic group to invest in further, more focused, activities. This taxon allows multisensory activities on adaptation and evolution to be design relatively easily due to the ease of acquiring diverse feather types, the abundant availability of taxidermy specimens with different bill shapes among game species, and also the widespread availability of bird song recordings, easily adding an auditory component.
Despite lack of access to education, participants with visual impairment are interested in, and understand, evolution
We found that participants with visual impairment had a lower level of educational attainment (Fig. 3a). The majority of participants with visual impairment had not enrolled in high-school education (56%, 14 out of 25), only seven participants with visual impairment had attained post-secondary education, and none held a master's degree. In comparison, the majority of sighted participants had attained bachelor's degrees (59.1%, 13 out of 22; Fig. 3a). Only two of the seven participants with visual impairment (28.6%) with post-secondary education had taken courses in biology and evolution (at the bachelor's degree level), compared to six of the 17 sighted participants (35.3%; Table S2).
Interestingly, when enquired regarding their interest for evolution (for questionnaire scale description refer to Additional file 5: questionnaire; and for individual data refer to Additional file 3: Table S2), a similar percentage of participants from both groups described themselves as having a lot of interest in evolution (28% with visual impairment; 39% sighted). Indeed, two participants with visual impairment shared with the educators that they wanted to become biologists, having given up because "it was too visual", making it clear that the biological sciences are not equally accessible to everyone. However, basal knowledge, which was measured by the amount of correct answers scored before the activity, was quite high for both groups: all participants tended to score high both in the true and false questions and in the word association exercise (Additional file 6: Fig. S3). We did not find statistical differences in average performance between sighted people and people with visual impairment (Fig. 3; Additional file 6: Fig. S2 and S4), based on learning ( Fig. 3a; Pearson correlation: r = 0.79 for participants with visual impairment; Spearman correlation rho = 0.81 for sighted) and prediction scores ( Fig. 3b; t-test: p-value = 0.77). However, the higher 'learning' and 'prediction' scores tended to belong to sighted participants, while the lowest scores tended to be recorded for participants with visual impairment (Fig. 3; Additional file 6: Fig. S2 and S4; refer to Methods for a detailed description of all score metrics). Based on the results from the branchspecific activities, it is also notable that, when enquired about the consequences of environmental changes, both sighted people and people with visual impairment could successfully predict evolution outcomes in terms of expected phenotypic changes ( Fig. 3b; refer to Additional file 4 for detailed description of prediction exercises), which shows a general basic understanding of the mechanisms of natural selection and adaptation.
The fact that we have a small sample size, together with the substantially high basal knowledge (Additional file 6: Fig. S3) found for both participant groups (Fig. 3a), makes it difficult to assess the true effectiveness of the activity based exclusively on in loco data.
An inclusive description of all figures and plots above-mentioned is presented in the Additional file 7.
'Common ancestor' and 'Natural Selection' become more familiar terms, but artificial selection and neutral evolution might be hard to grasp without specific activities At the beginning and end of the activity, participants were read the same list of 33 terms and asked if they instinctively associated them with the concept of 'evolution' (Fig. 4; refer to the methods section for details). This meant to assess whether terms that are essential to understand evolution-like adaptation and common ancestor-were clearly present in the participant's minds.
After the activity, association of terms increased in general. Overall, the terms that increased the most were 'Common ancestor' and 'Natural selection' for the participants with visual impairment; and 'Ecology' and 'Natural selection' for sighted participants (Fig. 4, Additional file 6: Fig. S5). All these terms were heavily used on the scripts provided to the educators. However, for the participants with visual impairment, the association of terms that can lead to misunderstanding evolutionsuch as perfecting and progress-increased more (Fig. 4, Additional file 6: Fig. S5).
All participants associated all neutral terms-such as tree or ramification-with 'evolution' both before and after the activity (Additional file 6: Fig. S5).
Interestingly, the term "artificial selection" decreased in association for both groups. This is not surprising since there were no artificial selection activities or mentions. This term was not used during the workshop, as the majority of the examples and predictive exercises were based on well-studied responses to natural selection, such as the stickleback fish plating reduction or the food-availability-driven beak morphology of birds.
Also interesting is that while 'mutation' was highly associated with evolution by both groups, already before the activity (84% of sighted and 86% of participants with visual impairment), 'chance' was not. Together with the fact that 'Progress' and 'Perfecting' tended to increase, this might signal that a linear view of evolution towards perfection and humanization is rooted in the participant minds and calls for the need of including neutral forces in outreach activities to avoid promoting wrong or extremely adaptationist views of evolution. Fig. 3 Participant education data and scores, based on basic knowledge of evolution before and after the activity, and prediction of evolution outcomes. Blue data refer to sighted participants and black data to participants with visual impairment. a Depicts the relationship between participants age and education level. Lines visualize the linear model per participant category and vertical dashed lines mark the average age of each group. b Refer to the scores of true or false questions on basic evolution concepts applied before and after the MSToL activities. Lines show the correlation between before and after score. For sighted participants (blue) and visually impaired participants (black), R squared values are depicted from a regression analysis with 'Score after the activity' dependent on 'Score before the activity' . c Visualizes participants' predictive ability in scenarios where environments shape phenotypic responses, following the same color code Volunteer educators' emotional experience reveals no discomfort in communicating with participants with visual impairment and that through teaching they also learn a lot, while having a lot of fun.
To assess the benefits of inclusive outreach activities for the educators, we conducted a brief post-activity questionnaire on the volunteer educators.
Despite the majority of volunteers (93.3%; 14/15) not having previous experience communicating with people with visual impairment, during the activity, 53.3% felt more at ease communicating with people with visual impairment, especially with those participants above 60 years old (Fig. 5a). When asked what was their favorite aspect of communicating with participants with visual impairment, the volunteers expressed feelings of empathy and mentioned discovering the world from a new perspective. Furthermore, the great majority of volunteer educators reported that they learned immensely (Fig. 5b) while educating, especially communication skills, and reported they had a lot of fun while teaching evolution (Fig. 5c).
Discussion
The development of the multisensory tree-of-life and the data collected during the activity provide evidence for three main arguments: (1) there is a clear lack of accessibility to evolutionary biology education for people with visual disability, despite their interest in the matter; (2) the classic examples of evolution are capable of being transformed into multisensory activities; (3) touch and haptic models are essential for people with visual disability and a plus for sighted participants, making haptic activities the most powerful resource to increase accessibility and inclusion, benefiting everyone, independently of physical impairment.
Lessons learned to provide a comfortable and stimulating experience for people with visual impairment
As a pioneer activity, we were faced with a lot of hurdles due to inexperience. With this publication we hope to reduce those for anyone reproducing this activity or creating similar ones. However, our main and most valuable lesson concerns communication. As sighted organizers Fig. 4 Terms associated with the concept of evolution before and after the MSToL activity. Metric shown is the difference of percentage of people associating the word with evolution after the activity minus the percentage of people associating the word with evolution from the beginning. In dark green are terms essential to understand evolution, in blue are words related, but not essential; in grey neutral terms, and in red terms that can be misused or promote the misunderstanding of evolution. Notice that neutral term bars are absent because they were equally associated before and after the activity (see Additional file 6: Fig. S5). Vertical dashed lines depict the average difference of each word group, following the same color scheme we needed to make sure that our ideas and translation of visual into haptic were indeed accessible to people with visual impairment; and that the MSToL room was comfortable for everyone. Consulting with people with visual impairment or blindness is absolutely indispensable at every step. Initially, we had designed the activity as having braille instructions at the tables, which greatly diminishes the amount of volunteer educators needed. However, our blind consultant informed us that in our local community of people of visual impairment, only a small minority knew how to read braille. Another very important insight was the fact that people who have lived with visual impairment for different lengths of their lives will have different sensitivities to more subtle textures, might be more or less comfortable with the volume of sound in the room, and might be more or less experienced in navigating a room with floor textures. Thus, tripping hazards should be avoided when planning the floor phylogeny texture and the disposition of the branches-flat carpet for the phylogeny and tables closer to the walls of the room were our optimal design. If there are participants with motor disabilities or on wheelchairs, the height and shape of the display tables also need to be considered for accessibility. A big part of offering a safe and stimulating environment for people with visual impairment is the acoustics of the room: a lot of echo and noise easily becomes overwhelming. To avoid this, it is important to control the flow of people in the room and the volume of the sounds within. Following the available guidelines together with communicating and consulting with the local community of people with visual disability are crucial to ensure a comfortable learning experience for everyone.
Be careful with the 'evolution ladder towards humanization'
The aspect that we believe is the most urgent to improve in this and other activities, is the fact that the public easily retains the misconception of evolution as a linear processes towards humanization. Despite directly countering this idea through the branching phylogeny patterns presented, a question that seemed to hold low scores, across all participants, both before and after the activity, asks if 'mammals are more evolved than fish' (Refer to Additional file 6: Fig. S3 for individual question scores). In future activities, attention should be paid to the fact that having mammals and, more specifically, the Homo branch 'higher' on the phylogeny display-in our case towards the end of the experience for someone who started at the root (Fig. 2)-might promote the incorrect, and quite common, notion of evolution 'progressing' linearly towards humanization and taxa that are evolutionarily closer to our species.
Thus, as something to improve, topology display (e.g. display mammals more to the side instead of at the top, do not make Homo the last visited branch) and language use (e.g. never use 'more primitive' or 'basal' for any extant taxon) should be mindful of inadvertent contribution towards the adaptationist and directional evolution narratives.
Towards more effective and inclusive outreach activities
Global estimates are clear regarding the under-representation of people with disability in STEM, both in the classroom and in the academic community (Moon et al. 2012). Under this status quo, people with visual impairment are denied access to knowledge and to participation in scientific communities. Consequently, we lose diversity of thought and experiences that could promote more stimulating ways of teaching that everyone could benefit from.
In evolutionary biology, besides displaying data and concepts, images are a source of interest and wonder for biodiversity, which fuels curiosity. Being so, the promotion of scientific literacy demands the translation of evolution's patterns to senses other than vision. Although text and audio descriptions of graphical representations are useful, students with visual disability have reported that many important details are left out or misinterpreted by the translator (Shute et al. 2005). Furthermore, as the complexity of visual content intensifies so does the challenge of presenting it through auditory cues (Shute et al. 2005).
In a comparative study on sighted science students, tactile learners retained and understood concepts better, while also enjoying their lessons much more (Pashler et al. 2009). The preference of both MSToL participant groups (with and without visual impairment) for 'touch' as a learning sense, and the overall positive global learning scores reported in our activity, further suggest 'touch' as a generally inclusive and powerful vehicle of information delivery.
In fact, evolution concepts can be ideal for tactile learning, because much of the visual content represents descriptions of morphological and environmental variation, easily translated into 3D haptic images. When these resources are incorporated into science teaching, interest can increase for both sighted and blind students (Hasper et al. 2015). Thus, as shown by growing evidence from life sciences (Fraser and Maguvhe 2008), communication should be multisensory to increase teaching effectiveness for all students.
Conclusion
Without inclusive approaches, students with visual impairment often lose motivation due to real or perceived physical barriers to knowledge acquisition (Bell and Silverman 2018). However, when knowledge is made accessible they can realize their potential just as sighted students do (Sahin and Yorek 2009). Therefore, the inclusion of multisensory activities in outreach, which we have shown to be quite accessible for a lot of branches of the tree-of-life, can have important academic and social impacts. Just like museums, outreach activities should be 'inclusive and polyphonic spaces that address present social challenges and promote active partnerships with and for diverse communities, contributing to human dignity and social justice, global equality and planetary wellbeing' (Sandahl 2018). Involving complementary senses on future activities will not only promote equity for those with disability, but also move us faster towards an inclusive and diverse scientific community, and towards a public more aware of biodiversity, evolution and our connection to it.
==
Domain: Psychology Biology
|
[{"added": "2023-01-07T15:33:06.512Z", "created": "2021-03-11T00:00:00.000", "id": "255487335", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://evolution-outreach.biomedcentral.com/counter/pdf/10.1186/s12052-021-00143-1", "pdf_hash": "b84ff3099f1a8fd102ee969e48638d37168dd146", "pdf_src": "SpringerNature", "provenance": "peS2o-0075.json.gz:33769", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "b84ff3099f1a8fd102ee969e48638d37168dd146", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Myelinating Co-Culture as a Model to Study Anti-NMDAR Neurotoxicity
Anti-NMDA receptor (NMDAR) encephalitis is frequently associated with demyelinating disorders (e.g., multiple sclerosis (MS), neuromyelitis optica spectrum disorder (NMOSD), myelin oligodendrocyte glycoprotein-associated disease (MOGAD)) with regard to clinical presentation, neuropathological and cerebrospinal fluid findings. Indeed, autoantibodies (AABs) against the GluN1 (NR1) subunit of the NMDAR diminish glutamatergic transmission in both neurons and oligodendrocytes, leading to a state of NMDAR hypofunction. Considering the vital role of oligodendroglial NMDAR signaling in neuron-glia communication and, in particular, in tightly regulated trophic support to neurons, the influence of GluN1 targeting on the physiology of myelinated axon may be of importance. We applied a myelinating spinal cord cell culture model that contains all major CNS cell types, to evaluate the effects of a patient-derived GluN1-specific monoclonal antibody (SSM5) on neuronal and myelin integrity. A non-brain reactive (12D7) antibody was used as the corresponding isotype control. We show that in cultures at the late stage of myelination, prolonged treatment with SSM5, but not 12D7, leads to neuronal damage. This is characterized by neurite blebbing and fragmentation, and a reduction in the number of myelinated axons. However, this significant toxic effect of SSM5 was not observed in earlier cultures at the beginning of myelination. Anti-GluN1 AABs induce neurodegenerative changes and associated myelin loss in myelinated spinal cord cultures. These findings may point to the higher vulnerability of myelinated neurons towards interference in glutamatergic communication, and may refer to the disturbance of the NMDAR-mediated oligodendrocyte metabolic supply. Our work contributes to the understanding of the emerging association of NMDAR encephalitis with demyelinating disorders.
Introduction
Anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis is a severe neuropsychiatric immunological disorder, which is diagnosed by the presence of circulating anti-NMDAR autoantibodies in the cerebral spinal fluid (CSF) [1][2][3]. The clinical symptoms developed by patients are thought to be due to substantial alterations of glutamatergic transmission, and include psychiatric and cognitive manifestations that progress to movement disorder, speech dysfunction, seizures, impaired consciousness, and autonomic instability. Often anti-NMDAR encephalitis is also associated with tumors, mostly ovarian teratoma [4]. Early immunotherapy and/or tumor removal leads to a substantial or even complete recovery in about 80% of patients [5]. Histopathological examination demonstrates extensive microgliosis and the presence of IgG deposits in the hippocampus, basal forebrain, basal Int. J. Mol. Sci. 2023, 24, 248 2 of 15 ganglia, and spinal cord, including signatures of endoneural edema and Wallerian degeneration in affected neural tissue [6]. Approximately 5% of patients develop clinical and radiological evidence of a demyelinating disorder [7].
Autoantibodies to glutamate receptors with a proven pathological role include the immunoglobulins G (IgG) directed against ionotropic receptors AMPAR-GluR3, NMDAR-GluN1, -GluN2A, and -GluN2B, as well as AABs against metabotropic receptors (anti-mGluR1 and anti-mGluR5 glycoproteins) (reviewed in [8]). In contrast to other anti-NMDARs (e.g., anti-GluN2A or anti-GluN2B) eliciting excitotoxicity via receptor activation [8], crosslinking of NMDARs by autoantibodies to the extracellular domain of the obligatory GluN1 subunit affects glutamate signaling via mechanisms involving receptor internalization, decreased surface cluster density, and synaptic localization of the NM-DARs [9,10]. Of note, compensatory synaptic plasticity mechanisms activated following the exposure to anti-GluN1 antibodies do not induce counterbalancing changes in the expression of the glutamate receptor; however, they cause a decreased inhibitory synapse density onto excitatory neurons [11]. The effect of human anti-GluN1 IgG on inhibitory and excitatory neurons demonstrates brain region-specific alterations. In contrast to hippocampal neurons, dysfunction of inhibitory neuron output by anti-GluN1 IgG triggers a hyperexcitable state in cultivated cortical neurons leading to network hyperactivity [12]. NMDAR hypofunction has been proposed to be a part of the pathophysiological mechanisms underlying anti-GluN1-driven anti-NMDAR encephalitis [13].
We have previously isolated an anti-NMDAR-GluN1 (SSM5) autoantibody from intrathecal plasma cells of a patient with anti-NMDAR encephalitis and demonstrated in vivo its pathogenic relevance [14]. We showed that the effect of SSM5 is not only restricted to neurons, and may also affect other cell types in which glutamatergic transmission plays a role. For example, anti-GluN1 IgG affects the NMDAR-induced Ca 2+ response and the glucose transporter 1 (GLUT1; SLC2A1) surface expression in myelinating oligodendrocytes [15]. This effect may strikingly influence trophic oligodendrocyte-neuron communication and energetic support of myelinated axons through the recently described glutamate-mediated signaling in axon-myelin synapses [16].
In the present study, we examine the effects of SSM5 and a corresponding isotype control, non-brain reactive monoclonal IgG1 (12D7) [14], particularly in myelinated neurons. We apply a robust model of long-term cultures derived from dissociated embryonic mouse spinal cord [17]. During cultivation, spinal cord cultures produce compactly myelinated internodes separated by nodes of Ranvier [18] and generate neurons that recapitulate spinallike electrical activity [19], serving as a valuable tool for functional studies in myelinated, synapse-forming CNS neurons [20]. Besides neurons and oligodendrocytes that constitute the major cell population, the typical E12.5-E13.5 murine spinal cord-derived culture contains astrocytes and microglia that recapitulate innate CNS immune functions [21,22]. We show that myelinating spinal cord culture is a suitable model to investigate the effects of human anti-GluN1 antibodies and report that targeting GluN1 disturbs neuronal integrity and myelination, an effect that has been disregarded in other culture models.
Defining an Optimal Stage of Culture for Treatment
Mouse embryonic spinal cord cells were isolated from 12.5 days old embryos (E12.5) and maintained in culture, as described previously [17]. Cells were cultivated for 5 weeks, and progression of neuronal growth and myelination were examined. By day 14 in vitro (DIV), neurite lengthening, visualized by antibodies to neuronal differentiation marker neurofilament 200 kDa (NF200) [23,24], was completed ( Figure 1). Upon differentiation, oligodendrocytes developed most of their processes aligned with axons, and were labelled by antibodies against myelin basic protein (MBP). Myelination starts between 3 and 4 weeks of culture (DIV 21-28) and peaks around DIV 35 (Figure 1b). Although not all oligodendrocytes become fully mature, about two-third of MBP-positive cells provided single or multiple myelin segments. The extent of neuronal insulation at the peak of myelination (DIV 35) was proportional to neurite density in individual wells ( Figure 1c). ment ( Figure S1a) and widespread neuronal degeneration. Neuronal vulnerability to NMDA coincided with the loss of myelin (Figures 1b and S1b).
Supplementation of cultures for one week with 1 μM thyroid hormone triiodothyronine (T3) that supports functional maturation and myelination during oligodendrocyte development [26], increased density of myelin segments in differentiating cultures maintained for DIV 21-35 ( Figure S1b). Notably, treatment with T3 starting after DIV 35 had no effect on myelination, indicating that oligodendrocyte maturation is mainly completed during the first 5 weeks in vitro (Figure 1b). Taken together, characterization of long-term cultures demonstrated that at DIV 21-35, differentiating neural cells develop functional NMDARs and are sensitive to cell growth-promoting mediators. DIV 14). Progression of myelination over 5 weeks in culture was identified using antibody to intermediate neurofilament heavy chain (NF200; red) and MBP (green), the lineage markers specific for neurons and mature myelinating oligodendrocytes, respectively. (b) Quantification of myelination. Overlap MBP signal with NF200 enable to visualize myelin segments. Double immunostained images were processed with a CellProfiler pipeline (example shown for DIV 21) that enables quantification of the pixel area for each signal. Myelin segments were calculated as an overlay of red and green lines. (c) At the peak of myelination (DIV 35), the extent of neuronal insulation is proportional to neurite density in individual wells. Data summarized results from three DIV 14). Progression of myelination over 5 weeks in culture was identified using antibody to intermediate neurofilament heavy chain (NF200; red) and MBP (green), the lineage markers specific for neurons and mature myelinating oligodendrocytes, respectively. (b) Quantification of myelination. Overlap MBP signal with NF200 enable to visualize myelin segments. Double immunostained images were processed with a CellProfiler pipeline (example shown for DIV 21) that enables quantification of the pixel area for each signal. Myelin segments were calculated as an overlay of red and green lines. Next, we tested developing spinal cord cultures for sensitivity to NMDA in the presence of glycine (co-agonist; available in the culture medium at estimated concentration 0.2-0.25 mM). In line with previous data [25], overstimulation of spinal cord neurons with 100 µM NMDA starting at DIV 21, DIV 28 or DIV 35 lead to a massive loss of neurofilament ( Figure S1a) and widespread neuronal degeneration. Neuronal vulnerability to NMDA coincided with the loss of myelin (Figure 1b and Figure S1b).
Supplementation of cultures for one week with 1 µM thyroid hormone triiodothyronine (T3) that supports functional maturation and myelination during oligodendrocyte development [26], increased density of myelin segments in differentiating cultures maintained for DIV 21-35 ( Figure S1b). Notably, treatment with T3 starting after DIV 35 had no effect on myelination, indicating that oligodendrocyte maturation is mainly completed during the first 5 weeks in vitro (Figure 1b). Taken together, characterization of long-term cultures demonstrated that at DIV 21-35, differentiating neural cells develop functional NMDARs and are sensitive to cell growth-promoting mediators.
Immunoreactivity of SSM5 in Spinal Cord Cultures
Immunodetection with patient-derived anti-NMDAR-GluN1 antibody (SSM5) showed a dot-like immunoreactivity to neuronal processes starting around DIV 14-21 ( Figure 2). SSM5 also co-localized with a fraction of chondroitin sulphate proteo-glycan (NG2)-positive cells ( Figure S2) and myelin (MBP) aligned with axons ( Figure 2). These results suggest that SSM5 affects glutamatergic transmission in neuronal and oligodendroglial cells derived from the embryonic spinal cord.
SSM5 Antibody Induces Neuronal Damage and Myelin Loss
Next, we studied the potential effect of SSM5 on neuronal survival and myelination. In order to avoid the negative effects of GluN1 targeting on neurite outgrowth and spine maturation of progenitor cells [27,28], we used 3-to 5-week-old cultures (DIV 21-35)-the period at which most neurons exhibited differentiated morphology ( Figure 1a). 10 μg/mL SSM5, a concentration in culture media shown to be effective for the blocking of NMDAevoked Ca2+ influx in neurons [14], were added to the myelinating cultures, and the effects were compared to isotype (12D7; 10 μg/mL) and non-treated control cultures. Oneweek treatment with SSM5 revealed a significant decrease in neurofilament content in late (DIV 28-42) cultures, while isotype control (12D7; 10 μg/mL) had no toxic effect ( Figure 3). Loss of neurofilament immunoreactivity occurred progressively and reached highest values in cultures treated starting from DIV 35 (6% less of neurofilament at DIV 21-28; 18% at DIV 28-35; 23% at DIV 35-42 in comparison to the respective 12D7 controls), indicating that susceptibility to SSM5 toxicity is related to the stage of neural differentiation ( Figure 3b). In parallel to the neurotoxic effect, SSM5 significantly lowered myelination in late cultures (7% less myelin segments at DIV 21-28; 18% at DIV 28-35; 21% at DIV 35-42 as compared to 12D7 group) ( Figure 3c). We also found a significant difference in the percentage of nuclei exhibiting signs of pyknosis or karyorrhexis (DIV 21-28: 17% vs. 26% of abnormal nuclei; DIV 35-42: 19% vs. 28% of abnormal nuclei, compared to 12D7 group) ( Figure S3).
SSM5 Antibody Induces Neuronal Damage and Myelin Loss
Next, we studied the potential effect of SSM5 on neuronal survival and myelination. In order to avoid the negative effects of GluN1 targeting on neurite outgrowth and spine maturation of progenitor cells [27,28], we used 3-to 5-week-old cultures (DIV 21-35)-the period at which most neurons exhibited differentiated morphology ( Figure 1a). 10 µg/mL SSM5, a concentration in culture media shown to be effective for the blocking of NMDAevoked Ca2+ influx in neurons [14], were added to the myelinating cultures, and the effects were compared to isotype (12D7; 10 µg/mL) and non-treated control cultures. One-week treatment with SSM5 revealed a significant decrease in neurofilament content in late (DIV 28-42) cultures, while isotype control (12D7; 10 µg/mL) had no toxic effect ( Figure 3). Loss of neurofilament immunoreactivity occurred progressively and reached highest values in cultures treated starting from DIV 35 (6% less of neurofilament at DIV 21-28; 18% at DIV 28-35; 23% at DIV 35-42 in comparison to the respective 12D7 controls), indicating that susceptibility to SSM5 toxicity is related to the stage of neural differentiation (Figure 3b). In parallel to the neurotoxic effect, SSM5 significantly lowered myelination in late cultures (7% less myelin segments at DIV 21-28; 18% at DIV 28-35; 21% at DIV 35-42 as compared to 12D7 group) ( Figure 3c). We also found a significant difference in the percentage of nuclei exhibiting signs of pyknosis or karyorrhexis (DIV 21-28: 17% vs. 26% of abnormal nuclei; DIV 35-42: 19% vs. 28% of abnormal nuclei, compared to 12D7 group) ( Figure S3). Immunofluorescence analysis of neurites (NF200; red) and myelin (MBP; green) after treatment with SSM5 (10 μg/mL) or 12D7 (10 μg/mL) for one week at different stages of myelination. Scale bar: 200 μm. Quantification of neurofilament density (b) and myelin segments (c) following treatment with SSM5 starting at DIV 21, DIV 28, DIV 35. One-way ANOVA with Tukey multiple comparison, showed significant differences between SSM5 vs. 12D7 vs. control groups; n = 4; technical replicates = 3; * p < 0.05, ** p < 0.01; ns, non-significant. Note: detrimental effect of SSM5 was more pronounced at later stages of culture and reached statistical significance after DIV 28. Mean ± SEM.
To characterize the affected neurons, we analyzed neurite blebbing and fragmentation as markers of neurodegeneration during both spinal cord neuronal injury [29,30] and autoimmune antibody-mediated neurodegeneration [31,32]. Evaluation of neuronal morphology in cultures exposed to SSM5 revealed an increase in the number of injured neuronal processes (up to 8% at week DIV 21-28, 26% at DIV 28-35 and 33% at DIV 35-42; compared to 12D7 group) ( Figure 4). Importantly, the severity of SSM5 neurotoxicity increased in late cultures characterized by higher proportion of myelinated axons (see Figure 1b). Thus, it is tempting to speculate that axonal myelination made them more vulnerable to neurotoxic stimulus. The number of injured neurons in cultures treated with 12D7 antibodies appeared similar to non-treated cultures. Quantification of neurofilament density (b) and myelin segments (c) following treatment with SSM5 starting at DIV 21, DIV 28, DIV 35. One-way ANOVA with Tukey multiple comparison, showed significant differences between SSM5 vs. 12D7 vs. control groups; n = 4; technical replicates = 3; * p < 0.05, ** p < 0.01; ns, non-significant. Note: detrimental effect of SSM5 was more pronounced at later stages of culture and reached statistical significance after DIV 28. Mean ± SEM.
To characterize the affected neurons, we analyzed neurite blebbing and fragmentation as markers of neurodegeneration during both spinal cord neuronal injury [29,30] and autoimmune antibody-mediated neurodegeneration [31,32]. Evaluation of neuronal morphology in cultures exposed to SSM5 revealed an increase in the number of injured neuronal processes (up to 8% at week DIV 21-28, 26% at DIV 28-35 and 33% at DIV 35-42; compared to 12D7 group) ( Figure 4). Importantly, the severity of SSM5 neurotoxicity increased in late cultures characterized by higher proportion of myelinated axons (see Figure 1b). Thus, it is tempting to speculate that axonal myelination made them more vulnerable to neurotoxic stimulus. The number of injured neurons in cultures treated with 12D7 antibodies appeared similar to non-treated cultures.
We anticipated the myelin and neurite swellings induced by the SSM5. Analysis of damaged processes that remain individual myelinated segments indicates that changes in the myelin sheath structure (presumably myelin blisters [33]) occur at the site of axonal injury characterized by structural blebbing ( Figure 5). Demyelinated axonal lesions frequently flank normally myelinated fragments (without axonal blebbing), hinting that the local area of damage serves as a platform for further myelin destabilization and demyelination. However, the exact sequence of events leading to structural abnormalities in myelin and axons remains to be clarified. week with 12D7 (middle panels) or SSM5 (right panels) and immunostained for NF200 (grey). Red arrowheads indicate affected neuronal processes. Scale bar: 100 μm. (b) Quantification of damaged neurites. Data show the percentage of processes exhibiting morphological abnormalities (swelling, fragmentation) of total number of neurites. Three independent experiments (technical replicates = 2), Mean ± SEM. One-way ANOVA with Tukey multiple comparison showed significant differences between SSM5 vs. 12D7 vs. control groups only when treatment was performed in late cultures (DIV 28-42); ns, non-significant; ** p < 0.01.
We anticipated the myelin and neurite swellings induced by the SSM5. Analysis of damaged processes that remain individual myelinated segments indicates that changes in the myelin sheath structure (presumably myelin blisters [33]) occur at the site of axonal injury characterized by structural blebbing (Figure 5). Demyelinated axonal lesions frequently flank normally myelinated fragments (without axonal blebbing), hinting that the local area of damage serves as a platform for further myelin destabilization and demyelination. However, the exact sequence of events leading to structural abnormalities in myelin and axons remains to be clarified. We anticipated the myelin and neurite swellings induced by the SSM5. Analysis of damaged processes that remain individual myelinated segments indicates that changes in the myelin sheath structure (presumably myelin blisters [33]) occur at the site of axonal injury characterized by structural blebbing (Figure 5). Demyelinated axonal lesions frequently flank normally myelinated fragments (without axonal blebbing), hinting that the local area of damage serves as a platform for further myelin destabilization and demyelination. However, the exact sequence of events leading to structural abnormalities in myelin and axons remains to be clarified.
Discussion
In this study, we describe the effects induced by patient-derived anti-NMDAR (GluN1) IgG1 in mouse myelinating spinal cord cultures, and demonstrate neurotoxic and demyeli-nating effects of GluN1 targeting in vitro. This finding is in accordance with previous histopathological and MRI examinations in patients with anti-NMDAR-associated encephalitis, demonstrating the development of neurodegeneration in spinal neurons [6], grey matter atrophy [34] and white matter damage [7,[35][36][37]. Moreover, diaplacentally transferred anti-NMDAR AABs cause severe neurotoxic effects on neonatal development [27] that may result in long-lasting neuropathological effects [38], corroborating our finding using in vivo models.
Examination of neurite blebbing and fragmentation, which are features of neurodegeneration, revealed a higher percentage of lesioned neurons induced by SSM5 application in late myelinated cultures (DIV [28][29][30][31][32][33][34][35][36][37][38][39][40][41][42]. This is the period (approximately DIV25 ± 5 days) when most myelinating oligodendrocytes formed mature myelin sheathes, as demonstrated in Figure 1 and previous reports [17,18]. Cultures treated in the early stages of the myelinating process (DIV 21-28) remained largely unaffected, indicating that susceptibility to SSM5 toxicity may likely be related to sheath maturation and compaction of the oligodendrocyte membrane around axons. In line with our previous data demonstrating the negative effects of SSM5 on NMDA-mediated Ca 2+ responses and glucose transporter GLUT1 plasma membrane translocation in mature oligodendrocytes [15], we observed SSM5 immunoreactivity on the myelin compartment. Oligodendroglial NMDARs mediate spontaneous synaptic currents [39] and function as typical receptor-gated cation channels. Their stimulation can result locally from glutamatergic axons, with which oligodendrocyte form direct synapses [40,41]. Alterations in myelin NMDAR signaling and glucose uptake induced by SSM5 correlate with structural degenerative changes in neurites and myelin loss observed in the present study. These findings may hint at impaired glutamate-dependent axonmyelin communication ( Figure 6). Our findings are in line with the proposed role of myelin NMDAR signaling in glycolytic support of myelinated axons that face energetic challenges when being physically isolated from the trophic metabolites (e.g., glucose, lactate) in the extracellular space [16,42,43]. The neurosupportive role of myelin NMDAR signaling was demonstrated in mutant mice, in which genetic ablation of the GluN1 subunit in cells of oligodendrocyte lineage underlies neurological deficits (hind limb clasping, hunchback, and ataxia). Particularly, GluN1 deficiency leads to ongoing axonal degeneration coinciding with myelin abnormalities (myelin delamination) in brain and spinal cord white matter [16]. With a greater similarity, we observe these processes in long-term myelinated co-cultures exposed to SSM5 ( Figure 5). More recently, an altered expression of myelin NMDAR subunits has been linked to destabilization occurring in axon-myelin units in normal-appearing white matter of patients with multiple sclerosis [33]. Aberrant NMDAR signaling in axon-myelin synapses has been considered as a potential cause for the local detachment of myelin from axon, myelin blister formation that consecutively leads to axonal swelling, fragmentation, and demyelination [33]. In this regard, it is worth mentioning that a demyelinating disease can manifest along with anti-NMDAR encephalitis [7,44,45] and may share the hallmarks of classic multiple sclerosis lesions without complement deposition [46]. Accumulating clinical evidence indicate the presence of anti-NMDAR AABs in some patients with multiple sclerosis [47,48]. A recent study involving 200 patients revealed an overlap between anti-NMDAR encephalitis and MS (the German Network for Research on Autoimmune Encephalitis (GENERATE); in preparation). Nevertheless, it remains largely unclear whether antibodies against NMDARs or other neuronal surface structures [49] are causative for axonal pathology and demyelination in vivo. [56,83], autism spectrum disorder [84], and Alzheimer's disease [11,85,86].
Generation of Anti-NMDAR/GluN1 IgG1 Antibody
The generation and functional characterization of a recombinant IgG1 monoclonal antibody (SSM5) derived from intrathecal plasma cells of a patient with anti-NMDAR encephalitis and a control isotype against an irrelevant Cancer/Testis (CT) antigen, NY-ESO-1 (12D7) have previously been reported [14]. The effect of the biological activity of SSM5 and 12D7 on NMDA-induced Ca 2+ responses in oligodendrocytes have previously been described [15].
Animals
All experimental procedures were conducted following the guidelines and protocols approved by the local animal welfare committee (Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen [LANUV]) under protocol number O74/08. Mice on C57BL/6 background were used for preparation of cell cultures.
Myelinating Cell Cultures
The method of generating myelinated spinal cord neurons was essentially based on previous publications [17], with minor modifications. Briefly, dissection and meninges stripping of mouse spinal cords (12.5 day of gestation) was performed in ice-cold Dulbecco's Modified Eagle's Medium (DMEM; 1000 mg/L glucose, 11880028). Six cords were combined in 1 mL of Hank's Balanced Salts solution (HBSS; without Ca 2+ /Mg 2+ , 14175053) Figure 6. Graphic representation of the putative mechanisms involved in neurotoxic effects of human anti-GluN1 autoantibodies on myelinated neurons. Anti-GluN1 IgG1 may interfere with neurotransmission through NMDARs localized at different parts of neuron. Targeting NMDAR at axon-myelin synapses will affect glutamate-mediated localization of glucose transporter (GLUT1) to surface membrane of oligodendrocyte and diminish trophic support of the axon. Subsequently, the insufficient local ATP production by axonal mitochondria may trigger neurodegenerative processes at internodes. Anti-GluN1-mediated surface depletion of extra-synaptic NMDARs may affect inhibitory effects of basal NMDAR activity on expression of proapoptotic genes, such as Apaf1, Bid, Puma. Lowering synaptic NMDAR signaling may interfere with its known role in CREB-mediated neuronal survival, neuroprotective µ-calpain activation, antiapoptotic and antioxidant gene expression.
However, axonal injury following prolonged treatment with SSM5 might be initiated by a direct impairment of the NMDAR transmission in spinal cord neurons. Internalization of NMDARs following crosslinking with anti-GluN1 antibody decreases synaptic cluster density of the receptor and affects glutamate signaling [9][10][11]. Several lines of evidence indicate that NMDAR hypofunction (NRHypo) may play a detrimental role in neuronal survival and may lead to apoptotic neurodegeneration. Chronic low neuronal activity induced in cultivated neurons by tetrodotoxin or blockade of basal extra-synaptic NMDAR activity with AP5 antagonist initiates expression of pro-apoptotic genes that leads to increased susceptibility to neuronal damage [50][51][52] (Figure 6). Synaptic NMDAR activity is coupled with the expression and recycling of several antioxidant enzymes [53,54] (Figure 6), while NRHypo may lead to neuronal injury upon oxidative damage [55,56]. Consistently, genetic ablation of GluN1 from cortical GABAergic neurons initiated increased formation of reactive oxygen species in cortical neurons [57]. Particularly relevant for the naturally occurring neurodegenerative process in long-term cultures [58], it was shown that synaptic NMDAR signaling activates µ-calpain and protects cortical neurons against starvation and oxidative stress-induced damage [59]. Transient blockade of NMDAR activity during early stages of development leads to widespread neurodegeneration and neuronal apoptosis [60]. Alternatively, NRHypo may lead to neuronal damage by altering the neural network circuit [12,61,62].
Effector mechanism of AABs can also be attributed to the antibody-dependent cellular cytotoxicity (ADCC) [63][64][65]. ADCC is a complement-independent mechanism involving Fc-receptor (FcR) activation by target-bound antibodies on the surface of effector cells, such as microglia. Microglial cells invade the spinal cord parenchyma at E11.5 [66] and take part in the early development of neuronal network, including the activity-dependent refinement of myelination via myelin sheath phagocytosis [67]. Myelinating co-cultures derived from the embryonic spinal cord comprise microglia/macrophages that efficiently recapitulate innate CNS immune properties [19]. The engagement of Fc-receptor (FcR) signaling on microglia by antigen-bound immunoglobulins triggers inflammatory [68] and proliferative [69] responses in the CNS. In cultured myeloid cells, Fc-driven reaction initiates innate effector responses including phagocytosis, inflammatory cytokine release and ADCC independent of complement activation [70][71][72][73]. Therefore, it would be of interest to determine the potential effect of anti-GluN1 on activation of microglia.
Neural cells are able to produce virtually all components of a complement system [74,75] that upon activation, may direct AABs' pathogenicity via complementdependent cytotoxicity (CDC). Indeed, complement activation is recognized as a major pathogenic or contributing factor in neuromyelitis optica spectrum disorders (NMOSDs) and myasthenia gravis (MG), where AABs form stable surface complexes on aquaporin-4 (AQP4) or acetylcholine receptors (AChRs), respectively [76][77][78]. In contrast, engagement of NMDAR upon cross-linking with anti-GluN1 lead to rapid internalization and receptor depletion from the cell surface, making it unlikely the antibodies will form IgG clusters and triggering subsequent complement activation. In support of this, histological examinations in anti-NMDAR encephalitis demonstrated prominent microgliosis but showed no evidences of complement deposits or complement-mediated tissue injury [79,80].
Our data expand the list of encephalitis-derived autoantibodies that mediate neurodegenerative changes in cultured neurons such as anti-GluR3B IgG [81] and anti-IgLON5 [31,32], and support the causative role of autoantibodies in autoimmune encephalitis. The results reported herein are of particular relevance for neurodevelopmental alterations caused by maternal-to-fetal transfer of NMDAR antibodies [82]. We propose that anti-GluN1 IgG1-induced neurotoxicity is associated with NMDAR hypofunction and disturbance of axon-myelin communication. The results of this study are relevant for other CNS disorders with similar neuropsychiatric manifestations, where NRHypo state is a critical component of the pathophysiology, such as schizophrenia [56,83], autism spectrum disorder [84], and Alzheimer's disease [11,85,86].
Generation of Anti-NMDAR/GluN1 IgG1 Antibody
The generation and functional characterization of a recombinant IgG1 monoclonal antibody (SSM5) derived from intrathecal plasma cells of a patient with anti-NMDAR encephalitis and a control isotype against an irrelevant Cancer/Testis (CT) antigen, NY-ESO-1 (12D7) have previously been reported [14]. The effect of the biological activity of SSM5 and 12D7 on NMDA-induced Ca 2+ responses in oligodendrocytes have previously been described [15].
Animals
All experimental procedures were conducted following the guidelines and protocols approved by the local animal welfare committee (Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen [LANUV]) under protocol number O74/08. Mice on C57BL/6 background were used for preparation of cell cultures.
Treatment of Cells Grown on Multiwell Plates
To minimize 'edge-effects' due to the increased rate of evaporation or warming of media, wells on the outer edges of the plate were filled with HBSS, while cells were grown in the wells of the middle part of the plate. The effect of NMDA (100 µM; solubilized in water, Tocris Bioscience, Cat. No. 0114), dimethylsulfoxid (DMSO, 0.01% v/v, Sigma-Aldrich, D8418), triiodothyronine (T3, 1 µM, solubilized in DMSO, Sigma-Aldrich, T6397), SSM5 (10 µg/mL) and 12D7 (10 µg/mL) in myelinating cultures was investigated between DIV 21-28, DIV 28-35 and DIV 35-42. In order to avoid potential effects related to the location of the wells on the microplates, substances were added to the wells following a pre-generated random pattern that varied from one experimental repeat to the next. Four technical replicates per condition were used for each of four independent culture preparations. Treatments were performed three times a week by replacing half the medium.
Image Acquisition and Analysis
The Operetta CLS High Content Analysis System (PerkinElmer, Waltham, MA, USA) was used to acquire multiplexed wide-field fluorescent images in 96-well plate format with a 20 X objective, with the following channels 490/20 nm and 525/36 nm (Cy2, Alexa Fluor 488), 579/24 nm and 624/40 nm (Cy3) and 350/50 nm and 455/50 nm (Hoechst 33258, Thermo Fisher Scientific, Waltham, MA, USA). Images were captured from 25 fields of view per well and myelin (area stained with anti-MBP) and neurites (area stained with anti-NF200) were quantified using CellProfiler software [87,88] version 2.1.0. The pipelines used are based on dapi.cp and myelin.cp (last indexed on 24 March 2021) available at [URL]. Briefly, images were coded with parental metadata (imageID, wellID) and with row/column metadata. Quality control of the images was based on cell numbers (area stained with Hoechst) and artefacts. Shape was used to segment a closely-spaced cell. Total area, neurite area and myelin area measurements were exported to a comma-delimited spreadsheet for data analysis. The average value obtained from all images per well from each of the treatment conditions, from a single independent cell culture, were considered as one independent experimental unit.
For analysis of the neuronal injury and degeneration, the number of neurites that contained multiple or localized swelling (blebbing) or fragmentation of the neuronal process was counted manually using Adobe Photoshop software and expressed as a percentage of all neurites per automated optical inspection (AOI). The experimenter was blinded to the treatment condition during cell quantification.
Coverslips were imaged using an Olympus BX51 fluorescence microscope (Olympus France SAS, Rungis, France) and monitored by CellˆA imaging software (Soft Imaging System GmbH, Münster, Germany) and ImageJ, a free public-domain software developed by the National Institutes of Health ( [URL]). Detection of SSM5 and 12D7 immunoreactivity was performed on images acquired using a Leica SP8 confocal laser-scanning microscope (Leica, Wetzlar, Germany) and analyzed with LAS X software (Leica Application Suit X analysis tools (v. 3.7.4.23463).). Images were contrast-enhanced using Adobe Photoshop software (version CS3, Adobe Systems Ltd., Europe) to facilitate visibility in composite figures.
Statistical Analysis
Statistical analysis was performed using Prism software (GraphPad Software, San Diego, CA, USA). Statistically significant differences in myelination between multiple time points during culture cultivation and multiple treatments groups were determined by using one-way analysis of variance (ANOVA) followed by Tukey's Honest Significant Difference (HSD) post hoc test. Data are represented as mean ± standard error of mean (SEM). In all experiments a p value: * < 0.05; ** < 0.01; *** < 0.001 was defined as statistically significant. Each n-value reported are derived from a single independent cell culture, comprising multiple wells or coverslips, representing technical replicates of the various treatments.
==
Domain: Psychology Biology
|
[{"added": "2022-12-25T16:11:07.389Z", "created": "2022-12-23T00:00:00.000", "id": "255086984", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/1422-0067/24/1/248/pdf?version=1671786282", "pdf_hash": "1498fdd7ecf7129aa81d762b7afa205513649a6e", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:38691", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "7ca529035da2ecd7eeda29a46d7ecca66ddc939e", "year": 2022}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Plasma BDNF is a more reliable biomarker than erythrocyte omega-3 index for the omega-3 fatty acid enrichment of brain
Enriching brain DHA is believed to be beneficial for the prevention and treatment of several neurological diseases, including Alzheimer’s disease. An impediment in assessing the effectiveness of the treatments is the lack of a reliable biomarker for brain DHA. The commonly used erythrocyte omega-3 index is not suitable for brain because of the involvement of unique transporter at the blood brain barrier (BBB). We recently showed that dietary lysophosphatidylcholine (LPC)-DHA significantly increases brain DHA, which results in increase of brain BDNF. Since there is bidirectional transport of BDNF through the BBB, we tested the hypothesis that plasma BDNF may be used as biomarker for brain DHA enrichment. We altered the brain DHA in rats and mice over a wide range using different dietary carriers of DHA, and the correlations between the increase in brain omega-3 index with the increases in plasma BDNF and the erythrocyte index were determined. Whereas the increase in brain omega-3 index positively correlated with the increase in plasma BDNF, it negatively correlated with the erythrocyte index. These results show that the plasma BDNF is more reliable than the erythrocyte index as biomarker for assessing the effectiveness of omega-3 supplements in improving brain function.
The brain contains a very high concentration of the essential omega-3 fatty acid (FA) docosahexaenoic acid (DHA), which plays an important role in the normal development and function of the brain. Deficiency of DHA is associated with several neurological diseases, including Alzheimer's, schizophrenia, Parkinson's, and major depressive disorder [1][2][3] . Furthermore, epidemiologic 4 and pre-clinical studies [5][6][7][8] show beneficial effects of dietary omega-3 FA in the prevention and management of these diseases. Therefore, nutritional supplements such as fish oil are widely used in order to increase brain DHA with a hope to prevent these diseases or mitigate their effects. Although some beneficial effects have been reported [9][10][11] , majority of the controlled clinical trials using the currently available supplements failed to show improvement in Alzheimer's disease [12][13][14] , Huntington's disease 15 , or schizophrenia 16 . A possible reason for the failure of these trials is that the supplements do not significantly enrich brain DHA at clinically relevant doses, and therefore it is necessary to measure the brain DHA levels in order to test their effectiveness. Since a direct measurement of brain DHA is not possible, reliable non-invasive biomarkers are needed to determine the brain enrichment. Currently, the most widely used biomarker is the percentage of eicosapentaenoic acid (EPA) + DHA in the erythrocyte membrane lipids (omega-3 index) 17 . The basis for using this biomarker is the epidemiologic data showing that the dietary intake of omega-3 FA is positively correlated with the changes in the erythrocyte omega-3 index 18 . Furthermore, the increase in omega-3 FA of erythrocytes, following fish oil feeding correlated positively with the changes in brain DHA content in aged rats 19 , as well as neonatal baboons 20 . In contrast, other studies reported no positive correlation between erythrocyte DHA and brain DHA in swine which were fed fish oil 21 or in weanling rats fed alpha linolenic acid 22 . The mechanism of uptake of DHA and EPA by the brain is unlike the uptake by the systemic tissues because of the involvement of a transporter at the blood brain barrier which is specific for the lysophosphatidylcholine (LPC)-form of DHA 23 , whereas most other tissues obtain their omega-3 FA via lipoprotein uptake or by exchange with plasma lipids. Therefore, the enrichment of brain DHA may not correlate with that of other tissues, including the erythrocytes. We recently demonstrated that the brain DHA can be increased by up to 100% 24,25 . Whereas triacylglycerol (TAG)-DHA had minimal effect on brain DHA, di-DHA PC (phosphatidylcholine) and LPC-DHA markedly and dose dependently increased the DHA in all regions of the brain 25 . Furthermore, brain BDNF levels were increased significantly in proportion to the increase in DHA 24 . Since there is a bidirectional transport of BDNF across the BBB 28 , we determined whether the increase in brain BDNF also results in an increase in plasma BDNF. As shown in Fig. 1, there was indeed a positive correlation between the increase in plasma BDNF and the increases in BDNF levels of cortex and hippocampus after treatment with various molecular carriers of dietary DHA. The absolute values of the BDNF (and the statistical significance determined by ANOVA) are shown in the insets. These results show that the changes in plasma BDNF levels reflect the changes in brain BDNF levels, as also reported by others 29 . Since the increase in brain BDNF is correlated with the increase in brain DHA 24 , we tested the hypothesis that plasma BDNF may be a valid biomarker for the changes in the brain DHA content. As shown in Fig. 2A, B, the increase in plasma BDNF correlated positively with the increase in DHA in both cortex and hippocampus over a wide range of values. The insets show absolute percentages of brain omega-3 FA (EPA + DHA) and the concentrations of plasma BDNF under various dietary conditions. We have also determined the correlation between the increases in brain DHA levels and erythrocyte omega-3 index, the most commonly used biomarker for measuring the incorporation of dietary omega-3 FA into brain and other tissues 18,30 . As shown in Fig. 2C, D, the increases in erythrocyte omega-3 levels were actually negatively correlated with the increases in cortex or hippocampus omega-3 levels. This is due to the fact that TAG-DHA significantly increased the erythrocyte omega-3 FA without appreciably increasing the brain omega-3 FA. On the other hand, PC-DHA and LPC-DHA which increased the brain omega-3 FA, had only modest effect on erythrocytes. These results therefore show that the erythrocyte omega-3 index is not a suitable marker for the changes in brain omega-3 FA altered by dietary lipids. Correlation of plasma BDNF levels with brain BDNF in rats. Two month old rats were gavaged daily with the indicated DHA-compounds (40 mg DHA/kg body weight) for one month, and the BDNF levels in plasma and brain regions were determined by ELISA. Two doses of LPC-DHA (5 mg and 10 mg) equivalent to 20 mg DHA and 40 mg DHA/kg body weight respectively were used. The insets show the absolute values (mean ± SD, n = 10 rats/group) of BDNF in the control (untreated) and DHA-treated groups. Bars of the same color without common superscripts are significantly different from each other (one-way ANOVA, with Tukey multiple comparison correction). The increase in BDNF by DHA treatment was calculated by subtracting the average of the control values from the individual samples of the treated groups. Pearson correlation was calculated between the increase in plasma BDNF vs the increase in cortex or hippocampus (Graphpad, Prism 8.0).
It should be pointed out that the negative correlation observed above does not mean a reciprocal relationship between erythrocyte omega-3 and brain omega-3. Instead, this is due to the divergent mechanisms of uptake by the two tissues. Brain acquires its DHA through the Mfsd2a transporter pathway which prefers LPC-DHA over other forms of DHA 23 , whereas the DHA uptake by the erythrocytes is most likely through the exchange with plasma lipids. Therefore, it is possible that the erythrocyte index may not reflect the brain index, but could reflect the uptake of DHA by other peripheral tissue that acquire DHA through non-Mfsd2a pathways, including uptake of free FA through diffusion, and the receptor-mediated uptake of lipoproteins. To investigate this, we determined the correlation of changes in erythrocyte omega-3 index with changes in this index of other tissues. In addition, we determined the correlation of the omega-3 index of these tissues with plasma BDNF. As shown in Fig. 3, the liver omega-3 index was negatively correlated with the erythrocyte index, but positively correlated with the increase in plasma BDNF, similar to the brain. In contrast, the changes in erythrocyte index were positively correlated with the changes in heart (Fig. 4A) and adipose tissue (Fig. 4C). In both these tissues, TAG-DHA was more efficient than LPC-DHA or PC-DHA in increasing the omega-3 FA 25 . The plasma BDNF changes, on the other hand, were negatively correlated with the changes in omega-3 index of adipose tissue (Fig. 4B) as well as heart (Fig. 4D). In conclusion, these results show that the erythrocyte index, which has been widely used as a surrogate for the tissue incorporation of dietary omega-3 FA, reflects only selected tissues such as adipose tissue and the heart, but not the brain or liver.
Studies in normal mice; effect of dietary free (unesterified) DHA versus LPC-DHA.
Although some previous studies suggested that BDNF is absent in mouse plasma 29,31 , more recent studies showed the presence of measurable amounts 32,33 . We previously showed that while dietary free DHA did not appreciably www.nature.com/scientificreports/ increase brain DHA, LPC-DHA (both sn-1 acyl and sn-2 acyl isomers) markedly increased brain DHA, as well as BDNF, and improved brain function in normal male mice 24 . We now determined whether the increase in brain DHA by LPC-DHA resulted in an increase of plasma BDNF in mouse plasma also. As shown in Fig. 5A, B (insets), free DHA did not increase plasma BDNF levels compared to controls, whereas both isomers of LPC-DHA significantly increased it in cortex as well as hippocampus. There was a positive correlation between the increase in plasma BDNF and the increase in brain BDNF (Supplementary Figs. 1 and 2 online). The increase in plasma BDNF above the control value correlated positively with the increase in omega-3 index in both the brain regions. In contrast, the increase in erythrocyte index was negatively correlated with the increases in cortex and hippocampus (Fig. 5C, D). These results are similar to those obtained in rats, and thus show that plasma BDNF is a valid marker for changes in brain omega-3 FA levels not only in rats but also in mice.
Effect of dietary LPC-EPA in mice.
Whereas previous studies reported that brain EPA levels cannot be increased through diet 34,35 , we have demonstrated that feeding LPC-EPA to normal mice not only increases brain EPA levels by several fold, but also increases brain DHA by about 100% in normal mice 36 . We determined whether the increase in plasma BDNF can be used as a biomarker for the increase in brain EPA and DHA after feeding free and LPC-EPA. As shown in Fig. 6, the increase in plasma BDNF above the average of control values positively correlated with the increase in the brain omega-3 index. However, the increase in erythrocyte omega-3 index was negatively correlated with that of brain because free EPA increased the omega-3 content of erythrocytes but not the brain, similar to the effects of TAG-DHA. The insets show the absolute values for all groups, including the controls. The increase in plasma BDNF also correlated positively with its increase in the brain ( Supplementary Fig. 3 online). These results show that the effect of feeding LPC-EPA on plasma BDNF are similar to those of feeding LPC-DHA.
Studies with lipase treated krill oil and fish oil. We recently showed that the brain omega-3 index can be significantly increased by feeding krill oil which has been pre-treated with a lipase (thus generating LPC-EPA and LPC-DHA), but not by similarly treated fish oil, which cannot generate LPC 37 . Since many of the clinical studies are carried out with fish oil or krill oil, we determined whether the plasma BDNF can be used as surro- www.nature.com/scientificreports/ gate for brain omega-3 index in the mice treated with the lipase-modified and unmodified krill oil and fish oil. As shown in Fig. 7A, B (insets), only the lipase-treated krill oil significantly increased the omega-3 FA content in both cortex and hippocampus. The increase in plasma BDNF correlated with its increase in cortex and hippocampus (Supplementary Figs. 4 and 5 online). Furthermore, the increase in plasma BDNF above the control value positively correlated with the increases in the omega-3 indexes of cortex and hippocampus. On the other hand, the increase in erythrocyte index was negatively correlated with the increases in the indexes in cortex and hippocampus (Fig. 7C, D). These results are similar to the results obtained with pure LPC-DHA or LPC-EPA in the mice.
Discussion
Although there are several nutritional supplements of omega-3 FA in the market claiming to improve brain function and to protect against neurological diseases, controlled clinical trials supporting these claims are lacking. While some studies did report positive results 9-11 many other studies reported negative results in improving brain function and memory [12][13][14] . An impediment for testing the effectiveness of the various supplements in humans is the lack of a reliable biomarker for the enrichment of brain omega-3 FA in response to them. Although the erythrocyte omega-3 index has been effectively used to evaluate the cardiovascular benefits of the omega-3 supplements 38 , the utility of this index in determining the effectiveness of these supplements for brain enrichment has not been demonstrated. In fact, the study by Fenton et al. 17 , which showed positive correlation of the erythrocyte omega-3 index with most other tissues excluded brain and liver, the two most important tissues relevant to the omega-3 FA function in the brain. Previous studies by Berliner et al. 21 showed no significant correlation between the DHA concentration of erythrocyte membranes and that of brain membranes in miniature swine fed menhaden oil. Similarly, Tu et al. 22 reported that after feeding α-linolenic acid to weanling rats, the erythrocyte omega-3 index correlated positively with that of most other tissues, but not the brain. Some epidemiologic studies also showed no correlation between erythrocyte index and depression 39 or white matter hyperintensity 40 . Many experimental studies in animals, on the other hand, have reported a positive correlation between erythrocyte omega-3 index and that of the brain, but the range of brain DHA values achieved in these studies was narrow, since TAG-omega-3, which does not efficiently enrich brain omega-3, was used for feeding 19,22,41 . Interestingly we also found a positive correlation between the increase in erythrocyte and brain indexes if we plot only the values of the rats fed TAG-DHA ( Supplementary Fig. 6 online). However, this correlation turned negative when the effects of PC-DHA and LPC-DHA are included, since the latter induce a much greater increase in brain www.nature.com/scientificreports/ DHA without a concomitant effect on erythrocytes (Fig. 2C, D). Therefore, the negative correlation does not mean that there is a reciprocal relationship between the erythrocytes and the brain, but instead is indicative of the divergent incorporation profiles of TAG-DHA and LPC-DHA. Whereas DHA from dietary TAG is incorporated significantly into erythrocytes, it is inefficient in enriching brain DHA. In contrast, LPC-DHA efficiently increased brain DHA by up to 100%, without significantly altering erythrocyte levels.
The current study makes a strong case for the plasma BDNF as a reliable non-invasive biomarker for the increase in brain omega-3 FA levels in response to treatments in patients. BDNF is an important neurotrophin with a role in neurogenesis, neuronal survival, learning and memory, as well as in regulation of body weight and energy homeostasis 42 . Plasma levels of BDNF are significantly decreased in patients with psychiatric disorders 43 , and are increased after treatment with anti-depressants 44 , as well as high doses of omega-3 FA 45 . Plasma BDNF is also significantly increased after vigorous exercise 46 , which is further enhanced by feeding DHA 27 . Importantly, it has been shown that there is a bidirectional transfer of BDNF between the brain and the plasma 27,28 , and that up to 80% of BDNF in the plasma may be derived from the brain 46 . There is convincing evidence that DHA increases the expression of BDNF in the brain possibly through the activation of Akt 27 or GPR40 47 . Many of the beneficial effects of DHA may be through the expression of BDNF. Therefore, there is a physiological basis for using the plasma BDNF as a functional surrogate for brain omega-3 FA levels, unlike the erythrocyte omega-3 FA levels, which are metabolically unrelated to the brain levels.
In addition to the brain omega-3 FA status, the plasma BDNF may be a reliable marker for the omega-3 FA level of the liver, which does not correlate with the erythrocyte omega-3 index. There is ample evidence from experimental studies that high dietary omega-3 FA diets are beneficial in the treatment of fatty liver 48 , but mixed results were obtained in the clinical trials. Measuring plasma BDNF, which correlates positively with the www.nature.com/scientificreports/ hepatic enrichment of omega-3 FA would be helpful in determining the effectiveness of the treatments with nutritional supplements. It may be pointed out that the BDNF concentration of serum is much higher than the plasma, since large amounts of BDNF are released during the activation of platelets 49 . Therefore, it is important to measure the BDNF levels in the plasma, not in the serum, for more accurate reflection of the brain DHA levels. Another important consideration is that since BDNF expression and its plasma levels are also increased by vigorous exercise 46 and anti-depressant treatments 44 , such factors should be controlled for, if present, in order to specifically measure the effects of DHA. For example, any exercise regimen and anti-depressant therapy should be continued as usual during omega-3 FA treatment, and plasma BDNF should be measured before and after the treatment period.
Materials and methods
Animals and dietary treatments. Most of the analyses were carried out on samples obtained from our studies published previously 24,25,36,37 . All animal protocols were approved by the UIC institutional animal care committee, and all methods were carried out in accordance with the relevant guidelines and regulations. Male Sprague-Dawley rats (8 week old) were purchased from Harlan laboratories (Indianapolis, IN). Male c57BL/6 mice (2-4 months old) were purchased from Jackson Laboratories (Bar Harbor. Maine).
In study 1, the rats were gavaged daily with 10 mg of DHA in the form of TAG-DHA, di-DHA PC, or 5 and 10 mg of DHA in the form of LPC-DHA for 30 days 25 . In study 2, male mice (4 month old) were gavaged daily with 40 mg DHA/kg body weight in the form of free DHA, sn-1 acyl LPC-DHA, or sn-2 acyl LPC-DHA for 30 days as described previously 24 . In study 3, male mice (2 month old) were gavaged daily with 40 mg EPA/kg body weight in the form of free EPA or LPC-EPA for 15 days, as described previously 36 . In study 4, male mice www.nature.com/scientificreports/ (2 month old) were fed diets enriched with natural or lipase-treated fish oil or natural or lipase-treated krill oil for 30 days. The FA composition of most of the tissues, determined by GC/MS, has been presented in our previous studies 24,25,36,37 . In addition, we analyzed the FA composition of the erythrocytes in all animals by GC/MS for this study. The values of EPA and DHA (percentage of total) were combined to give the omega-3 index of the tissues.
Analytical procedures. The FA analysis of tissues was carried out by GC/MS as described previously 24 .
BDNF in plasma and brain regions was assayed by ELISA, using Promega Emax Immunoassay system kit (Promega Inc., Madison, WI, USA), according to the manufacturer's protocol. Rat brain regions (Cortex and hippocampus) were homogenized in the lysis buffer, and the homogenates were centrifuged at 10,000×g, for 20 min. The supernatants were collected and used for the quantification of BDNF levels.
Statistics and correlations. The significance of differences between treatment groups was determined by one-way ANOVA, with Tukey post hoc multiple comparison corrections. For each study, the average of control values (untreated group) was first calculated. This average was then subtracted from individual values of the treatment groups to calculate the increases in omega-3 FA of tissues or plasma BDNF due to the treatment. The increases in omega-3 FA in the brain and other tissues were plotted against the increases in plasma BDNF or erythrocyte omega-3 of each animal to determine the Pearson correlation coefficients (Graphpad Prism 8.0). Figure 7. Correlation of brain omega-3 index with plasma BDNF and erythrocyte omega-3 index in mice fed krill oil or fish oil. Natural or lipase-treated fish oil and krill oil were incorporated into AIN-93G diet to provide 2.64 g of EPA + DHA per kg diet. These diets were fed to normal male mice for 30 days, and the tissue FA composition as well as plasma BDNF contents were measured. The top 2 panels (A, B) show the correlation of the increase in omega-3 indexes of cortex and hippocampus with the increase in plasma BDNF levels, whereas the bottom 2 panels (C, D) show the correlation of the increases in cortex and hippocampus omega-3 indexes with that of erythrocytes. The insets show the absolute values (mean ± SD, n = 5 mice/group) of omega-3 indexes (% of EPA + DHA) and the plasma BDNF levels (pg/ml) for all groups including the controls (which were fed unsupplemented AIN-93G diet). Bars of same color with different superscripts are significantly different from each other by one-way ANOVA.
==
Domain: Psychology Biology
|
[{"added": "2023-02-20T15:04:26.700Z", "created": "2020-07-02T00:00:00.000", "id": "257026520", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/s41598-020-67868-9.pdf", "pdf_hash": "40c9b9f912438bb20afd692717b835f014ab7d7b", "pdf_src": "SpringerNature", "provenance": "peS2o-0075.json.gz:33125", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "40c9b9f912438bb20afd692717b835f014ab7d7b", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Reactive gliosis and neuroinflammation: prime suspects in the pathophysiology of post-acute neuroCOVID-19 syndrome
Introduction As the repercussions from the COVID-19 pandemic continue to unfold, an ever-expanding body of evidence suggests that infection also elicits pathophysiological manifestations within the central nervous system (CNS), known as neurological symptoms of post-acute sequelae of COVID infection (NeuroPASC). Although the neurological impairments and repercussions associated with NeuroPASC have been well described in the literature, its etiology remains to be fully characterized. Objectives This mini-review explores the current literature that elucidates various mechanisms underlining NeuroPASC, its players, and regulators, leading to persistent neuroinflammation of affected individuals. Specifically, we provide some insights into the various roles played by microglial and astroglial cell reactivity in NeuroPASC and how these cell subsets potentially contribute to neurological impairment in response to the direct or indirect mechanisms of CNS injury. Discussion A better understanding of the mechanisms and biomarkers associated with this maladaptive neuroimmune response will thus provide better diagnostic strategies for NeuroPASC and reveal new potential mechanisms for therapeutic intervention. Altogether, the elucidation of NeuroPASC pathogenesis will improve patient outcomes and mitigate the socioeconomic burden of this syndrome.
Introduction: As the repercussions from the COVID-pandemic continue to unfold, an ever-expanding body of evidence suggests that infection also elicits pathophysiological manifestations within the central nervous system (CNS), known as neurological symptoms of post-acute sequelae of COVID infection (NeuroPASC). Although the neurological impairments and repercussions associated with NeuroPASC have been well described in the literature, its etiology remains to be fully characterized.
Several theories have been proposed to explain these neurocognitive symptoms, including inflammatory changes, hypoxia, coagulopathy, vascular endothelial, dysfunction and direct viral invasion of the neurological tissue (22). Although the precise mechanisms remain elusive, six mechanisms have been proposed: i) systemic immune response-mediated neural dysregulation; ii) direct CNS invasion; iii) auto-immune responses; iv) latent pathogen reactivation; v) cerebrovascular thrombosis; and vi) multi-organ dysfunction (23). In this mini-review, we have highlighted the leading hypotheses and pathological mechanisms supporting NeuroPASC, through the consequential disturbance of reactive microglia and astroglia, which lead to persistent neurocognitive symptoms of PASC.
. . Glial cell reactivity
Maintenance of optimal cognitive function is a complex process that requires coordination between neuron function and glial cells (24). In recent years, significant interest has been allocated to glial cell (i.e., microglia, astrocytes, and oligodendrocytes) dysfunction during cognitive impairment. In fact, the dysregulation of glial cell function leads to cognitive impairment associated with numerous neuropathologies, including metabolic syndromes (24) and neurodegenerative diseases such as Parkinson's (25) and Alzheimer's (26) diseases. Microglia, the resident immune phagocytes of the CNS, are essential for learning, memory, and behavior regulation in the adult brain (27). In addition to immune surveillance and phagocytosis, microglia are also responsible for other crucial functions in the CNS, including synaptic pruning and synaptogenesis, axon fasciculation and neurite formation, programmed cell death, astrocyte activation and proliferation, and oligodendrocyte differentiation and myelogenesis (27) (Figure 1). Based on the concept of cellular polarization, cells were separated into two phenotypically distinct sub-populations characterized by opposing effects on the CNS. Specifically, the classical (M1) microglial subset was believed to be responsible to produce pro-inflammatory mediators, which induced inflammation and neurotoxicity. Conversely, M2 was assumed to release antiinflammatory factors, which confer neuroprotectivity. With the advent of technology, M1 and M2 microglia are portrayed as brute oversimplifications to illustrate antagonistic states in both healthy and diseased brains (28). Microglia are likely to be significantly more complex as microglial subset identity and function are intricately regulated by microglial metabolic states and the environmental profiles of signaling mediators (e.g., cytokines and neurotransmitters) (24).
Complex microglial-astrocyte interactions also form a delicate equilibrium in CNS health. Indeed, cellular dysfunction from either cell population or the maladaptive synergistic interactions between microglia and astrocytes can result in neurotoxicity and alter synaptic plasticity through numerous mechanisms (29, 30). With a crucial role in brain homeostasis, astrocytes regulate CNS blood flow, glucose metabolism, and the recycling of neurotransmitters (24). Astrocytes are also depicted as master regulators of synaptic activity by controlling synaptic junction plasticity and mediating synapse elimination to avoid excitotoxicity (31,32). Reminiscent of microglia's obsolete nomenclature, astrocytes are classified into two distinct sub-populations (A1 and A2) based on their reactivity and function (30). On the one hand, A1 reactive astrocytes produce pro-inflammatory soluble mediators, which are mainly induced by the NF-κB signaling cascade (33). On the other hand, A2 reactive astrocytes generate anti-inflammatory mediators and many neurotrophic factors induced by STAT3 activation. As a result, reactive A1 astrocytes provoke neurotoxicity and neuronal death, whereas A2 astrocytes promote survival and neuron growth (33).
Upon cerebral insult, astrocytes undergo drastic phenotype change referred to as reactive astrocytosis, induced due to an upregulation of pro-inflammatory cytokines by neuroinflammatory microglia such as (interleukin) IL-1α, IL-1β, tumor necrosis factor-alpha (TNF-α), and the complement component 1q (C1q) (29, 30, 33). As a result, neurotoxic A1 reactive astrocytes display decreased function in synaptic formation and phagocytic capability. Furthermore, A1 reactive astrocytes promote significant neurotoxicity, which leads to cell death of cortical neurons and mature differentiated oligodendrocytes (30,34). Moreover, inflammatory microglia further accentuate NF-κB signaling, leading to A1 astrocyte population remodeling and neurodegeneration (35). A study by Saggu et al. (36) has shown that astroglial-mediated NF-κB activation is associated with white matter damage and cognitive impairments in vascular dementia models (36). While microglial activation alone is insufficient to initiate cell death in the CNS, microglial activation potentially enhances neurological damage by inducing reactive astrocytosis, resulting in neurodegeneration (30).
As for oligodendrocytes, they are responsible for axonal myelination, which regulates action potential conduction velocity, essential for neural circuit dynamics (37). Oligodendrocytes are also important contributors to neurodegenerative diseases including Alzheimer's disease, amyotrophic lateral sclerosis, and multiple system atrophy. More recently, studies have shown that, in addition to myelination, oligodendrocytes are required for the integrity and survival of axons independent of myelin itself (38). Mechanistically, oligodendrocytes foster glycolytic metabolism, which provides axons with energy-rich metabolites.
Altogether, the coordinated signaling between microglia, astrocytes, and oligodendrocytes is essential for homeostasis and CNS health.
. SARS-CoV--mediated activation of glial cells . . Indirect pathway: peripheral immune cell activation and CNS infiltration Acute and chronic CNS inflammation alike have drastic repercussions on glial circuitry and cytokine expression profiles, which result in dysfunctional immune signaling and synaptic plasticity (39). As a result of the intricate equilibrium that composes glial cell homeostasis, various neuroinflammatory states including chemotherapy (40) and notably COVID-19 infection (41), disrupts glial lineage, pertaining to glial population proliferation, differentiation, and maturation.
Following COVID-19 infection, an upregulation of proinflammatory chemokine-enhanced microglial populations and an impairment of oligodendrogenesis in mice models led to neurological disturbance in the absence of direct viral invasion (41).
Neuroinflammation underlies one of the leading theories to explain CNS injury during SARS-CoV-2 infection and is a consequence of the well-documented systemic cytokine storm and subsequent increase in blood-brain barrier (BBB) permeability (14,42). Through its spike surface glycoprotein, SARS-CoV-2 enters the host cells by binding to its angiotensin-2 converting enzyme (ACE-2) receptors, which consequently initiates an important inflammatory response (13,43). Brainblood barrier disruption from systemic inflammation facilitates neuroinflammation through neural invasion of inflammatory cytokines, which further stimulates cytokine secretion from the microglia (42). Accordingly, a study in rats has shown that exposure to a partial subunit of the SARS-CoV-2 spike protein (i.e., S1 protein subunit) elicits innate immune response through a pathogen-associated molecular pattern (PAMP), which triggers microglial activation and neuroinflammation in the absence of active virions (44). The S1 spike protein also activates the NRLP3 inflammasome that plays a pivotal role in innate immunity and inflammatory signaling triggered by PAMPs (45). This pathway leads to NF-κB activation, pro-inflammatory cytokine production (i.e., IL-1β and IL-18), and subsequent glial reactivity, all of which are associated with neurodegenerative diseases (46). Meanwhile, microglial activation via NF-κB signaling induces reactive astrocytosis, which in turn leads to excitotoxicity, white matter damage, and loss of myelin plasticity, in addition to oligodendrocyte and neuronal cell death (14,35,44).
Neuroinflammatory pathways that alter CNS homeostasis are linked to cognitive and neuropsychiatric complications (43). The systemic immune-inflammation index, which reflects the immune response and systemic inflammation based on a ratio of peripheral lymphocyte, neutrophil, and platelet counts (SII = platelets × neutrophils/lymphocytes), has been found to predict depressive symptomatology and cognitive dysfunction 3 months following initial infection (47). Even in the absence of direct CNS viral infiltration, consequential production of peripheral cytokine profiles associated with the host's antiviral response may be sufficient to induce neuroinflammatory reactions and/or compromise the integrity of the blood-brain interface. As a result, peripheral immune cells migrate through the BBB into the CNS and induce microglia-derived cytokines, which interfere with neurotransmission (14,42). These mechanisms have mostly been established using experimental models. For example, mild respiratory illness in AAV-hACE2 mice (48) following intranasal delivery of SARS-CoV-2 was sufficient to induce potent microglial reactivity in the sub-cortical white matter upon pathological examination of the mice brain tissue (41). Moreover, Klein et al. (49) compared the hamster models of SARS-CoV-2 to pathological specimens of human patients deceased from COVID-19, demonstrating similar pathological changes in the absence of viral neuroinvasion. These changes included abnormal BBB permeability, microglial activation, loss of hippocampal neurogenesis, and expression of IL-1β and IL-6 within sub-cortical structures (49).
Neuroinflammation during acute SARS-CoV-2 infection may consequently induce brain parenchyma and vessel alterations that further foster the inflammation of neurons and supportive cells (14). Additionally, such neuroinflammation could be a catalyst for microvascular thrombosis and ischemic brain injury during the COVID-19 infection (50). Magnetic resonance imaging (MRI) from a deceased COVID-19 patient revealed volumetric and micro-structural brain abnormalities, which were accompanied by several neuropathological lesions reminiscent of vascular and demyelinating etiology (51). Any combination of these events could lead to BBB disruption and subsequent immune cell infiltration of the CNS causing microglial activation and neuroinflammation in the absence of direct viral invasion of the CNS.
. . Direct pathways
Glial activation and neurotoxicity may result from the direct routes of SARS-CoV-2 infection. In a study where transgenic mice models expressing recombinant human ACE-2 were infected with SARS-CoV-2, investigators found viral particle (spike protein) infiltration within the CNS and an abundance of activated microglia in the proximity of the infected tissue (46). The utilization of human monocyte-derived microglia infected with SARS-CoV-2 revealed that viruses enter these cells through ACE-2 receptor binding in the absence of viral replication. More interestingly, they observed that the infected cells induced NLRP3 inflammasome activation and a potent pro-inflammatory response accompanied by IL-1β overexpression (46). Mechanistically, these neuroinflammatory events were shown to be NF-κB dependent as the utilization of NF-κB inhibitors led to complete inhibition of Il-1β release. Another study conducted by Samudyata et al. (52) established a brain organoid model with innately developing microglia (52). Such in vitro invasion assays on microglial cells co-cultured with SARS-CoV-2 demonstrate the loss of postsynaptic termini and neuronal cell death. Transcriptomic profiling of microglia exposed to SARS-CoV-2 revealed gene expression signatures that closely resembled neurodegenerative disorders (52). Nevertheless, it is worth noting that SARS-CoV-2 antigens and RNA have rarely been detected in the CSF of COVID-19 patients (53,54) while only detected in a minority of human brain autopsies (55). Heterogenous study results have resulted in controversy surrounding the neuroinvasive properties of SARS-CoV-2. This section will explore pathways by which CNS infiltration of SARS-CoV-2 of viral proteins may result in microglial activation and neuroinflammation during COVID-19 infection.
. . . Olfactory route
The presence of ACE-2 receptors along the olfactory tract suggests that the neurological manifestations of COVID-19 could be caused by direct neurological infiltration via the olfactory route (56, 57), a common entry site to several other respiratory viruses (58). CNS viral dissemination to the amygdala, hippocampus, and entorhinal cortex could then be possible through the connecting olfactory bulb, where SARS-COV-2 RNA has been found in approximately 20% of post-mortem brains from deceased COVID-19 patients (59). Numerous imaging studies also support this hypothesis (59)(60)(61). For example, neuroimaging from a cohort of 785 participants (including 401 participants scanned before and after COVID-19 infection) discovered significant longitudinal effects in SARS-CoV-2 cases including a decrease of thickness and tissue contrast from the orbitofrontal cortex and the parahippocampal gyrus gray matter, changes in tissue damage markers in olfactory cortex-related regions, and a global reduction in brain volume (61). Previously infected individuals from the latter cohort also demonstrated cognitive decline post-infection. Together, imaging data originating mainly from the limbic system could highlight COVID-19-mediated neurodegeneration through the olfactory pathways, neuroinflammatory events, and loss of sensory input caused by anosmia (61). Other imaging studies in COVID-19 patients using MRI cerebral imaging have enabled researchers to observe an increase in olfactory bulb signal intensity and volume size (60). Positron emission tomography (PET) has also shown reduced 18-fludeoxyglucose of orbitofrontal hypometabolism in patients with anosmia (62). Altogether, these findings suggest a role for imaging technologies in the detection and progression of direct neurological infiltration and pathogenesis of COVID-19 infection through the olfactory tract.
. . . Hematogenous spread and endothelial pathology
Perturbation of BBB permeability has been well documented during the infection of various respiratory viruses (63). Of note, cerebral endothelial cells, which comprise the BBB, are prone to SARS-CoV-2 infection through cell surface expression of receptors NRP1, BSG, and low levels of ACE-2 (64). Furthermore, SARS-CoV-2 has been shown to cross the BBB by transcellular pathways, accompanied by basement membrane disruption in mice models (65). As a result, vascular permeability increases and leads to perivascular cell infiltration and neuronal cell death. Wenzel et al. (64) have demonstrated brain endothelial cells infection; the expression of SARS-CoV-2 main protease (Mpro) cleaves the host protein NF-κB essential modulator (NEMO), which is an essential modulator of NF-κB-mediated survival (64). By ablating NEMO, M pro induces microvascular pathology, BBB disruption, endothelial cell death, and neuroinflammation. Similarly, ACE-2 (66) and NRP1 (67) receptors can be found in astrocytes, which are in direct contiguity with the BBB. Astrocyte infection by SARS-CoV-2 is further supported by the detection of the S1 spike gene transcripts and protein in the cerebral vasculature of COVID-19 patients (64) and the description of S1 spike-positive astrocyte in post-mortem human samples (67). Subsequently, in vitro neural stem cellderived human astrocytes were exposed to SARS-CoV-2, resulting in astrocyte infection through spike-NRP1 interactions (67). The resulting astrocyte phenotype decreased neuronal viability while promoting neuronal apoptosis (67).
Previous studies have also demonstrated the occurrence of neuropathological events mediated by the S1 protein of SARS-CoV-2. Accordingly, SARS-CoV-2 virions are known to spontaneously shed S1 protein subunits, which can be found in the plasma of COVID-19 patients (44,68). This pro-inflammatory protein has also been found in human cerebral endothelial cells upon autopsy in the absence of viral RNA and is strongly co-localized with inflammatory mediators including caspase-3, TNF-α and IL-6 (69). In mouse models, S1 spike protein injection leads to endothelial cell damage with increased expression of TNF-α and IL-6, which co-localized with the S1 spike subunit (69). Similarly, nonprimate models have demonstrated the presence of SARS-CoV-2 nucleocapsid protein in endothelial cells of the cerebral vasculature (70). Altogether, the expression of SARS-CoV-2 compatible receptors in cerebral structures, in addition to the discovery of SARS-CoV-2 genetic material and viral proteins in the endothelial tissue and astrocytes, suggests that viral invasion or viral protein infiltration of cerebral vasculature could be a mechanism that leads to microglial activation and neuroinflammation.
. . . Cerebrospinal fluid
Another proposed route for SARS-CoV-2 infection is through the cerebrospinal fluid (CSF). In a study utilizing humanpluripotent-stem-cell-derived brain organoids to examine SARS-CoV-2 neurotropism, ACE-2 positive choroid plexus epithelial cells were amenable to infection, which leads to an initial disruption of the blood-CSF barrier followed by a subsequent complete breakdown of barrier integrity (71). Infection of these organoids has been associated with transcriptional dysregulation and cell death, suggestive of a neuroinflammatory response and deficits in cellular functions (72). Although some studies have shown SARS-CoV-2 PCR positivity in patient's CSF samples, other studies have contradicted this notion (73). While the neuroinvasive properties of SARS-CoV-2 through the blood-CSF barrier have not been confirmed, a more likely mechanism involves barrier leakage, leading to the translocation of immune cells and cytokines that sustain neuroinflammation (71).
. Reactive gliosis as a culprit of NeuroPASC . . Current evidence of microglial reactivity in NeuroPASC
A recent study on AAV-hACE2 mice models with mild SARS-CoV-2 respiratory infection has demonstrated a prominent increase in pro-inflammatory cytokine and chemokine profiles (e.g., IFN-γ, IL-6, TNF-α, CXCL10, CCL7, CCL2, and CCL11) in the CSF and serum samples as rapid as 7 days post-infection (41). Longitudinal evaluation of pro-inflammatory mediators revealed that while serum levels of these mediators normalized after 7 weeks, there was a progressive increase of CSF cytokines/chemokines levels over time. Notably, CCL11, a cytokine associated with cognitive impairment (74), remained persistently elevated in the CSF over time, suggesting that isolated respiratory infection with SARS-CoV-2 can result in prolonged changes in CSF cytokine profiles, leading to persistent neuroinflammation (41). The latter study has also demonstrated that mice infected with SARS-CoV-2 displayed white matter microglial reactivity for at least 7 weeks, which culminated in oligodendrocyte death, axonal demyelination, and impaired mechanisms of cellular homeostasis and neuron generation in the hippocampus. These findings align with recent studies highlighting BBB disruption, microglial activation, aberrant cytokine expression, and suppression of hippocampal neurogenesis in brain samples from post-mortem COVID-19 patients (49). Moreover, Schultheiß et al. (75) demonstrated elevated serum cytokine profiles up to 8 months post-infection in a cohort of COVID-19 patients manifesting mostly mild-to-moderate infection severity (75). Interestingly, persistently elevated levels of serum IL-1β, IL-6, and TNF-α correlated with PASC symptoms of dyspnea, fatigue, and cognitive impairment. Further examination also suggested that these cytokines were constitutively secreted by resident monocytes/macrophages in the lungs (75). In parallel, a study by Peluso et al. (76) revealed that an increase in plasma IL-6, TNF-α and glial fibrillary acidic protein (GFAP), an axonal structural protein and biomarker of glial cell activation, predicts NeuroPASC symptoms in SARS-CoV-2 infected patients (76).
Cognitive dysfunction is also correlated with increased immunoregulatory pathway protein expression and a downregulation of inflammatory and antiviral response proteins (77). Moreover, individuals with NeuroPASC exhibit deficient systemic humoral immunity response to various SARS-CoV-2 antigens (Spike, S1, S2, RBC, and Nc) when compared to non-PASC COVID-19 control patients. Elevated levels of serum IgG specific to SARS-CoV-2 are associated with improved NeuroPASC clinical outcomes possibly due to enhanced viral clearance (78), while individuals who experience severe neurological injury following acute COVID-19 infection tend to elicit elevated levels of CSF SARS-CoV-2 specific antibodies (79). Distinct T-cell response and effector signatures in addition to unique CSF humoral responses highlight the significance of humoral immunity alterations and pathogenic outcomes of NeuroPASC (77, 78). Taking into consideration that mild respiratory infection and systemic inflammation can lead to BBB permeability disturbances combined with microglial reactivity (41), one could suggest that immunologic alterations (77, 78) and persistent systemic inflammation following COVID-19 (75) may be a catalyst for chronic neuroinflammation and glial reactivity in previously primed microglia.
. . Microglial priming and persistent neuroinflammation in NeuroPASC
Considering the detrimental role of persistent microglial reactivity in neurodegenerative diseases, such reactive states could also be key to NeuroPASC pathogenesis. Accordingly, a key concept in AD trajectory known as microglial priming is associated with aging and systemic inflammation (80). Fundamentally, microglia priming renders them more susceptible to secondary inflammatory events, which in turn promotes microglial differentiation to proinflammatory subtypes and triggers an exaggerated inflammatory response in response to subsequent stimuli (80). This phenomenon may explain why the prevalence of NeuroPASC is higher in older adults (20). Although the specific mechanisms initiating microglial priming remain to be elucidated, it is generally accepted that chronic inflammation and/or repetitive inflammatory stimuli are a governing factor. Recently, Albornoz et al. (46) have demonstrated that the SARS-CoV-2 S1 spike protein acts as an NLRP3 inflammasome and microglial primer, setting the stage for increased reactivity to inflammatory stimuli (46). Persistent glial reactivity and chronic neuroinflammation in neurodegenerative diseases can be attributed to an exaggerated inflammatory response upon repeated exposition to pathological stimuli (80), such as βamyloid plaques and alpha-synuclein in AD (80) and Parkinson's disease (PD) (81), respectively. Similarly, the persistent systemic inflammation in PASC (75) could represent a stimulus with the capacity to longitudinally promote microglial reactivity, leading to maladaptive neuroinflammation in microglia previously primed during the wake of SARS-CoV-2 infection. Keeping these mechanisms in mind, SARS-CoV-2 infection and pathogenesis could potentially trigger neurodegenerative events reminiscent of AD and PD (82). As such, there exists a positive correlation between COVID-19 infection and its severity with the risk of AD development (83). Moreover, COVID-19 may exacerbate motor and non-motor symptoms in PD patients (84).
There are considerable parallels between SARS-CoV-2 and influenza sequelae. Iosifescu et al. (20) and Taquet et al. (85) compared neurological and psychiatric sequelae following these viral infections. The incidence of long-term COVID-19 and influenza-related neuro-sequelae was 2.58 and 2.06% (20) and 3.01 and 1.83% (85), respectively. The average onset of NeuroPASC symptoms was 138 days following the initial infection vs. 238 days for influenza sequelae (20). The occurrence of altered mental status was significantly greater in NeuroPASC patients (17%), but there were no statistically significant differences in other clinical signs and symptoms when compared to influenza. These symptoms include anxiety, depression, dizziness, fatigue, headaches, nausea, seizures, and strokes (20). From a pathophysiological perspective, respiratory influenza infection elicits neuroinflammation through pro-inflammatory cytokines secretion and microglial reactivity (86,87). These processes alter BBB permeability, structural hippocampal plasticity, and may underlie cognitive dysfunction (86,87). Fernández-Castañeda et al. (41) compared CSF proinflammatory cytokine profiles at 7 days and 7 weeks post-infection between mice models of SARS-CoV-2 and H1N1 influenza, revealing distinct profiles, with some overlap. Of note, CCL11, a cytokine associated with cognitive impairment (74) remained persistently elevated in both SARS-COV-2 and H1N1 models. A comparison of microglial reactivity revealed similar hippocampal pathology at 7 days and 7 weeks post-infection. However, unlike respiratory COVID, sub-cortical white matter integrity in H1N1 mice was preserved at 7 weeks, with a resolution of acute microglial reactivity and oligodendrocyte loss (41).
Alternatively, microglial activation during acute SARS-CoV-2 infection could be sufficient to induce maladaptive inflammatory pathways, leading to chronic neuroinflammation and NeuroPASC in the absence of longitudinal peripheral stimuli. This phenomenon has been described following traumatic brain injury (TBI) in human brain samples, where densely packed reactive microglia are responsible for chronic neuroinflammation and white matter degradation (88). In fact, persistent inflammatory pathology was observed in over a quarter of TBI cases and for up to 18 years following the initial brain injury (88). Studies also showed that the ensuing microglial activation and neuroinflammation from TBI results in cognitive impairment and predispose to AD (89).
A comparable syndrome is cancer-therapy-related cognitive impairment, commonly referred to as "chemo-fog, " which is characterized by mild-to-moderate impairments in memory, attention, executive functioning, and processing speed (90). The term itself and the affected neuropsychological domains resemble the "brain fog" currently used to describe NeuroPASC cognitive impairment. Furthermore, accumulating evidence suggests that chemotherapies and cranial radio-irradiation elicit a persistent microglial activation beyond the duration of treatment, leading to neuroinflammation, loss of hippocampal neurogenesis, and neuronal plasticity in addition to white matter pathology, all of which represent the core features of NeuroPASC pathology (91). Hence, it is plausible that microglial activation persists beyond the initial inflammatory stimuli in NeuroPASC, aligning with the findings observed in traumatic brain injury (TBI) and cancerrelated cognitive impairment.
Globally, the resolution of neuroinflammation is essential to mitigate neurological damage. Accordingly, this is the precise role of microglia and astrocytes subsets with tissue repair and antiinflammatory functions (33, 92). However, in neurodegenerative conditions, neuroinflammation is a crucial pathological driver as it tends to be chronically active and fails to resolve (92). Moreover, the anti-inflammatory phenotypes of microglia, which promote the clearance of inflammation in a healthy setting, are altered in neurodegenerative diseases (92). Comprehension of the delicate balance in glial cell networks and function is therefore essential to understand the complex processes governing neurodegeneration. For example, while M1 and M2 microglia are portrayed as oversimplifications to illustrate antagonistic states in both healthy and diseased brains, studies have reported distinct microglial sub-populations known as disease-associated microglia (DAM) in Alzheimer's disease (AD) (93). This unique subset of microglial cells has been specifically associated with neurodegenerative disorders and remains undetectable in healthy human brain samples. Similarly, distinct microglia populations with unique signatures have been identified in mice models, characterized by altered homeostatic gene expression and chemokine profiles that show significant overlap with DAM (41). Although the complete elucidation of DAM cells and their role in neurological disorders remains under investigation, further studies are required to map the intricate networks and function of glial cells in NeuroPASC.
. Discussion
This mini-review has explored numerous cellular processes and pathways by which SARS-CoV-2 affects the CNS leading to glial reactivity and NeuroPASC (Figure 2). We illustrate an indirect pathway, characterized by the absence of direct viral invasion of the CNS, where microglial activation and neuroinflammation are consequential repercussions of systemic inflammation and BBB breakdown. These events, therefore, result in the translocation of peripheral cytokines and immune cells to the CNS, culminating in microglial activation and neurological damage. Of note, the S1 spike protein subunit of the SARS-CoV-2 could also lead to microglial priming, setting the tone for microglial reactivity and neuroinflammatory response in a viral neuroinvasionindependent manner. We herein discussed three pathways of direct neuroinvasion that could potentially lead to microglial reactivity: i) through the olfactory bulb; ii) via a hematogenous/endothelial path; and iii) through the CSF. It is likely that microglial reactivity results from a combination of these mechanisms as they are not mutually exclusive (Figure 2).
Reactive microglia are responsible for a plethora of CNS repercussions, including synaptic plasticity impairment (94, 95), inappropriate synaptic elimination, dysfunction of hippocampal neurogenesis, and memory loss (96). Secretion of the microglial pro-inflammatory cytokines also leads to numerous neuropsychiatric manifestations, including apathy, cognitive impairment, anxiety, depression, and learning disability (97). The impact of reactive gliosis has also been well documented using SARS-CoV-2 experimental models (41,49,52). In sum, recent data suggest that persistent neuroinflammation could explain the significant prevalence of neuropsychiatric symptoms observed in COVID-19 patients.
While controversy surrounds the legitimacy of NeuroPASC as a distinct neuroinflammatory syndrome, evidence suggests that it possesses distinct microglial subtypes (41), humoral immunity signatures (78), and T-cell activation and effector signatures (77). Despite arising from different CNS insults, the consequences of microglial reactivity, such as white matter injury, impaired hippocampal neurogenesis, and loss of myelin plasticity are similar across various syndromes. These include NeuroPASC, cancer-related cognitive impairment, cognitive dysfunction following traumatic brain injury, and influenza infection. Consequently, it should come as no surprise that the clinical translation of these shared pathological lesions takes nearly identical forms. Pharmacologically targeting these reactive pathways may hold the key to treating numerous neurodegenerative and chronic neuroinflammatory diseases.
A thorough understanding of NeuroPASC pathophysiology and microglial reactivity is primordial to the development of disease-altering therapy. It is the first step toward alleviating the important socioeconomic burden of post-acute COVID-19 syndrome and its neurocognitive sequelae, a global health problem (98).
FIGURE
FIGURE Direct and indirect pathways of NeuroPASC. Schematic representation of various pathways and their consequences on microglial reactivity and neuroinflammation of the CNS following the SARS-CoV-infection are depicted.\===
Domain: Psychology Biology Medicine. The above document has 2 sentences that start with 'Although the neurological impairments and',
5 sentences that start with 'As a result',
2 sentences that start with 'In a study',
2 sentences that end with 'sequelae of COVID infection (NeuroPASC)',
2 sentences that end with 'of SARS-CoV-2 infection',
2 sentences that end with 'and neuronal cell death',
3 paragraphs that start with ' . '. It has approximately 4516 words, 209 sentences, and 40 paragraph(s).
|
[{"added": "2023-08-27T15:05:40.296Z", "created": "2023-08-24T00:00:00.000", "id": "261185972", "metadata": {"extfieldsofstudy": [], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.frontiersin.org/articles/10.3389/fneur.2023.1221266/pdf", "pdf_hash": "adb06d3e360b3b50b90c8367f1e29393555f3d17", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:39426", "s2fieldsofstudy": ["Biology", "Psychology", "Medicine"], "sha1": "1e4dba95d696842e89685c7ea3f344e9722650de", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The cerebellar clock: predicting and timing somatosensory touch
Humans are adept at predicting what will happen next and when precisely it will occur. An activity as everyday as walking at a steady pace through a busy city while talking to a friend can only happen as smoothly as it does because the human brain has predicted most of the sensory feedback it will receive. It is only when the sensory feedback does not match what was expected, say, a sudden slippery spot on the pavement, that one becomes aware of the sensory feedback. The cerebellum is known to be involved in these predictions, but not much is known about the precise timing of them due to the scarcity of time-sensitive cerebellar neuroimaging studies, such as ones conducted with magnetoencephalography. We here investigated the timing of sensory expectations as they are expressed in the cerebellum using magnetoencephalography. We did this by comparing the cerebellum’s response to somatosensory omissions from regular trains of stimulation to its response to omissions from irregular trains of stimulation. This revealed that omissions following regular trains of stimulation showed higher cerebellar power in the beta band than those following irregular trains of stimulation, precisely when the omitted stimulus should have appeared. We also found evidence of cerebellar theta band activity encoding the rhythm of new sequences of stimulation Our results furthermore strongly suggest that the putamen and the thalamus mirror the cerebellum in showing higher beta band power when omissions followed regular trains of stimulation compared to when they followed irregular trains of stimulation. We interpret this as the cerebellum functioning as a clock that precisely encodes and predicts upcoming stimulation, perhaps in tandem with the putamen and thalamus. Relative to less predictable stimuli, perfectly predictable stimuli induce greater cerebellar power. This implies that the cerebellum entrains to rhythmic stimuli for the purpose of catching any deviations from that rhythm.
Abstract
Humans are adept at predicting what will happen next and when precisely it will occur. An activity as everyday as walking at a steady pace through a busy city while talking to a friend can only happen as smoothly as it does because the human brain has predicted most of the sensory feedback it will receive. It is only when the sensory feedback does not match what was expected, say, a sudden slippery spot on the pavement, that one becomes aware of the sensory feedback. The cerebellum is known to be involved in these predictions, but not much is known about the precise timing of them due to the scarcity of time-sensitive cerebellar neuroimaging studies, such as ones conducted with magnetoencephalography.
We here investigated the timing of sensory expectations as they are expressed in the cerebellum using magnetoencephalography. We did this by comparing the cerebellum's response to somatosensory omissions from regular trains of stimulation to its response to omissions from irregular trains of stimulation. This revealed that omissions following regular trains of stimulation showed higher cerebellar power in the beta band than those following irregular trains of stimulation, precisely when the omitted stimulus should have appeared. We also found evidence of cerebellar theta band activity encoding the rhythm of new sequences of stimulation Our results furthermore strongly suggest that the putamen and the thalamus mirror the cerebellum in showing higher beta band power when omissions followed regular trains of stimulation compared to when they followed irregular trains of stimulation.
We interpret this as the cerebellum functioning as a clock that precisely encodes and predicts upcoming stimulation, perhaps in tandem with the putamen and thalamus. Relative to less predictable stimuli, perfectly predictable stimuli induce greater cerebellar power. This implies that the cerebellum entrains to rhythmic stimuli for the purpose of catching any deviations from that rhythm.
Introduction
Interest in the cerebellum has surged recently. This is likely due to the realization that it does not only subserve motor behaviour and coordination as previously believed (Schmahmann, 1997). In a recent study, using functional magnetic resonance imaging (fMRI), King et al. (2019) showed that the functions of the cerebellum span many domains, such as hand movements, saccades, divided attention, verbal fluency, autobiographical recall, word comprehension, action observation, mental arithmetic, emotion processing and language processing. Thus, it is involved in many functions hitherto believed to be associated with the cerebrum. In terms of surface area, it has also recently been shown that the cerebellum is only 20% smaller than the cerebral cortex . Furthermore, it has become clear that the cerebellum actively predicts what feedback it should receive from the external world (Hull, 2020). Important questions here are what the spatiotemporal nature of these predictions is. In this study we focus on the temporal aspects, especially how the degree of irregularity in the environment affects cerebellar predictions. In a recent study (Andersen and Lundqvist, 2019), using magnetoencephalography (MEG), we found that the cerebellum was involved in maintaining and updating expectations when regularly timed somatosensory stimulation was interrupted by unexpected omissions of stimulation. We furthermore found that an evoked response in the secondary somatosensory cortex (SII) was elicited by the omitted stimuli at the exact time somatosensory stimulation was expected, despite an inter-stimulus interval (ISI) of 3 s between stimuli. This SII response has also been found with a shorter ISI, i.e. 500 ms (Naeije et al., 2018). This suggests that the temporal expectations are very precise even over long intervals. For comparison, ISIs must be shorter than at least 500 ms for auditory omissions to elicit responses (Joutsiniemi and Hari, 1989;Yabe et al., 1997).
The cerebellum has had a reputation of being impossible to study with non-invasive electrophysiological methods due to suspicions of it being too finely folded, which should result in signal cancellation, to produce signal measurable at a distance. However, we, (Andersen et al., 2020), recently reviewed the MEG and electroencephalographic (EEG) literature on cerebellar findings and found at least thirty MEG or EEG studies reported cerebellar activity. Supporting the validity of these findings, of which many did not have cerebellum as their primary target, Samuelsson et al. (2020) recently showed, using a combination of MEG signal simulation and highresolution magnetic resonance imaging (MRI) capturing the fine structure of the cerebellum, that cerebellar signals are strong enough to be detectable by state-of-the-art MEG. The time thus seems ripe for investigating the cerebellum with a targeted study. We here follow the working hypothesis that the cerebellum functions as a clock creating temporal predictions about future stimulation.
The earliest MEG evidence pointing to the cerebellum playing an important role in updating and maintaining expectations is a study by Tesche and Karhu (2000) who estimated time courses for dipolar sources that they placed in the cerebellar vermis. They found oscillatory responses to omitted stimuli in the ranges 6-12 Hz and 25-35 Hz. These were maximal around the time the stimulus should have happened, indicating a relationship to expectational processes. We, (Andersen and Lundqvist, 2019), found more direct evidence of cerebellar oscillatory responses using a dynamic imaging of coherent sources (DICS) beamformer (Gross et al., 2001). We found that the cerebellum was more strongly activated in the first repetition of a stimulus compared to the very first stimulation for the ranges 4-7 Hz and 10-30 Hz. Furthermore, we found a tonic difference between omitted stimulations, i.e. unexpected absences of stimulation, and non-stimulations, i.e. expected absences of stimulation in the range of 3-15 Hz. The tonic difference between nonstimulations and omissions was in the direction of more power in the non-stimulations in the murhythm. This may be an effect of more power in the mu-rhythm when the body is at rest (Kuhlman, 1978), as is the case for non-stimulations. Thus, to find a time-dependent cerebellar response to omissions, as is to be expected if the cerebellum functions as a clock, we reasoned that omissions of different kinds should be contrasted against one another instead of against non-stimulations.
To this end, we have created a paradigm which in many ways is similar to those of Andersen and Lundqvist (2019) Naeije et al. (2018) and Tesche and Karhu (2000). Crucially, however, we manipulated the regularity of stimulation trains by adding different levels of temporal jitter. The basic idea of the paradigm is that an underlying rhythm is introduced with three tactile stimulations following one another with the same temporal spacing between them. Then three extra stimulations follow, either jittered (irregular) or with the same temporal spacing (regular), finally followed by an omission of stimulation. The paradigm also included longer periods of non-stimulation that would follow an omission.
The main hypothesis of the current study is that the cerebellum will be more strongly activated to omissions of otherwise expected stimulation during regular than during irregular stimulation trains. We expect this difference in cerebellar activation to occur around the expected timing of the omitted stimulus following the underlying rhythm established by the first three stimuli. Based on the findings in Andersen and Lundqvist (2019) and Tesche and Karhu (2000) we expect these differences to occur in the theta and beta bands.
Following the recent findings of Andersen and Lundqvist (2019), Fardo et al. (2017), Naeije et al. (2018) and Allen et al. (2016), we also investigated other areas that are implicated in somatosensation and in updating sensory expectations, specifically the primary and secondary somatosensory cortices (SI and SII), and the inferior parietal cortex (IPC).
Finally, it must be mentioned that when speaking of timing capabilities, structures of the basal ganglia and thalamus are likely to be heavily involved (Harrington and Haaland, 1998). In fact, it has been proposed that the basal ganglia and the cerebellum together with the thalamus are nodes of an integrated network responsible for cognitive timing (Bostan and Strick, 2018;Gibbon et al., 1997), and dysfunction of this network has been implicated in Parkinson's Disease (Caligiore et al., 2016). We thus also should expect differences in the basal ganglia and the thalamus, but due to both the depth and the closed structures of neurons (Lorente de Nó, 1947) in basal ganglia structures and the thalamus, MEG is deemed anything but insensitive to these structures (Hari and Puce, 2017). However, Attal et al. (2012) (see also: (Attal and Schwartz, 2013)) argue that MEG should be at least as sensitive to the putamen of the basal ganglia as MEG is to the hippocampus, which is recently seen as a more and more viable target of MEG (Garrido et al., 2015;Ruzich et al., 2019). Attal and Schwartz (2013) also showed experimental evidence that thalamic activation may be within reach of MEG (see also : Pizzo et al., 2019). Therefore, we investigated the putamen of the basal ganglia and the thalamus.
Participants
Thirty right-handed participants volunteered to take part in the experiment (seventeen males and thirteen females, Mean Age: 26.7 y; Standard Deviation: 6.3 y; Range: 18-41 y). The experiment was approved by the local institutional review board in accordance with the Declaration of Helsinki.
Stimuli and procedure
Tactile stimulation was generated by two ring electrodes driven by an electric current generator (Stimulus Current Generator, DeMeTec GmbH, Germany). The ring electrodes were fastened to the tip of the right index finger. One was placed 10 mm below the bottom of the nail and the other 10 mm below that. Three kinds of sequences were administered. 1) a steady sequence, where all stimuli happened exactly on time; 2) a jittered sequence, where stimuli 4-6 were 5 % jittered, i.e. they happened from ~-75 ms to ~75 ms (in steps of 1 ms) relative to where the stimulus would have happened, had the sequence been steady; 3) a heavily jittered sequence, where stimuli 4-6 were 15 % jittered, i.e. they happened from ~-225 ms to ~225 ms (in steps of 1 ms) relative to where the stimulus would have happened, had the sequence been steady. For both cases, the stimulation had to be a minimum of 13 ms away from the time where the steady stimulus would have happened.
150 sequences of each kind were administered, resulting in 3,300 events of which 2,700 were stimulations and 450 were omissions. The remaining 150 were non-stimulations. Stimuli were administered in trains of six with an ISI of 1,487 ms. This made sure that stimulation would not lock to the 50 Hz power coming from electrics. After the sixth stimulus an omission of stimuli occurred and a new train of stimulation was begun. Thus, between the last stimulus of a train and the following train 2,974 ms elapsed (Fig. 1). An exception to this was that after each fifteen sequences, a train of five non-stimulations would occur, i.e. 5 x 1,487 ms of no stimulation. Thus, omissions and non-stimulations differ by whether or not an expectation was present in the time leading up to it. This structure resulted in 450 First Stimulations, 450 Repeated Stimulations (2) and 450 Repeated Stimulations (3). Note that the first three stimulations are identical in all three conditions and are thus collapsed. Furthermore, this resulted in 3 x 150, one for each of the three conditions, Repeated Stimulations (4), Repeated Stimulations (5) and Repeated Stimulations (6). Finally, it resulted in 3 x 150, one for each of the three conditions, Omitted Stimulation and a total of 150 Non-Stimulations.
Fig. 1: Stimulation sequence:
There are three conditions. A steady rhythm where all stimulations are on-time, with an inter-stimulus interval of 1,487 ms. For the two jittered conditions, the first three stimuli are on time while the last three are off-time with different amounts of jitter, randomly chosen. The omission is timed to where the stimulus would have occurred, had it been on-time, i.e. 8,922 ms after the first stimulation. Black vertical lines indicate that the stimulation is on-time, and the red lines indicate that they are jittered. The red lines furthermore indicate where the stimulus should have occurred, had it been regular.
During the stimulation procedure, participants watched a nature programme with sound from panel speakers. Participants were instructed to pay full attention to the movie and pay no attention to the stimulation of the finger.
Electro-oculography, -cardiography and myography (EOG, ECG and EMG) were recorded. For EOG, eye movements and eye blinks were recorded to monitor participants. EMG was recorded over the splenius muscles. These were used to monitor that the participant would not build up tension in the neck muscles. Lastly, for explorative purposes, respiration was measured using a respiratory belt but is not analysed here.
Preparation of participants
In preparation for the measurement, each participant had two pairs of EOG electrodes placed horizontally and vertically, respectively, around the eyes. ECG was measured by having a pair of electrodes on each collarbone. Finally, two pairs of EMG electrodes were placed on either side of the splenius muscles. Four head-position indicator (HPI) coils, two behind the ears and two on the forehead, were placed on the participants. Subsequently, each participant had their head shape digitized using a Polhemus FASTRAK. Three fiducial points, the nasion and the left and right preauricular points, were digitized along with the positions of the HPI-coils. Furthermore, around 200 extra points digitizing the head shape of each participant were acquired. Participants were subsequently placed in the supine position of the MEG system and great care was taken, such that they would lie comfortably in the scanner, thus preventing neck tension.
Acquisition of data
Data was recorded on an Elekta Neuromag TRIUX system inside a magnetically shielded room (Vacuumschmelze Ak3b) at a sampling frequency of 1,000 Hz. As data was acquired online, lowpass and high-pass filters were applied, at 330 Hz and 0.1 Hz respectively.
Processing of MEG data
We analysed evoked responses and oscillatory responses. MaxFilter was not applied to preserve the rank of the data. The four first signal space projections (SSPs) were projected out for the sensor space analyses due to the strong influence of gradients on the magnetometers.
For the evoked responses, we low-pass filtered the data at 40 Hz (finite impulse response; zerophase; -6 dB cutoff frequency: 45 Hz; order: 198) and then cut the raw data into epochs of 800 ms, 200 ms pre-stimulus and 600 ms post-stimulus. The epochs were demeaned using the pre-stimulus period. Segments of data including magnetometer responses greater than 4 pT or gradiometer responses greater than 400 pT/m. Subsequently, epochs were averaged to create evoked responses for each of the trial types.
For the oscillatory responses, the data were band-pass filtered into the theta and beta bands, both zero-phase with a -6 dB cutoff frequency: from 4-7 Hz (Theta) and 14-30 Hz (Beta). A Hilbert transform was applied to both bands and the data were cut into epochs of 1,500 ms, 400 ms prestimulus and 1,100 ms post-stimulus for the stimulations and 750 ms pre-stimulus and 750 ms poststimulus for the omissions. The stimulation epochs were demeaned using the pre-stimulus interval up to -50 ms, whereas the omission epochs were demeaned using the whole time-interval. The difference in demeaning was due to the lack of evoked responses for omission epochs. Segments of data including magnetometer responses greater than 4 pT or gradiometer responses greater than 400 pT/m were rejected. Epochs were not rejected based on EOG and EMG due to beamformers (see below) being good at suppressing the artefacts arising from eye blinks and movements. We calculated the envelopes of the Hilbert transformed data, and, finally, to minimise the effect of outliers on the average of epochs, for each contrast of interest, we converted that contrast to a zscore, using a Wilcoxon signed-rank test entering each condition of the contrast into the test.
Source reconstruction
For both the evoked responses and the oscillatory responses, a linearly constrained minimum variance (LCMV) beamformer (van Veen et al., 1997) was applied. For the former, it was applied to evoked responses and for the latter to the Hilbert transformed epochs.
We acquired sagittal T1 weighted 3D images for each participant using a Siemens Magnetom Prisma 3T MRI. The pulse sequence parameters were: 1 mm isotropic resolution; field of view: 256 mm x 256 mm; 192 slices; slice thickness: 1 mm; bandwidth per pixel: 290 Hz/pixel; flip angle: 9°, inversion time (TI): 1,100 ms; echo time (TE): 2.61 ms; repetition time (TR): 2,300 ms. Based on these images, we did a full segmentation of the head and the brain using FreeSurfer Fischl et al., 1999). Subsequently we delineated the head and brain surfaces using the watershed algorithm from MNE-C (Gramfort et al., 2013). We created a volumetric source space from the brain surface with sources 7.5 mm apart from one another (~4000 sources). Single compartment boundary element method (BEM) solutions (volume conduction models) were calculated from the brain surfaces. For each participant, the T1 was registered to the participant's head shape with the fiducials and head shape points acquired with Polhemus FASTRAK. Finally, with the co-registered T1, we created two forward models based on the volume conduction model, the positions of MEG sensors, and the volumetric source space.
For estimating source time courses, the data covariance matrix was estimated based on the poststimulus period (from 0 ms to 600 ms) for the evoked responses. For the oscillatory responses, the data covariance matrix was estimated based on the post-stimulus time period (from 0 ms to 1,100 ms) for the stimulations, and for the omissions the data covariance matrix was estimated based on the whole time period (-750 ms to 750 ms). No regularisation was applied when estimating the filter weights. For the filters, we chose the source orientation that would maximise source output. Finally, for the oscillatory responses, we took the absolute value of the complex-valued LCMV-source-timecourses, and averaged over all the epochs to acquire an averaged source time course for both bands (4-7 Hz and 14-30 Hz). For the evoked responses, we also used the absolute value of the estimated source time courses.
For all LCMVs only magnetometers were used since these are the sensors most sensitive to deep sources. These source reconstructions were morphed onto fsaverage, a common template from FreeSurfer. Note, that signal space projection vectors were not applied on the data to be beamformed since beamformers suppress low-rank external noise well (Sekihara and Nagarajan, 2008).
To investigate contrasts in an unbiased manner, the spatial filter of the LCMV beamformer was estimated based on the covariance of the combined data. Subsequently, this filter was used to estimate the power for each condition in the contrast and finally the contrast for the oscillatory responses was calculated as: (condition_1 -condition_2) / (condition_1 + condition_2), meaning that the contrast expresses the difference in power as a percentage of total power in the two conditions. For the evoked responses, we looked at the difference (condition_1 -condition_2).
Statistical analysis
For the oscillatory responses, we focused on the comparisons between the Omissions, with the difference between Omission 0 and Omission 15 expected to be the strongest. Also the comparison between First Stimulation and Repeated Stimulation (2) was interesting due to the difference in expectation and our earlier findings of a difference in the theta band (Andersen and Lundqvist, 2019). We used a cluster permutation test (Maris and Oostenveld, 2007) to address the multiple comparisons problem arising from doing all-sensor and whole-brain analyses. The analyses done in sensor space were done on the whole time range (-400 ms to 1,100 ms for the stimulations and -750 ms to 750 ms for the omission). We used this to inform the time range of our subsequent tests for the whole-brain analysis. The null hypothesis of such a test is that data are exchangeable, meaning colloquially speaking that our labelling does not matter. The alternative hypothesis then is that the labelling does matter. The procedure was as follows. First we conducted a t-test for each of the time-sensor or time-source combinations (e.g. for the beamformer: 801 time samples and 4,342 sources; > 3 million tests). Then spatio-temporal clusters were formed based on connecting neighbouring sensors/sources and neighbouring time points. Only time-sensor/source combinations that were significant at α = 0.05 were included in clusters. 1024 permutations were then run where the condition labels, e.g. Omission_0 and Omission_15, were shuffled between trials. Using the same procedure as for the correctly labelled trials, spatio-temporal clusters were formed from the significant tests for each of the permutations. A null hypothesis distribution was sampled by adding the value of the largest cluster to the distribution for each permutation. The largest cluster in each permutation was defined as the one that had the largest sum of absolute t-values. For each cluster in the correctly labelled trials, the likelihood of that cluster, or one more extreme, was found, its pvalue, given the sampled null hypothesis distribution. The null hypothesis that the data is exchangeable was then rejected if the largest cluster in the correctly labelled data was associated with a p-value that is lower than the alpha value, which we set at 0.05.
For the evoked responses, we did the cluster permutation tests on the source time courses directly, since the SI and SII responses were expected. We focused on the comparisons between First Stimulation and Repeated Stimulation (2) and between Omitted Stimulation and Non-Stimulation following Andersen and Lundqvist (2019). We tested on the full time range (-200 ms to 600 ms).
Evoked responses
The electric stimulations gave rise to the expected SI and SII activations ( Surprisingly, no omission responses were found ( Fig. 2A). This is in contrast to two recent studies (Andersen and Lundqvist, 2019;Naeije et al., 2018), in which evoked responses source localized to SII were found peaking around 130 ms. A difference between these two studies and the B A C current, is that we in the current study used electric stimulation, whereas the two other studies used an inflatable membrane that pressed on the finger when inflated. Differences were found between First Stimulation and Repeated Stimulation (2) for the source time courses when using the cluster permutation procedure described above in the time range -200 ms to 600 ms (p BIGGEST_CLUSTER = 0.00098). No differences were found between Omission 0 and Non-Stimulation (p BIGGEST_CLUSTER = 0.73).
Oscillatory responses
For the oscillatory responses, we focused on the theta and beta band responses, following the results of Andersen and Lundqvist (2019) and Tesche and Karhu (2000). For the stimulations, the main contrast of interest was between First Stimulation and Repeated Stimulation (2). In the beta band, the null hypothesis could not be rejected (p BIGGEST_CLUSTER = 0.72). In the theta band, we found that 47 magnetometers formed part of two spatiotemporal clusters with p-values < 0.05 (p BIGGEST_CLUSTER = 0.00098) (Fig. 3A). Given that the responses peaked around 200 ms, we centred the subsequent source cluster analysis on the window of -200 ms to 600 ms. We found one cluster with a p-value < 0.05 (p BIGGEST_CLUSTER = 0.00098). The maximal value was found for the left inferior parietal cortex at 145 ms (Fig. 3D), beginning at 0 ms (Fig. 3F). A left cerebellar activation was also found after 290 ms (Fig. 3C),. For the comparison between Repeated Stimulation (2) and Repeated Stimulation (3) the null hypothesis could not be rejected (p BIGGEST_CLUSTER = 0.12). For descriptive purposes, we reconstructed all the first three stimulations based on a common spatial filter and estimated time courses for the sources shown in Fig. 3CD (Fig. 3EF). These show the cerebellar and inferior parietal cortical activity for each of the conditions in separation. For the omissions, the main contrast of interest was between Omission 0 and Omission 15, as we expected the cerebellar contrast to be the strongest there. In the theta band, the null hypothesis could not be rejected (p BIGGEST_CLUSTER = 0.25). In the beta band, we found that 14 magnetometers formed part of three spatiotemporal clusters with p-values < 0.05 (p BIGGEST_CLUSTER = 0.014 ) (Fig. 4A). The sensor topography (Fig. 4B) was compatible with an underlying cerebellar source on the right side. For the source reconstruction, we found one cluster with a p-value < 0.05 (p BIGGEST_CLUSTER = 0.0078) testing on the time interval from -400 ms to 400 ms. A right cerebellar source was part of the cluster (Fig. 4C), and the maximally responding source 0 ms was in the putamen (Fig. 4D). Running similar tests for Omission 0 vs Omission 5 and Omission 5 vs Omission 15 resulted, however, in not being able to reject the null hypothesis (p BIGGEST_CLUSTER = 0.18 and p BIGGEST_CLUSTER = 0.11 respectively). Again, for descriptive purposes, we reconstructed all three types of omissions based on a common spatial filter and estimated time courses for the sources shown in Fig. 4CD (Fig. 4EF). These show that cerebellar and putamen activity starts increasing around -250 ms of stimulation when the omission is preceded by a regular train of stimulation (Omission 0) and starts decreasing again around 0 ms. The opposite pattern emerges for the omissions preceded by irregular trains of stimulation (Omission 5 and Omission 15), which decrease from -250 ms and start increasing from 0 ms. We investigated the other areas of interest besides cerebellum and putamen. For the contrast between Omission 0 and Omission 15 in the beta band, we also found differences in the thalamus, the inferior parietal cortex, SI and SII (Fig. 5) In summary, we found that non-cortical structures, i.e. cerebellum, putamen and thalamus, followed the expected timing, i.e. the contrast peaking at ~0 ms, whereas cortical structures such as SI and inferior parietal cortex peaked before the expected stimulus (~-250 ms) and SII peaking after the expected stimulus (~250 ms).
for the individual time courses.
We then investigated the same areas for the contrast between First Stimulation and Repeated Stimulation (2) for the theta band (Fig. 6). Here, we found differences in the very same areas. Except for the cerebellum, all differences were expressed around the peak of the left inferior parietal cortex (~145 ms) (Fig. 3D&F&H). The cerebellar differences peaked around 285 ms for the left side and around 500 ms for the right side. (Tzourio-Mazoyer et al., 2002) with the exception of SII, which is from the Harvard Oxford cortical atlas (Desikan et al., 2006). See Supplementary Fig. 2
Discussion
In this study we investigated if and how the cerebellum is involved in timing expectations. We presented trains of electrical stimulation that were either regularly timed or irregularly timed and followed by an omission of stimulation. We found that our hypothesis, that cerebellum more strongly activates to omissions appearing in a regularly timed train than in an irregularly timed train, to be corroborated by our beta band findings, showing that the contrast and the omission activity related to the regular train was at their strongest close to 0 ms (Fig. 4E & 4G).
A cerebellar clock (the beta band)
That the differences in cerebellum is maximal around the expected time of stimulation indicates that cerebellar power reflects the strength of the temporal expectation, which theoretically should be strongest in the predictable condition. In this regard, the cerebellum functions as a clock keeping track of when sensory information is supposed to arrive. On the contrary, when there is temporal uncertainty regarding when the stimulus should arrive cerebellar activity decreases around the time of the expected stimulation, and for the most uncertain one (Omission 15) most so (Fig. 4E). Cerebellar power thus reflects the prediction of an upcoming somatosensory stimulation, as beta band power is expected to increase when predicted information is expected to arrive (Engel and Fries, 2010) (Fig. 4E). The cerebellar findings here are also similar to findings from the auditory domain, where it has been shown that beta band power in the auditory cortex synchronizes with the rhythm of auditory stimuli, peaking at their occurrence and then decreasing again (Arnal and Giraud, 2012;Fujioka et al., 2012).
Interestingly, the most irregular condition (Omission 15) showed a dip for oscillatory power around the time, the stimulation should have occurred, had the pattern been regular (Fig. 4G). In this condition, a stimulation is expected, but its exact timing is uncertain. Given that a function of prediction is to inhibit the processing of the expected stimulus, we can interpret this as the cerebellum uninhibiting the processing of the spatially expected, but temporally unexpected, stimulus. Interestingly, the dip begins just before the earliest time that the stimulation could occur (~-225 ms) and is back to pre-expected-stimulus levels at around 225 ms. This suggests that the clock is even tracking the range within which the stimulus should occur. This possibility is seemingly contradicted by the observation that the slightly jittered condition (Fig. 1) also starts decreasing around 225 ms (5 % of 1,487 ms = 74 ms). This can however be explained by us having used an interleaved design; i.e. when the cerebellum detects jitter in the stimulation, there is no principled way of deciding, in the current design, whether a given stimulation train is 5 % jittered because any 5 % jittered train is compatible with being a 15 % jittered train. To test whether the decrease adapts this precisely to the temporal uncertainty, a blocked design could prove useful in subsequent studies.
Our results furthermore indicated that the putamen and thalamus are more strongly activated for the omission preceded by regular trains than omissions preceded by irregular trains. This fits well with the knowledge that the thalamus and basal ganglia are involved in timing as a recent metaanalysis of fMRI timing studies showed (Teghil et al., 2019). Of notice is that they follow the timing of the cerebellum, peaking around 0 ms, indicative of a network of cerebellum, putamen and thalamus (Bostan and Strick, 2018;Caligiore et al., 2016;Gibbon et al., 1997). For these, we also saw the dip in power for the most irregular condition (Omission 15) (Fig. 4H, Supplementary Fig. 1).
Importantly though, activations of putamen and thalamus found using MEG must be treated with caution. Both are not optimal targets for MEG, due to their deep locations and the closed field of their neurons (Lorente de Nó, 1947). This results in a weak signal when measured outside the head that makes it challenging to say with confidence which deep source produced the MEG signal. However, theoretical work and simulation studies show that MEG should be sensitive to thalamus and putamen (Attal et al., 2012;Attal and Schwartz, 2013), albeit at a degree much lower than its sensitivity to the neocortex. In terms of experimental evidence, it was recently shown that MEG can retrieve patterns of activation from the putamen and thalamus (Pizzo et al., 2019). This theoretical and experimental evidence together with the knowledge that the present paradigm should elicit activity in the basal ganglia and the thalamus combined with the high number of omission trials (N=150) mean that we believe that the present evidence is supportive, but not conclusive, of actual putamen and thalamus activations measured by MEG.
In contrast to the cerebellum, the SI and inferior parietal cortical contrasts peaked before (Fig. 5) the expected arrival of the stimulus and the SII contrast peaked after the expected arrival of the stimulus. The SI and inferior parietal cortex differences may indicate a dampening of cortical activity before the onset of an expected stimulus, compatible with the decrease in theta band activity for repeated stimulations (Figs. 3 & 6). The SII activity may reflect an update of the current state of affairs, i.e. that the train of stimulation has been broken. Similar results have been reported for omissions in the auditory domain, with beta band power increasing in the time interval after the expected, but omitted, stimulus in the auditory cortex (Fujioka et al., 2009).
We unexpectedly did not find an SII evoked response to the omissions (Fig. 2), despite two recent studies having reported this (Andersen and Lundqvist, 2019;Naeije et al., 2018). Both these studies used tactile stimulation by inflating a membrane fastened to the fingertips of participants and found an SII response around 150 ms time-locked to the expected, but omitted, stimulation. A possible explanation for not finding such a response in the current study is that the instantaneousness of the electrical stimulation used in the current study relative to the longer extended touch of the membrane makes it harder for the brain to time-lock to the exact time point. This would have the consequence that evoked analyses would not be sensitive to the SII response, but that the Hilbert transformation strategy that we applied would, given its sensitivity to responses that are not precisely time-locked. Thus, it is possible that the reported SII response (Fig. 5) is similar to the response reported by Andersen and Lundqvist (2019) and by Naeije et al. (2018).
To summarize, the current findings provide evidence that the cerebellum, together with the putamen and thalamus, tracks the timing of upcoming stimulation, as reflected by beta band power. This fits the interpretation of the beta band as predicting the when of upcoming stimulation (Arnal and Giraud, 2012), and supports the notion of the cerebellum as a clock that keeps track of upcoming stimulation. Cortical structures such as the SI, SII and the inferior parietal cortex are also timed relative to the upcoming stimulation. However, the SI and the inferior parietal cortex seem to be downregulating before the expected stimulus, and SII activates after the omitted stimulus, possibly reflecting an updating of the current state of affairs. Notably, the omission-related SII difference is strongest in the right hemisphere, mirroring the evoked difference found by Andersen and Lundqvist (2019).
Encoding of expectations (theta band)
We furthermore found in the theta band that the cerebellum reacted more strongly to the first stimulation of a train compared to the subsequent stimulation around 280 ms after stimulation onset. This may reflect the unexpected nature of the first stimulation compared to the subsequent stimulation. This is similar to what has been found using electrical stimulation (Tesche and Karhu, 2000; see also : Naeije et al., 2018), but opposite to what has been found using tactile stimulation (Andersen and Lundqvist, 2019). Future studies will have to tell whether this difference is dependent on the type of stimulation. Similarly to the omission contrast, cortical areas, i.e. SI, SII, and inferior parietal cortex, and non-cortical areas, i.e., cerebellum, putamen and thalamus, were found to differ in activation between the first and the subsequent stimulation (Fig. 6). Except for the cerebellum, all contrasts peaked around 145 ms (Fig. 3F), whereas cerebellum peaked later, around 285 ms (Fig. 3E). All the cortical areas were expressed most strongly on the contralateral side of the stimulation as expected. The difference found for the SII is likely to be the same activity caught by the evoked analysis (Fig. 1C), as the evoked response is in the theta range. The later cerebellar activations are potentially an updating of the new state of affairs (Engel and Fries, 2010), i.e., that the stimulation train has been interrupted.
Cerebellum's role in forming expectations
The current study reveals that cerebellum (along with its potential network members, thalamus and putamen (Bostan and Strick, 2018)) was the only area that peaked around 0 ms when comparing omissions (Fig. 5). We interpret this as the cerebellum clocking that stimuli arrive as expected, possibly in unison with the putamen and the thalamus. This clocking activity is expressed in the beta band. Similar clocking activity has been revealed in the auditory domain for the beta band (Arnal and Giraud, 2012;Fujioka et al., 2012;Ruiz et al., 2017).
We also found cerebellar activity in the theta band when comparing the first stimulation to the subsequent stimulation. In contrast to the beta band activity, the theta cerebellar activity was unique in peaking later (~285 ms) than the first stimulation related activations (~145 ms) (Fig. 6). We interpret this as the cerebellum updating its clock for subsequent stimulation. Of note is also that, in contrast to the omission-related beta band activity, the putamen and the thalamus peak out of sync with the cerebellum. Our findings are similar to recent findings (Dave et al., 2020), where transcranial magnetic stimulation was used to knock out cerebellar theta and cerebellar beta respectively. Cerebellar theta was found to be related to encoding episodic memories and cerebellar beta was found to be related to semantic predictions related to those memories. We can thus interpret our current findings as cerebellar theta setting the clock and cerebellar beta checking whether the clock is set correctly.
Control analyses
An alternative explanation for why we found differences in cerebellar activity when comparing omissions following regular and irregular trains of stimulation may be that the proposed cerebellar clock is based on local timelocking rather than global timelocking. By global timelocking, we mean that the timing of the expected, but omitted, stimulus is set to 8,922 ms (six times the ISI (1,487 ms)) after the first stimulation of the train (Fig. 1). By local timelocking, we mean that the timing of expected, but omitted, stimulus is set to the temporal distance between the fifth and sixth stimulation, whatever that may be in each particular stimulation train. Using local timelocking, we were also able to reject the null hypothesis (p BIGGEST_CLUSTER = 0.0029), and we found similar activations of the cerebellum and the putamen (Supplementary Fig. 3). We also used mean timelocking, where we calculated mean distance between the last three stimulations (Fig. 1). Using mean timelocking, we found similar activations of the cerebellum and the putamen ( Supplementary Fig. 4) (p BIGGEST_CLUSTER = 0.0049). The way we timelocked the data in the main analysis (Fig. 4) thus does not change the results significantly.
It is also possible that the contributions of the gradient could not be filtered by the beamformer. We therefore re-ran the beamformer analysis on data where the signal space projections had been applied. This again resulted in very similar results (Omission 0 vs. Omission 15; p BIGGEST_CLUSTER = 0.0049; Supplementary Fig. 5).
Conclusion
In conclusion we find that the cerebellum predicts the timing of prospective somatosensory stimuli, functioning like a clock. This is evidenced by cerebellar beta-band power increasing and subsequently decreasing around the expected timing of stimulation, when stimuli are perfectly predictable, whereas when stimuli are less than perfectly predictable, cerebellar beta-band power decreases around around the expected time of stimulation, indicating that cerebellar beta band activity is related to prediction. This decrease is suggestive evidence of the clock tracking the uncertainty range of the expected upcoming stimulation, but further studies are needed for ascertaining that the range is also encoded. Also of note, we found evidence of cerebellar theta-band activity potentially encoding memories on which the subsequent predictions can be based. We interpret this as cerebellar theta activity reflecting setting the proposed cerebellar clock and the cerebellar beta activity reflecting checking the prediction of the clock. Furthermore, we find intriguing evidence of putamen and thalamus activation, tracking the timing of somatosensory stimuli as well. This fits well with knowledge from other domains, such as functional magnetic resonance imaging. However, given the low sensitivity of magnetoencephalography to the putamen and the thalamus, further research is warranted.
==
Domain: Psychology Biology
|
[{"added": "2020-10-10T13:15:48.500Z", "created": "2020-10-01T00:00:00.000", "id": "222232948", "metadata": {"extfieldsofstudy": ["Biology", "Psychology"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2021/01/19/2020.10.01.321455.full.pdf", "pdf_hash": "c8604948566cd37b8dcbba3ed3a88263e7952088", "pdf_src": "BioRxiv", "provenance": "peS2o-0075.json.gz:32199", "s2fieldsofstudy": ["Biology", "Psychology"], "sha1": "c8604948566cd37b8dcbba3ed3a88263e7952088", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The fronto-parietal network connects more strongly to central than peripheral V1
Central and peripheral vision are important for distinct aspects of everyday life. We use central vision to read and peripheral vision to get the gist of a scene. To understand how these differences are reflected in connectivity between V1 and higher-order cognitive areas, we examined the differential connectivity of V1 that represent central and peripheral vision. We used diffusion-weighted-imaging and resting-state blood-oxygen-level-dependent data to examine structural and functional connectivity. The present results demonstrate strong evidence that centrally-representing portions of V1 are more strongly functionally and structurally connected to the fronto-parietal network than are peripherally representing portions of V1. This suggests that these patterns of connections between central V1 and the fronto-parietal network are direct and support attention-demanding visual tasks. Overall, our findings contribute to understanding how the human brain processes visual information and forms a baseline for any modifications in processing that might occur with training or experience.
Resting-state functional networks are groups of brain regions whose activity tends to be temporally correlated at rest (Zalesky, Fornito, Cocchi, Gollo, & Breakspear, 2014) . One way a functional network can be identified is through clusters of correlated activity in the cortex (Yeo et al., 2011) . The fronto-parietal network (FPN) directs attentional control (e.g., (Zanto & Gazzaley, 2013) , the cingulo-opercular network (CON) supports the maintenance of task demands (e.g., (Coste & Kleinschmidt, 2016) , and the default mode network (DMN) is less active when there are attentional or task goals (e.g., (Raichle, 2015) and instead supports tasks such as memory retrieval and semantic processing (Binder, 2012;Gerlach, Spreng, Gilmore, & Schacter, 2011;Sestieri, Shulman, & Corbetta, 2010;Spreng, 2012) . Since functional networks can influence visual information processing in different ways, networks likely differ in how they connect to representations of the visual field across V1. There are also feedforward connections between V1 and functional network regions, as evidenced by the fact that strong stimulus-driven signals are observed in higher-order brain regions (Katsuki & Constantinidis, 2014) .
Previous work in our lab has investigated how the central-to-peripheral cortical organization of V1 influences functional connectivity between V1 and the rest of the corte x (Griffis et al., 2017) . This prior work indicated that there were retinotopic patterns of functional connectivity between V1 and functional networks during resting fixation. Specifically, the central representing portion of V1 was more functionally connected to regions belonging to the FPN and the far-peripheral representing portion of V1 was more functionally connected to regions belonging to the DMN. The fact that, as described above, central vision is under different top-down control than peripheral vision, might underlie its preferential connection to the FPN which is related to directing attentional control and has been shown to facilitate bottom-up and top-down attentional processes for visual information (Katsuki & Constantinidis, 2014) . The role of far-peripheral vision in environmental monitoring and the need for suppression of visual information from this portion of the visual field during central fixation might be the reason for preferential connectivity to the task-negative DMN (Li, 2002) . However, our previous work was limited by a small sample size (i.e., 20 participants) and design (i.e., data were acquired during the resting fixation blocks of a visual task) issues (Griffis et al., 2017) . These limitations leave questions regarding whether previous functional findings extend to free viewing during rest in a larger sample and, critically, how these effects relate to differences in the anatomical (i.e. white matter) connections of central vs. peripheral V1.
Anatomical connections can be examined by an invasive method in which retrograde or anterograde tracers are injected into the brain and resulting staining of brain regions are examined. These types of studies inject one location and look for staining in other locations, and this can give fine detail information about the connection patterns of the brain (e.g. (Felleman & Van Essen, 1991) . Because of the slow and arduous nature of these studies, examinations of connections from visual regions have been primarily focused on defining structural connectivity across areas thought to be primarily involved with vision (Andersen, Asanuma, Essick, & Siegel, 1990;Lysakowski, Standage, & Benevento, 1988;Neal, Pearson, & Powell, 1990) . Thus, it is not clear whether prior work did not find connections between V1 and higher-order processing areas or simply did not report those findings from their analysis. A more comprehensive study aimed to close this gap in the literature by creating a connectivity matrix throughout the whole cerebral cortex in macaque (Markov et al., 2014) . An injection of anterograde tracer was placed into central-representing V1. Connections that projected from V1 to regions of the frontal and parietal cortex were found including: F5 (involved in motor planning), 8l (frontal eye fields), and 7A (involved in attention modulation and planning) (Markov et al., 2014) . Of the frontal and parietal regions which were examined by Markov et. al, only area 8l (frontal eye fields) showed top-down connections to V1. Importantly, in the taxonomy put forward by Yeo (2011), the "fronto-parietal network" does not overlap with frontal eye fields (instead the frontal eye fields are within the Dorsal Attention Network) (Glasser et al., 2016;Wang, Mruczek, Arcaro, & Kastner, 2015) . No tracer was injected into the peripheral representing portion of V1. Thus, it is not clear whether these V1-fronto-parietal connections exist for peripheral portions of V1. While brain anatomy is relatively fixed in adulthood, a healthy brain changes the structural connections between regions with experience and age (Davis et al., 2009) . White matter tracts correspond to direct connections between brain regions. Structural connections can be studied in-vivo in humans with diffusion-weighted MRI and tractography. Major white matter tracts that connect to the occipital lobe such as the inferior fronto-occipital fasciculus (connects occipital lobe to lateral prefrontal cortex) and the inferior longitudinal fasciculus (connects occipital lobe to anterior temporal lobe) have been well documented using tractography methods in humans. However, the inferior fronto-occipital fasciculus has not been found in the macaque brain, which may explain why tracer studies investigating V1 have not helped to inform occipital-prefrontal cortex connections (Takemura et al., 2016) .
The goals of the current study are (1) to assess the reproducibility and generalizability of retinotopic effects on functional connections between V1 and functional networks, found in prior work (Griffis et al., 2017) ), in a new dataset collected under different task conditions (previous work used blocks of rest during a task with central fixation and the current data was collected as part of a resting-state only scan), (2) extend this prior work to the retinotopic effects of structural connections between V1 and functional networks, and (3) examine the relationship between these functional and structural connections. Since functional connectivity between two brain regions could come from both direct and indirect structural connections, we used DWI to examine direct connections between regions (Adachi et al., 2012;Honey et al., 2009) that were previously found to show functional connections.
To address these goals, we used resting-state BOLD fMRI and DWI to examine if portions of V1 that represent different visual eccentricities differ in their functional and structural connectivity to functional networks. We found 1) strong evidence that centrally representing portions of V1 are more strongly functionally connected to the frontoparietal network than are peripherally representing portions of V1, 2) Structural connections show the same pattern, with stronger connections between central V1 and frontoparietal regions, in particular a lateral frontal portion of the frontoparietal network, and 3) the pattern of structural and functional connections is similar, suggesting that this lateral frontal connection pattern arises from a direct connection, rather than from an indirect connection.
Figure 1. Graphical representation of methods.
Moving from left to right: data from the HCP dataset through the analysis stream and representations of the results. The top portion shows the analyses for functional connectivity and the bottom portion shows the analysis for structural connectivity. These analysis streams are largely parallel.
Participants
Diffusion weighted imaging, resting-state functional imaging, and structural imaging data from the 900-subject release of the Human Connectome Project (HCP) dataset were used in the study (Figure 1). Participants in this dataset were healthy young adults between 22-36 years of age who had normal or corrected-to-normal vision. Most subjects had at least one relative in the group, many of them are twins. Our hypotheses are not about individual differences, and due to the large sample size of the data there is still a great deal of diversity in the sample, therefore we did not treat related and unrelated samples separately. Using a relatively large sample size has the advantage of facilitating replication and extension of findings for the researchers. Participants with structural abnormalities (e.g. tumors and large area brain damage) that were identified through HCP quality control were excluded from the study. We then visually inspected the remaining data for white matter abnormalities. Participants were excluded if their structural scans displayed large or punctate white matter hyperintensities that were easily detected by eye. In total, 114 subjects were excluded from the original 900 subject dataset. Seven hundred eighty-six healthy subjects passed quality control standards which included 335 males and 449 females. fMRI data from four of them were dropped due to quality standards for functional comparison. Otherwise, the same participants were used in both structural and functional connectivity analyses.
Data Acquisition
T1-weighted structural MRI, resting state fMRI and multi-shell diffusion weighted (DWI) MRI data were acquired using a customized Siemens 3T "Connectome Skyra" (Sotiropoulos et al., 2013). High-resolution three-dimensional MPRAGE, T1-weighted anatomical images (TR = 2400 ms, TE = 2.14 ms, flip angle = 8, FOV = 320 x 320 mm 2 , voxel size 0.7 x 0.7 x 0.7 mm 3 , number of slices = 256, acceleration factor (GRAPPA) = 2) were used. Functional magnetic resonance imaging (fMRI) data were acquired with a multi-band gradient echo (GE) EPI sequence (voxel size = 2 x 2 x 2 mm 3 ; TR= 720 ms; TE = 33.1 ms; flip angle = 52 degrees; FOV = 208 x 180 mm 2 ; number of slices = 72) in four runs (each of them took approximately 15 minutes) with eyes open and related fixation on a cross on a dark background. Phase encoding direction was obtained in right-to-left for half of the scans and in left-to-right for the other half of the resting-state scans.
For DWI, multi-band diffusion-weighted echo-planar (EP) images (voxel size = 1.25 x 1.25 x 1.25 mm 3 ; TR= 5520 ms; TE = 89.5 ms; flip angle = 78 degrees; MB = 3; FOV = 210 x 180 mm 2 ; number of slices = 111; b=1000, 2000 and 3000 s/mm 2 , diffusion directions = 95, 96 and 97) were used. DWI data includes six runs (each of them took approximately 9 minutes and 50 seconds). Each gradient table was acquired with right-to-left and left-to-right phase encoding polarities which were then merged after distortion correction as part of the HCP Preprocessing Pipeline (Glasser et al., 2013). V1 Eccentricity Segment Definitions V1 eccentricity segments were hand-drawn within the Freesurfer fsaverage V1 label as described in a previous publication from our lab (Burge et al., 2016;Griffis et al., 2017Griffis et al., , 2015 . Nine segments of V1, approximately 10mm wide were created for both the left and right hemisphere of the Freesurfer fsaverage brain. The V1 segments were then transformed from fsaverage space to individual anatomical space, producing subject-specific segmentation of V1. Previous work has shown that cortical anatomy is a reliable predictor of the retinotopic organization of V1 (O. Hinds et al., 2009;O. P. Hinds et al., 2008) so that the more posterior parts of the visual cortex represent more central portions of the visual field. The average eccentricity of each segment was estimated from Benson and colleagues' probabilistic retinotopy template (Benson et al., 2012) and 3 retinotopic regions were identified: central vision (mean eccentricity estimates of 0-2.2 degrees visual angle), mid-peripheral vision (mean eccentricity estimates of 4.1-7.3 degrees visual angle) and far-peripheral vision (mean eccentricity estimates of 14.1-25.5 degrees visual angle) (Figure 2). These ROIs were defined in the gray matter on the cortical sheet for the freesurfer template, then moved into the individual anatomical space for each participant. To avoid the potential for artifacts due to differences in ROI size, the number of segments per eccentricity region were assigned to more evenly distribute ROI size. The 'central vision' ROI included three segments, while mid-peripheral and far-peripheral regions included two segments each to keep the number of vertices similar between sections. Figure 2. V1 Eccentricity segments. The Far-peripheral representing section of V1 is shown in yellow, the mid-peripheral representing section of V1 is shown in red, and the central representing section of V1 in green. These regions of V1 were used as target regions for tractography analyses and seed regions in functional connectivity analyses. For brevity, we sometimes use shorthands like "central V1" to refer to these centrally-representing regions.
Functional Network ROI Definitions FP, CO, and DMN labels created by (Yeo et al., 2011) were transformed from the Freesurfer fsaverage brain to individual anatomical space (Yeo et al., 2011) . Voxels within the grey matter corresponding to the network ROIs were used as seed voxels for the functional connectivity analysis and voxels within the white matter corresponding to the network ROIs were used as track seeds. Voxels were identified using the Freesurfer mri_aparc2aseg command and then transformed into individual diffusion space.
Resting-State Scan Image Preprocessing
The HCP minimal preprocessing pipeline that includes artifact removal, motion correction, and registration to standard space was used (Glasser et al., 2013) . Along with the preprocessing steps already described by Glasser and colleagues (2013), additional preprocessing steps were performed on the residual BOLD data to reduce spurious variance not associated with neural activity as described in this paragraph. Functional images were censored for movement according to validated techniques (Carp, 2013;Griffis et al., 2017;Power, Barnes, Snyder, Schlaggar, & Petersen, 2012) . Time points in which a participant moved more than 0.5 mm in one TR were replaced with an interpolated image from adjacent images. Runs were excluded if mean framewise displacement across the run was greater than 3 mm in any direction. Temporal band-pass filtering was applied between 0.009 and 0.08 Hz. In order to reduce artifactual noise, we applied regressors, including white matter and CSF signals and motion parameters that were extracted during motion correction for each subject from the previous step. Surface reconstruction, the region of interest (ROI) label generation, and image registration were also visually inspected for all subjects to ensure the accuracy of these automated computations. Next, right-to-left and left-to-right acquisitions were concatenated into a single 4D volume for the functional connectivity analysis.
Functional Connectivity Analysis
Functional connectivity refers to synchronization between time courses of activation between two brain areas due to the similar temporal signal profiles from these connected areas (Friston, Holmes, Poline, Frith, & Frackowiak, 1995) . Correlation maps for each participant were obtained from seed-to-voxel connectivity measurements between central, mid-peripheral, and far-peripheral ROIs within the primary visual cortex (V1) to each voxel in the brain. The resulting correlation coefficient maps were converted to z-score maps using Fisher's z transform. Fischer's transformed z-score maps were projected onto the individual cortical surface from 1mm below the white/gray matter boundary using Freesurfer's mri_vol2surf command. To test the functional connectivity differences, difference maps were compared by paired t-test using Freesurfer's mri_glmfit function.
Diffusion-weighted Image Preprocessing
The HCP minimal preprocessing pipeline was used for correction for B 0 and eddy current distortions (Glasser et al., 2013) . Further DWI data preprocessing was performed using the FMRIB's Diffusion Toolbox (FDT v3.0) using GPU for the acceleration of processing (graphics processing unit) (Hernández et al., 2013;Robinson et al., 2018) . For each voxel, a distribution of diffusion parameters was estimated by means of Markov Chain Monte Carlo sampling, which also allows for crossing fiber orientations (Behrens, Johansen-Berg, Jbabdi, Rushworth, & Woolrich, 2007) .
Tractography
The results from our previous study on functional connectivity, replicated here, led to the hypothesis that structural connections from central vs. peripheral regions would differ between the FP, CO, and DMN networks. Thus, we used these network regions as seeds in probabilistic tractography performed by FMRIB's Diffusion Toolbox (FDT) (Hernandez-Fernandez et al., 2019) and used the V1 ROIs as targets. For each seed voxel, 10,000 streamlines (along with default settings of maximum steps: 2000, step length: 0.5mm, curvature threshold: 0.2) were initiated and separate samples of the voxelwise diffusion distribution were calculated. A distance correction and loop-check, which prevents circular pathways, were applied. The tractography then resulted in each voxel within the seed ROI containing the number of streamlines that reached the target (V1 region) from that voxel. Tracking was performed in individual diffusion space. Seed and target regions were transformed into diffusion space for tractography analysis and tractography results were transformed into individual anatomical Freesurfer space for visualization.
Track frequencies (number of streamlines that reached the target) were transformed into track probabilities (likelihood of a track reaching the target) by dividing the log-scaled track frequency by the maximum log-scaled track frequency (Beer, Plank, & Greenlee, 2011;Wirth, Frank, Greenlee, & Beer, 2018) . This was done to mitigate possible biases arising from size differences of seeds (Smith, Beer, Furlan, & Mars, 2018;Wirth et al., 2018) . Track probabilities were projected onto the individual cortical surface from 1mm below the white/gray matter boundary using Freesurfer's mri_vol2surf command (Beer et al., 2011;Wirth et al., 2018) . Surface maps of the track termination probabilities were smoothed using a 2mm 2 Gaussian filter and averaged across all subjects.
Tractography Analysis
To test the hypothesis that patterns of functional connections previously found in V1 (Griffis et al., 2017) are similar to patterns of structural connections, comparisons were made between the central and far-peripheral eccentricity segments of V1 connectivity patterns to the FPN. Differences in track probabilities corresponding to V1 eccentricity segments connections were compared by paired t-test (using Freesurfer's mri_glmfit).
Comparison of Functional and Structural Connectivity
A subject-wise Pearson's correlation coefficient was performed for comparison across participants' structural connections and functional connections of central V1 eccentricity segment and 3 functional resting-state networks (FPN, CON, DMN).
RESULTS
We hypothesized that the connectivity between the eccentricity segments of primary visual cortex (V1) and functional networks (i.e., FPN, CON, DMN) differs in both structural and functional connections. We compared spontaneous BOLD activity within the same ROI and we performed probabilistic tractography for the comparison of structural connections.
Functional Connections to V1 depend on Eccentricity
We compared the whole-brain functional connectivity patterns of each segment of V1central, mid-peripheral, and far-peripheral. The t-test comparing functional connectivity to different eccentricity segments in V1 revealed significant effects ( p <.001) and brain regions belonging to FP, CO, and DMN functional networks (Figure 3). Notably, central representing V1 was preferentially connected (over mid-peripheral and far-peripheral V1) to regions associated with the FP network, including the mid orbitofrontal and inferior parietal regions of the FP network. While mid and far-peripheral representing V1 were not preferentially connected (over central V1) to any specific networks (Baldassano, Fei-Fei, & Beck, 2016) . This finding is similar to those found in a previous publication from our lab (Griffis et al., 2017) . Those previous results had also shown differences in connectivity between mid-peripheral-representing regions and far-peripheral representing regions, which were not observed here, (Figure 3). Thus, in this paper, we focus on distinctions between centrally-representing portions of V1 to far-peripheral portions of V1. There, vertices in yellow showed stronger (z>3) connectivity to central V1 than to both Far peripheral and mid-peripheral regions. Red indicates stronger connectivity to Central than Mid-peripheral regions, and orange indicates stronger connectivity to Central than Far-peripheral regions. The middle and right panels show analogous images highlighting vertices with significantly stronger connections to mid-peripheral and far-peripheral regions. For the middle panel, red is where mid-peripheral is greater than central and orange is where mid-peripheral is greater than far-peripheral. For the panel on the right, red is where far-peripheral is greater than central and orange is where far-peripheral is greater than mid-peripheral. Inferences regarding functional connectivity are based on these maps.
Bottom Row:
Functional Networks for comparison to the top-row. Previously documented FPN, CON, and DMN ( from Yeo et al., 2011). The FPN is shown in green, the CON is shown in tan, and the DMN is shown in blue. Note homologies between the FP network and the left panel.
To directly compare central and peripheral V1 functional connections to FPN we performed pairwise comparisons of functional connections between the Fronto-Parietal Network and the central and far-peripheral eccentricity segments of V1 (Figure 4). Results indicate that like our initial functional connectivity findings (Figure 3), there are preferential connections between central V1 and the Fronto-parietal network when compared to far-peripheral V1 ( Figure 4). Group average data was thresholded for significance (p<.001) and effect size (connectivity differences > .01). The Fronto-parietal network is outlined in green. Bottom: Group average for central and peripheral connectivity.
Structural Connectivity Eccentricity Differences
To follow up on functional connectivity findings (Figure 4), we performed pairwise comparisons of cortical track terminations between the FPN to the central eccentricity segment of V1. Results indicate that like our functional connectivity findings, there are also preferential structural connections between central V1 and the Fronto-parietal network when compared to far-peripheral V1 ( Figure 5).
Comparison of Functional and Structural Connectivity Patterns
A Pearson correlation coefficient was computed to assess the relationship between the pattern of vertex-wise track probabilities and functional connectivity correlations from 3 resting-state networks (Fronto-parietal network, Cingulo-opercular network and Default-mode network) to central V1. Structural connectivity (track probability values) and functional connectivity (functional correlations) were independently averaged across participants for each vertex. The resulting correlation coefficient of structural and functional connections between central V1 and resting-state networks was moderate [r =.3715, p <.0001]. The relationship indicates that the overall pattern of connectivity of central V1 is consistent across modalities.
DISCUSSION
Our goal with this study was to better understand the brain network basis for interactions between sensory and cognitive information, especially differences between central vs. peripheral vision. Understanding the structural and functional underpinnings of these interactions are essential for understanding the processing differences between central and peripheral vision and for future work examining plasticity of these systems.
Our approach compared structural and functional connections among different retinotopic eccentricities within V1 and large-scale functional networks (Figure 6, left column). Our results indicated that different visual eccentricities have different connectivity patterns to the rest of the brain, consistent with our previous data (Griffis 2017), and data from other analyses (Buckner & Yeo, 2014) . The present functional connectivity analyses replicated and extended previous findings on patterns of preferential connections between central V1 eccentricity segment and the fronto-parietal network (Griffis et al., 2017) . Our structural connectivity analyses further the field's understanding of the relationship between V1 and functional networks by describing the pattern of structural connections. A comparison between structure and function showed overall agreement, indicating that the functional connections are likely mediated by direct structural connections ( Figure 6, right column). The present study found differences between the connection patterns of central and peripheral representations, which is consistent with previously reported differences in information processing on central and peripheral visual information. Central vision appears to be under stronger top-down control than peripheral vision (Chen & Treisman, 2008;Lu et al., 2002;Zhaoping, 2017) . For example, stimuli presented within the peripheral visual field are harder to ignore than stimuli presented within central vision (Chen & Treisman, 2008) . The current work suggests that this distinction may come from anatomical relationships to attentional networks.
Functional Connectivity
Interestingly, prior findings are consistent with current findings, specifically finding preferential connections between central representing segments of V1 and regions belonging to the FP network (Griffis et al., 2017) . These results suggest that frontal areas influence not only cognitive control mechanisms but also primary visual processing areas, specifically central V1. On the other hand, the peripheral regions seemed to be preferentially connected more broadly across the cortex, with the exception of the FPN regions. The data provide further evidence to support the hypothesis of eccentricity dependent preferential connectivity of V1 to higher-order brain networks. One contribution in describing this connectivity is to extend previous work by Griffis et al., 2017, into a much larger dataset that was collected without fixation at rest, thereby, improving the generalizability of the findings.
Structural Connectivity
Findings from our analysis of structural connections, including preferential connections of the FPN to central V1, are consistent with our functional connectivity findings. These results support our hypothesis, based on functional connectivity findings, that the connections between V1 and brain regions associated with FPN depend on eccentricity. As previously discussed, Markov and colleagues (2012) investigated direct structural connections in the macaque brain and found weak long-range connections between V1 and regions that correspond to the human FPN. These connections were projections from V1 and could indicate that the structural connections observed here are bottom-up connections that provide visual information to aid in directing cognitive control within the FPN. However, previous work in macaques has not found major white matter tracts connecting the occipital lobe to the frontal lobe (Takemura et al., 2016) . Thus, the prior macaque literature provides limited insight into the structural connections between VA and FPN in humans. Additionally, prior work with macaques does not help to determine the directionality of the connections observed in the present study. Although diffusion tractography methods can conflate crossing fibers, the fact that tractography showed strikingly similar patterns to functional effects, bolsters the robustness of the effect.
Relationship Between Structural and Functional Connections
Statistical comparison of structural connections to functional connectivity across 3 resting state networks (FPN, CON, DMN) showed moderate correlation. Indicating that the functional connections observed between central V1 and the FPN are likely direct structural connections (functional and structural connections present). The direct, long-range connection between these regions may be related to the importance of speed in attentional control. For example, visual information needs to be processed quickly to impact attention selection. Thus, speed of processing appears to be important for providing attentional control from the FPN to central vision, which would be improved by direct structural connections. Top-down and bottom-up processing is supported by a large body of work on visual processing (Gandhi et al., 1999;Somers et al., 1999;Tootell et al., 1998;Yeshurun & Carrasco, 1998;Zhaoping, 2017) , but the present study contributes to this field by demonstrating the eccentricity dependent nature of the relationship between V1 and higher-order brain regions.
Relationship between central vision processing and Fronto-parietal Network
Complex biological systems are often driven by separate control mechanisms with distinct functional properties (Dosenbach, Fair, Cohen, Schlaggar, & Petersen, 2008) . Information processing during cognitive operations appears to rely upon the dynamic interaction of brain areas as large-scale neural networks including fronto-parietal network (FP), cingulo-opercular network (CO) and default mode network (DMN). Fronto-parietal network supports executive functions by initiating and adjusting top-down control (Dosenbach et al., 2008) . Also, the cingulo-opercular network supports salience-related functions and provides stable control over the entire task epochs. Moreover, the suppression of default mode network is critical for goal-directed cognitive processes (Anticevic et al., 2012;Spreng, Stevens, Chamberlain, Gilmore, & Schacter, 2010) . Cooperation among these top-down control systems of the brain is necessary for controlling attention, working memory, decision making, and other high-level cognitive operations (Anticevic et al., 2012;Dosenbach et al., 2007;Sonuga-Barke & Castellanos, 2007) .
FPN includes regions such as the intraparietal sulcus that play an important role in goal-directed cognitive functions (Spreng et al., 2010) , both spatial and non-spatial visual attention (Giesbrecht, Woldorff, Song, & Mangun, 2003;Scolari, Seidl-Rathkopf, & Kastner, 2015) . The role of central vision in visual processing and object recognition, as well as the need to inhibit distractors in the visual field could be the reason it was preferentially connected to the FP network. Contributions from high-order cognitive areas, like the FPN, help the brain to decide which visual areas will be prioritized for visual attention (Scolari et al., 2015) . Higher functional coupling and structural connections between central V1 and FPN is strong evidence of spatial prioritization based on both bottom-up and top-down information.
Limitations and Future Directions
Our study has several methodological limitations that should be discussed. Our study used only healthy young adults from the HCP dataset, which could influence the generalizability of our findings. Future work should include individuals from across the lifespan.
The task that participants completed during the HCP protocol was quite distinct from the task that participants performed in the previous dataset (Griffis et al., 2017) . Here, participants rested quietly, and though their instruction was to keep eyes open, no assessment of eyes open or eyes closed was performed. In contrast, the Griffis dataset included data from the rest period between blocks of a task, and eye position and lid opening were confirmed via eye-tracking. The fact that these data closely follow each other extends the possible interpretations of the original dataset: the distinction between peripheral and central V1 connectivity generalizes to a new task context. DWI-based tractography produces similar results to tracer methods (Donahue et al., 2016) ; however, probabilistic tractography indirectly traces axon bundles by modeling the path of most restricted water movement and then estimating white matter tracts. Fibers that cross, fan, or converge pose problems for estimating white matter tracts accurately (Johansen-Berg & Rushworth, 2009) . One way to improve track estimation is by modeling multiple angular compartments (e.g., ball-and-stick model) and using greater than 30 diffusion directions (i.e., 95, 96, and 97 directions in present study) both of which were used in the present study (Behrens et al., 2007) . Connections described in tractography are non-directional in that no determination of the direction of signaling is acquired. Therefore, the current study cannot interpret the direction (top-down versus bottom-up processing) of the described connections outside of the context of prior tracer studies.
Although the present tractography and functional connectivity analyses are aimed at measuring connections between eccentricity segments of V1 and functional networks, they are inherently different modalities, including but not limited to differences in the measurement of direct and indirect connections, whereas structural tractography analysis only identifies direct connections and functional connectivity analysis can identify direct and indirect connections. Therefore, the comparison between them is limited in scope. Since tractography describes direct connections between brain regions, inconsistencies where functional connectivity is present, but structural connectivity is not, could be due to indirect connections (Honey et al., 2009) . However, measuring both structural and functional connectivity provides valuable information to understand the relationship between brain regions that cannot be derived from one modality alone.
Future work could help determine the direction of the observed connections and further describe the complexity of direct and indirect connections that may exist between V1 and functional networks. While we only studied participants with healthy vision, future work should include participants with low vision in order to investigate possible connectivity changes related to vision loss. This work could serve as a baseline for these low vision studies. Future research could also help inform the plasticity of the described connections in the context of visual training and vision loss.
Conclusions
In summary, the main contribution of this work is a greater understanding of the connectivity of higher-order functional networks to the primary visual cortex (V1). Centrally-representing portions of V1 are strongly connected to the fronto-parietal network, both functionally and structurally. Strong structural connections, in particular to the lateral frontal portions of that network, implying that the functional relationship between central V1 and frontal regions is built upon direct, long-distance connections. Understanding how V1 is functionally and structurally connected to higher-order brain areas contributes to our understanding of the way the human brain processes visual information and forms a baseline for understanding any modifications in processing that might occur with training or experience.
==
Domain: Psychology Biology
|
[{"added": "2020-09-13T13:11:08.709Z", "created": "2020-09-09T00:00:00.000", "id": "221634909", "metadata": {"extfieldsofstudy": ["Biology", "Psychology"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.biorxiv.org/content/biorxiv/early/2021/02/03/2020.09.09.281725.full.pdf", "pdf_hash": "fb15cdbc23a9acf2b738a7118888b7c18227db11", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:30408", "s2fieldsofstudy": ["Psychology"], "sha1": "fb15cdbc23a9acf2b738a7118888b7c18227db11", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Field Efficacy of Some New Generation Insecticides Against Maize Aphid (Rhopalosiphum maidis) and its Effect on Yield
A field experiment was conducted at the Field Laboratory, Department of Entomology, Bangladesh Agricultural University (BAU), Mymensigh during Rabi season of 2017-18 to find out the efficacy of some new generation insecticides against maize aphid and its effects on yield. BARI Hybrid Butta-9 variety was used as experimental crop and four insecticides viz. Imidagold 20SL@ 0.1, 0.3 & 0.5 ml/L; Ambush 1.8 EC@ 1.5, 2.5 & 3.0 ml/L; Hadhak 45WP@ 0.2, 0.4 & 0.6 g/L and Suspend 5SG@ 0.5, 1.0 & 1.5 g/L were used as experimental treatments. Results clearly showed Imidagold 20SL resulted in the greatest reduction of aphid infestation and increased the grain yield compared to the all tested insecticides where the doses of 0.5 ml/L and 0.3 ml/L of this insecticide were statistically similar. Thereafter, the result was followed by Ambush 1.8 EC, Hadhak 45WP and Suspend 5SG, respectively, while the highest plant infestation, the lowest grain yield and the lowest benefit cost ratio were obtained from untreated control. Moreover, there had a positive and significant co-relation between percent reduction of plant infestation and yield increase. In addition, data also revealed that the use of Ambush 1.8 EC @ 3.0 ml/L performed second best treatment. Therefore, it can be concluded that the maize farmers can use Imidagold 20SL@ 0.3 ml/L or Ambush 1.8EC @0.3 ml/L for effective management of maize aphid under field condition. Article history Received: 03 Jul 2020 Accepted: 22 Sep 2020 Published online: 25 Sep 2020
Introduction
Maize (Zea mays L.) is the second important cereal crops after rice in Bangladesh (Ahmed, 2016). It is considered as a staple food in many countries of the world. In Bangladesh, the production of maize is about 2.81 MT in -2019(BBS, 2020. But the production of maize is very low compare to neighbor country. In country, maize is produced less than 30% in kharif season and more than 70% in Rabi season. About 90% of the home-grown maize is feeding a burgeoning poultry and fish feed industry and rests are used as human food (Alam et al., 2019a, c). It can be processed into a variety of food and industrial products including starch, flour, dent, waxy, salad, sweeteners, oil, beverages, glue, industrial alcohol, and fuel ethanol. Its grain contains about 45-50% of oil that is used in cooking. Maize is a photoinsensitive and C4-cycle crop that's why it can be grown throughout the year (Alam et al., 2019a). The infestation caused by insects on maize is increasing day by day due to continuous cultivation. Of different insect pests, maize aphid, Rhopalosiphum maidis is one of the important serious insect. It can attack maize at any stage of growth decreasing its yield. It is a polyphagous pest, i.e., it also attacks mustard, sorghum, barley and others horticultural crops. Nymphs and Adult are very much aggressive and cannibalistic in nature. So, they suck cell sap from all parts of the plant (Alam et al., 2014). Thus, it causes hamper on pollination and also introduces various types of virus and fungus on cobs and plant. In case of maize, it is reported to cause 5-7% yield loss when favorable environmental condition as well as susceptible varieties prevails (Patil et al., 2018 andAlam et al., 2015a, b;, 2020a. In this reason, the infestation of maize aphid is considered as one of the most limiting factors of maize yield (Alam et al., 2019a, c;2020a, b). However, insecticides are considered as main tools for controlling maize aphid in Bangladesh. Farmers usually use a lot of insecticides indiscriminately and frequently. As a result, it causes environmental pollution, outbreak of secondary pests, insecticide resistance, abatement in biodiversity of natural enemies, pesticides residues in grain and food adulteration. Finally, it also creates the imbalance in ecosystem. Therefore, to reduce the environmental pollution and to conserve the eco-system, suitable insecticide with optimum dose is necessary to develop & adopt ecofriendly and sustainable management system for crop cultivation. Of the many options, use of new generation insecticides is the alternative to manage the pest and develop the eco-system so that farmers can get a satisfactory yield as well as consumers can get fresh and safe food. From the above scenario and fact, in this present research, we have managed several new generation insecticides that are available in the local market from different groups for the management of maize aphid, R. maidis under field conditions and their effect on maize grain yield.
Experimental layout and location
The research was carried out in the Field Laboratory of the Department of Entomology, Bangladesh Agricultural University (BAU), Mymensingh during Rabi season of 2017-18. The research site was situated at 24.75 N latitude and 0.50 E longitudes at an average altitude of 18m above the mean sea level. The site of experiment belongs to the Sonatola series of the dark grey floodplain soil type under Old Brahmaputra Floodplain Agro-Ecological Zone (AEZ-9) (Alam et al., 2019a, b). The field experiment was laid out in Randomized Complete Block Design (RCBD) with three replications maintaining five treatments including untreated control.
Land preparation, variety and crop development
The land was prepared well with a power tiller followed by laddering and leveling the surface of the soil. All fertilizers were applied during land preparation except urea and Muriate of Potash (MOP). One-fourth of urea and MOP were applied at the time of final land preparation. The nitrogen, phosphorus, potassium, sulphur and zinc fertilizers were applied in the form of urea, triple super phosphate, muriate of potash, gypsum and zinc sulphate at the rate of 260, 80, 140, 50 and 4.5 kg/ha, respectively (FRG, 2012). Remaining urea and MOP were applied at three equal installments at prevegetative stage, full vegetative stage and early corn formation stage. Maize var. BARI Hybrid Butta-9 variety was used as experimental crop. The crop was sown in 1 st week of November, 2017 in line with raise bed, which is the normal sowing date being practiced in the experimental fields with a plot size of 10 m 2 (4m×2.5m), spacing of 60×30cm between row to row and plant to plant, respectively, and the distance was 70cm between the two plots. Total number of plots was 39. Weeding, irrigation and other intercultural operation were done properly as and when necessary for better growth and development of maize plants.
Treatments, doses and spraying
Four new generation insecticides viz. Imidagold 20SL (Imidacloprid), Ambush 1.8EC (Abamectin); Hadhak 45WP (Imidacloprid 25% + Thiram 20%) and Suspend 5 SG (Emamectin Benzoate) were selected as experimental treatments for this study. All chemical insecticides were collected from the local market of Sadar, Mymensingh. The detailed specifications (doses, group, etc.) of selected insecticides are presented in Table 1. At three stages of maize growth i. e. vegetative, inflorescence and cob formation, the plants were monitored regularly to confirm the infestation level caused by aphid and when considerable plants, inflorescence & cobs were found to be infested, then treatments were prepared and applied according to the experimental specifications stated below. Spraying was started at morning time to avoid bright sunshine and drift caused by strong wind.
Data collection parameters and procedure
Data were collected on 2, 5 and 7 days after spraying (DAS). The following parameters: Total number of plants, number of healthy and infested plants per plot, percent plant infestation, percent reduction of plant infestation, yield (t/ha) of healthy and infested grain, percent increase of yield over control and benefit cost ratio were considered for the efficacy evaluation of the selected insecticides. However, healthy and infested plants were counted from randomly selected five plants from each treated plots after each spraying of treatments. After harvesting of cobs, the grain was received from each treated plot along with control. Then these grains were weighted and recorded, and thereby, yield data were converted into ton per hectare according to treatment. All data of all stages were collected and compiled into average value. The percentage infestation of the plant was calculated by using the following formula (Alam et al., The percent reduction of infestation over control was calculated using the following formula: The percentage increase of yield over control was calculated by using the following formula: The benefit cost ratio (BCR) was calculated on the basis of prevailing market prices of maize grain, insecticides, spraying and cultivation cost etc. Benefit cost ratio was calculated using the following formula: Benefit cost ratio (BCR) =
Gross return
Total variable cost
Data analysis
All the recorded data were compiled and tabulated for statistical analysis. The Analysis of Variance (ANOVA) was done with the help of R statistics software version 3.5.3. The treatments means differences were adjudged with Duncan's Multiple Range (DMRT) Test (Gomez and Gomez, 1984) and Least Significant Difference (LSD) when necessary. Relation of variables between the percent increase of yield over control and percent reduction of insect infestation over control was calculated by using Pearson's Correlation Coefficient and Multiple Regression analysis with the R statistics software version 3.5.3 program.
Effect of different insecticidal treatment on aphid population
It was found that the tested four insecticides i.e. Imidacloprid, Abamectin, Imidacloprid 25%+Thiram 20% and Emamectin Benzoate showed efficacy on significant reduction of percent plant infestation by R. maidis compared to control (Table 2). Among, all the concentrations of the insecticides tested, the highest percent of plant infestation was found in control with a continuously increasing trend of plant infestation which was ranged from 86.01 to 96.73% and the cumulative mean of infestation was recorded 93.57%. The lowest percent of plant infestation was noted in the treatment of Imidacloprid 20SL @ 0.5ml/L having cumulative mean of plant infestation 21.30% ranged from 4.34 to 42.23% which was statistically significant (P≤0.01) than other treatments.
However, this treatment i.e. Imidacloporid 20SL @ 0.5ml/L was statistically similar to the doses of 0.3ml/L of Imidacloprid 20SL. Thus, the application of Imidacloprid 20SL@ 0.5 & 0.3ml/L performed best on percent plant infestation of maize. Hence, the cumulative mean of percent plant infestation for Abamectin 1.8EC@ 3.0ml/L, Hadhak 45WP@ 0.6g/L and Suspend 5SG@ 1.5g/L were 27.06, 33.34 and 54.72%, respectively. On the other hand, percent reduction of plant infestation over control was also calculated is shown in Figure 1. It was found that the highest percent (77.24%) of plants were reduced infestation of plant from the use of Imidacloprid 20SL@ 0.5ml/L against the R. maidis infestation which was statistically at par with the dose of 0.3ml/L of Imidacloprid 20SL. Besides, the second highest (71.08%) percent reduction of plant infestation over control was recorded in Abamectin 1.8EC @ 0.3ml/L. They were followed by Hadhak 45WP and Suspend 5SG, respectively. These clearly suggested the effective performance of Imidacloprid 20SL@ 0.3ml/L and Abamectin 1.8EC @ 03ml/L against R. maidis infestation of maize plant.
Grain yield (t/ha) and Benefit cost ration (BCR)
All the selected insecticides significantly increased the grain yield (t/ha) compared to untreated control (Table 3). In case of control condition, the grain yield was 3.23 t/ha. This yield was increased to 8.43 t/ha when maize plants were treated with Imidacloprid 20SL@ 0.5ml/L and this is the highest yield among the treatments, which was statistically similar to Imidacloprid 20SL@ 0.3ml/L (8.38t/ha).
This result (considering the efficacy of insecticides only) was followed by Abamectin 1.8EC, Hadhak 45WP and Suspend 5SG, respectively. On the other hand, it was also observed that the application of Imidacloprid 20SL@ 0.5ml/L has increased 61.48% grain yield over control (Figure 2), which was statistically at par with the dose of 0.3ml/L of Imidacloprid 20SL. The lowest percent increased of grain yield over control was recorded from Suspend 5SG@ 0.5g/L. In all field trials, the significant effects of the four insecticides on percent increase of yield over control were in the following rand order (lowest percent yield to highest percent yield): Imidagold 20SL<Ambush 1.8EC< Hadhak 45WP<Suspend 5SG. Benefit cost ratio (BCR) analysis of treatments applied against aphid on maize has been done and the results of this analyzed are presented in Table 3. The benefit cost ratio (BCR) in treated plots ranging from 0.54 to 1.46. Imidagold 20SL@ 0.5ml/L treated plots was recorded the highest benefit cost ratio (1.46) which was statistically similar with the dose of 0.3ml/L of Imidagold 20SL. They were followed by 1.22, 1.28, 1.37 in Ambush 1.8EC@ 1.5, 2.5, 3.0ml/L; 1.15, 1.25, 1.33 in Hadhak 45WP@ 0.2, 0.4, 0.6g/L and 0.96, 1.11, 1.20 in Suspend 5SG@ 0.5, 1.0, 1.5g/L, respectively, whereas the minimum benefit cost ratio (0.54) was found in case of untreated plots.
Correlation between percent reduction of plant infestation over control and percent increase of yield over control
Four new generation insecticides have a significant effect on the reduction of maize plant infestation. But the effects of four insecticides are not equal. Correlation study was done to establish the relationship between percent reduction of plant infestation over control and percent increase of yield over control. From Table 4, it was revealed that highly positive significant correlation was observed between two parameters. It was evident that the equation, Y=0.92 + (0.847)*X, where Y= yield increase over control (%) and X= reduction (%) of plant infestation over control gave a good fit to the data and the co-efficient of determination (R² = 0.92***) fitted regression line had a significant regression co-efficient. It may be concluded from the figure that the percent reduction of plant infestation over control was strong as well as positively correlated with the percent increase of yield over control. The yield of maize was increased due to the percent reduction of maize plant infestation.
Discussion
Among four insecticides, Imidagold 20SL@ 0.5ml/L found to be the best considering all the parameters studied viz. plant infestation, reduction of plant infestation, grain yield (t/ha), benefit cost ratio (BCR) and correlation between the reduction of infestation & percent increase of yield etc. which was statistically similar to the dose of 0.3 ml/L of Imidagold 20SL. The results of the present study are similar to the finding of the study ( Kumar et al., 2019) who indicated that Imidagold 20SL@ 0.3 ml/L showed maximum mortality against maize aphid. Our results have shown that Imidagold 20SL@ 0.3ml/L also caused statistically significant the highest reduction of plant infestation about 77%. Patil et al. (2018); Preetha et al.,(2012) ;Zewar, et al. (2007) also reported that Imidagold consistently performed superior to other tested insecticides against maize aphid. When considering grain yield of maize, Patil et al., (2018); Preetha et al. (2012); Zewar et al. (2007) noted that use of Imidagold returned higher yield and maximum reduction of plant infestation on various crops that supported our observations. On the other hand, we also found that, Ambush 1.8EC @ 0.3 ml/L showed the second best efficacy against aphid among four insecticides which supports the previous observations of Patil et al.(2018); Preetha et al. (2012); Suchail et al. (2001) on several crops. But, considering the efficacy of Abamectin, previously, Katare et al. (2015) and Singh et al. (1979) found that Abamectin (Ambush 1.8 EC) was found as the most effective one followed by Hadhak (Imidcloprid 25% + Thiram 20%) and Suspend (Emamectin Benzoate) against maize aphid. They reported that the efficacy of Abamectin was significantly better in comparison with that of the others. In this study, the Abamectin was also found effective against R. maidis but its effectiveness was comparatively lower than the other treatments such as Imidagold (Imidacloprid). The difference was possibly due to the dose of the Abamectin used. It was found that Katare et al. (2015) studied the effect of Abamectin using 3.5 ml/L dose and in our study it was 3.0 ml/L. In case of correlation between percent reduction of plant infestation over control and percent increase of yield over control, in our present study showed significantly strong positive correlation between mentioned parameters. Similar type of result was found by Alam et al., (2019a) and Katare et al. (2015). These findings is in agreement with the findings reported that Imidacloprid 20SL, Fipronil, Emamectin Benzoate, Imidacloprid 25%+ Thirma 20% and Abamectin provided strongly positive correlation between percent reduction of maize plant infestation over control and percent increase of yield over control (Alam et al., 2019a;Ahmad et al.,2017 andKatare et al., 2015).
Conclusion
Considering the findings from the present study, it may be concluded that the application of Imidagold 20 SL@ 0.5ml/L and 0.3ml/L is more effective together for controlling maize aphid considering the grain yield and minimum infested plant of maize. However, based on the benefit cost ratio, Imidagold 20 SL@ 0.3 ml/L is the best. This treatment, therefore, could be recommended to the maize grower for the effective management of R.
maidi. In addition, the result also showed that the use of Ambush 1.8 EC@ 3.0 ml/L performed second best treatment. Hence, in addition to Imidagold 20 SL and Ambush 1.8 EC could also be suggested for effective management of maize aphid.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-10-28T18:50:36.770Z", "created": "2020-01-01T00:00:00.000", "id": "226582159", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.5455/jbau.14970", "pdf_hash": "7965c645a598f5c86bbc47e1a86b438aad78ceb9", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:33577", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "5b80326b92558e1f73efe5c89bd9574f67fcae81", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of Integrated Use of Calliandra calothyrsus and Maize Stover with Urea on Maize Yield Components under Striga Infestation in Western Kenya
The effect of integrated use of Calliandra or maize stover with urea on Striga infestation and maize yield components was assessed in a field experiment carried out for five consecutive seasons on a clay loam Ferralsol in western Kenya. Urea and Calliandra or maize stover were combined to supply N at 75 kg ha from both sources in 0:0, 100:0, 80:20, 60:40, 40:60, 20:80, 0:100 ratios. A randomized complete block design (RCBD) with 12 treatments replicated four times was used with maize hybrid (WS 502) as a test crop. All the three ear traits (ears per plant, ear length, and ear diameter), kernels per row and grain yields were significantly higher under Calliandra (45 kg N ha) combined with urea (30 kg N ha) and Calliandra (30 Kg N ha) combined with urea (45 kg N ha) or maize stover (45 kg N ha) combined with urea (30 kg N ha) and maize stover (30 kg N ha) combined with urea (45 kg N /ha). Number of ears per plant was a major yield component and accounted for (R=0.74) of the variation in grain yield under Striga. Stepwise regression analysis showed ears per plant to be the most important yield component under Striga infestation (R=0.58) followed by kernels per row (R=0.38).
Introduction
Striga, commonly known as witchweed, is the most economically important parasitic weed seed plant in the world. It is a genus of 28 species of parasitic plants that occur naturally in parts of Africa, Asia and Australia. The genus is now classified in the family of Orobanchaceae although it was earlier placed in the family Scrophulariaceae [1]. Four species of the parasite cause economic losses in cereal crops and these are S. hermonthica (Del.) benth., S. asiatica (L.) kuntze, S. aspera (Willd.) Benth. And S. forbesii Benth. Among these, S. hermonthica is the most widespread and causes the greatest losses.
Striga hermonthica has the potential to threaten food security in many countries and is particularly significant in Africa [2]. It decimates maize which is the main staple food crop for close to 300 million people in Africa [3,4] with yield loss estimated at 10 million tons grain worth $ US 7 billion [5,6].
Striga infests about 212,000 hectares (about 15% of the arable land) in the Lake Victoria basin of Kenya [7], Urea on Maize Yield Components under Striga Infestation in Western Kenya causing yield losses of between 30-50% although losses of up to 100% have been reported [8]. Results of a survey on 83 farms in Western Kenya revealed that 73% of the farms are severely infected with S. hermonthica [9]. The potential maize yield in Western Kenya is 4-5 ton/ha [10] with the average yield loss attributed to Striga infection being 1.15 tons per hectare for maize [11]. Striga weeds are known to cause crop yield losses of between 20-100% for maize [12,13] and 20-50% in sorghum [14,13] although 100% yield loss is not uncommon. The build-up of Striga hermonthica (Del) Benth (witchweed) is associated with declining soil fertility resulting from continuous-intensive cropping without adequate fertilizer inputs [15,10], a common phenomenon in densely populated farming systems of western Kenya [16]. This profound yield loss necessitates the identification of management systems that increase and maintain adequate levels of soil fertility while at the same time reducing the Striga weed infestation. Integrated Soil Fertility Management (ISFM) has been demonstrated as a strategy that can address the complexities and peculiarities of soil fertility management on smallholder farms, help low resource endowed farmers mitigate problems of poverty and food insecurity by improving the quantity of food, income and resilience of soil productive capacity [17].
Yield a quantitative trait is functionally related to yield components. Information on the effect of integrated soil fertility management on maize yield components under Striga infestation could be useful to physiologists, modelers and agronomists. Research evidence regarding organic resource types and fertilizer application effects on Striga and maize performance under small-holder farming conditions is insufficient and fragmentary in Western Kenya and multiple agro-ecozones infested with Striga. This information could provide a vehicle for increasing efficiency of input and resource use by the staple food security crop, reducing the risks farmers face in using purchased inputs and encourage the adoption of improved technologies. The objective of this study was therefore to assess the effect of applying Calliandra and maize stover with urea on maize yield components under Striga infestation on small-scale farming conditions in western Kenya.
Experimental Site
This study was conducted at Nyabeda (N 0° 08', E 34° 24') in Siaya District of western Kenya. The area is classified as midlands (LM 2) with an altitude of approximately 1330 m above sea level [18]. Rainfall is bimodal, allowing two cropping seasons a year with the long rains starting from March ending in July and the short rains starting from August ending in November, with a mean annual of 1800 mm. Mean annual temperature ranges between 22°C and 24°C. The soils are classified as Ferralsols/Nitisols (Kandiudalfic Eutrudox), clayey, reddish, deep and well drained [18]. The soils chemical and physical properties sampled from the top 0-15 cm had the following characteristics; pH=4.9, total soil organic carbon 2.33 kg -1 , total soil N=0.23 g kg -1 , Olsen P=2.75 mg kg -1 , exchangeable Ca=7.95 cmol c kg -1 , exchangeable Mg=4.78 cmol c kg -1 , exchangeable potassium=0.05 cmol c kg -1 , exchangeable Na=0.40 cmol c kg -1 ,clay=23%, silt=14% and sand =63%.
Experimental Design, Establishment and Management
The trial had 12 treatments comprising of two plant residues applied solely, or in combination with inorganic fertilizer (urea), sole inorganic fertilizer and a control. The plant residues were maize stover and Calliandra calothyrsus ( Table 1). The experiment was laid out in a randomized complete block design (RCBD) with twelve treatments replicated four times. A substitution type of experiment was used whereby the total N rate was fixed at the recommended rate of 75 kg N ha -1 for maize in the study area. Treatments were combined in the following ratios i.e. 0:0, 100:0, 80:20, 60:40, 40:60, 20:80, and 0:100 so as to supply a total of 75 kg N ha -1 per treatment except the control i.e. treatment 0:0 where no N inputs were applied. Maize stover was obtained from neighbouring farms and Calliandra from an established demonstration plot within the area. At the beginning of every planting season, the plant residues were weighed, chopped and incorporated into the soil at a depth of 15 cm during land preparation. Fertilizer N in the form of urea was applied in splits with 33.3% of corresponding rates applied at planting and the rest (66.6%) six weeks later. Phosphorus (P) and potassium (K) were uniformly applied to all plots at the rate of 40 kg P and 20 kg K ha -1 as triple super phosphate and muriate of potash, respectively at the beginning of each season. One day after treatment application, a commercial maize variety WH 502, commonly grown by farmers in the area, was planted as a test crop at a spacing of 75 cm and 25 cm in inter and intra row spacing, respectively. Other agronomic procedures for maize production were appropriately followed after planting. Maize was harvested at physiological maturity and grain yield and yield components determined.
Striga Count and Collection in the Field
Data were collected on Striga infestation in all the plots. Striga shoots that emerged in the field were counted in the six middle rows (net plot area 27 m 2 ) at 2-week intervals starting from the day of Striga appearance and the maximum number of shoots recorded in each plot was recorded as Striga infestation. Striga emergence data were converted to the number of Striga plants m -2 . Data on maize damage score was based on a scale of 1 to 9 (1=normal plant growth, no visible damage; 9= severe damage or death) at 10 weeks after planting of maize [19]. Extensive leaf blotching with mostly frays necrotic spots. Some stunting and reduction in stem diameter, ear size and tassel size. 7 Definite leaf scorching, with gray necrotic spots, and leaf wilting and rolling. Severe stunting and reduction in stem diameter, ear size, and tassel size, often causing stalk lodging, brittleness, and husk opening at a late-growing stage. 8 Definite leaf scorching with extensive gray necrotic spots. Conspicuous stunting, leaf wilting, rolling, severe stalk lodging, and brittleness. Reduction in stem diameter, ear size and tassel size.
9
Complete scorching of all leaves, causing premature death or collapse of host plant and no ear formation
Statistical Analysis
Analysis of variance (ANOVA) and mean comparisons on the data to determine the effects of treatments on Striga incidence and maize yield components were done using Genstat 10 for windows (Release 8.1). Striga emergence data were transformed to log 10 (n+1) before statistical analysis. Average values of the respective traits in each season were used to generate correlation coefficients between grain yields and yield components. Stepwise regression analysis was employed to determine the most important yield components under Striga. Mention of statistical significance refers to P<0.05 unless otherwise stated. Significant means were separated using the LSD post-hoc tests.
Maize Damage and Striga Emergence
Seasonal trends in Striga incidence over the period of study are summarized in Table 3. There were highly significant (p<0.001) differences between seasons. The highest Striga densities (115 plants m -2 ) were observed in sole maize stover (75 kg N ha -1 ) and maize stover applied at 60 kg N ha -1 combined with urea at 15 kg N ha -1 ) (114 plants m -2 ). The two treatments had significantly higher Striga counts than all other treatments. Averaged over five seasons, Striga density was higher in the control and in treatments receiving more than 80% of N from maize stover (Table 3). In 2007 and 2008, Striga emergence was higher during long rains than the short rain seasons. The season by treatment interaction on Striga intensity was not significant (p=0.65).
There were highly significant differences (p<0.001) between treatments in damage scores ( Table 3). The highest damage score (6.6) was recorded in the control while the lowest (2.4) was recorded under Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ). Damage scores ranks of the maize under different treatment corresponded to the level of soil mineral nitrogen in the soil.
Ear traits
The effect of treatments and seasons on the number of ears per plant, ear length, and ear diameter were highly significant (P<0.001) ( Table 4). All the three ear traits (ears per plant, ear length, and ears diameter) were significantly higher under Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and Calliandra (30 Kg N ha -1 ) combined with urea (45 kg N ha -1 ) or maize stover (45 kg N ha -1 ) combined with urea (30 Kg N ha -1 ) and maize stover at (30 Kg N ha -1 ) combined with urea (45 Kg N ha -1 ) compared to sole maize stover and the control. Ears per plant were highest (0.82), under maize stover (30 kg N ha -1 ) combined with urea (45 kg N ha -1 ) while the lowest ears per plant (0.56) were recorded in the control. Ear length and ear diameter followed the same trend. Sole maize stover had the lowest ear diameter (0.39).
Number of grains per row
There were highly significant differences (p <0.001) between seasons and treatments in kernels per row ( Table 4). The highest kernels per row (35.0) was recorded under maize stover (30 kg N ha -1 ) combined with urea at (45 kg N ha -1 ) while the lowest (26.9) was recorded in the control.
Grain Yield
Grain yields and 100-grain weight were optimized at Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and Calliandra (30 Kg N ha -1 ) combined with urea (45 kg N ha -1 ) or maize stover (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and maize stover at (30 kg N ha -1 ) combined with urea (45 Kg N ha -1 ). Treatments with sole maize stover and the control had the lowest maize yields.
Relationship between Maize Grain Yield and Yield Components
All yield components except ear length and kernel rows were significantly (p<0.1-0.001) correlated with grain yields (Table 5). However, the degree of association for the various yield components was dependent on season. Correlations were strong in the long rainy seasons than the short rainy seasons. This may be attributed to higher stress levels (Striga, nutrient and moisture) in the short rainy seasons. Ears per plant showed the highest correlation with grain yield (r = 0.76, p<0.001). Stepwise regression analysis showed ears per plant to be the most important yield component under Striga infestation (R 2 =0.58), followed by kernels per row (R 2 =0.38) ( Table 6). This indicates that yield differences under Striga could be explained by variation in ears per plant and kernels per row. The overwhelming proportion of yield variation accounted by ears per plant shows that this trait was a major yield component under Striga, and it could be useful as a selection index.
Striga Emergence
The lower levels of Striga infestation under maize stover (30 kg N ha -1 ) combined with urea (45 kg N ha -1 ) and Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) could be attributed to higher mineral N levels that was provided by these treatments. High N levels have been reported to lead to a decrease in Striga infestation and an increase in crop yields [20]. Many scientists however, link the effect of nitrogen to delayed germination of the Striga seed, reduced radicle elongation, reduced production of Striga stimulant by the host plant and reduction of seeds response to the stimulants [21]. Higher densities of Striga in the control and sole maize stover may be attributed to low nitrogen levels in these treatments. Striga damage has been observed to be more pronounced in host plants grown in N deficient soils compared to those grown in soils well supplied with nitrogen [22]. Host plants produced in environments that are low in N produce higher levels of the Striga stimulant leading to higher Striga germination [23].
Ear Traits
Although ears per plant is known to be a genetically controlled characteristic [24], this trait was also influenced by nutrient availability. The higher ear per pant under maize stover (30 kg N ha -1 ) combined with urea (45 kg N ha -1 ) was mainly influenced by higher soil mineral N levels, while the control recorded low measures due to low soil nutrients. Low numbers of ears per plant may also be attributed to higher Striga numbers in the control and sole maize stover treatments (Table 3). This does suggest that Striga effects on maize grain yield relate more to pre-than post-flowering stress. Ear reduction from pre-flowering stress results from cessation of ear development and ear abortion [25]. This is in contrast to insidious and more delayed pathogenic effects, which have little or no effect on ears per plant [26].
The higher ear length in the above soil fertility management applications might be attributed to good photo assimilate supply. The maximum assimilate supply should be available during maize grain filling [27]. The two to three-week period after 50% silking is a critical stage in maize development that is highly dependent on assimilate supply; the period when final kernel number is determined [28]. Hussaini, [29] reported similar response where they linked this significant increase in yield to favorable effect of nitrogen on cob length and cob diameter, factors that are known have direct bearing on the final grain yield.
Number of grains per row
Kernels per ear is a product of rows per ear and kernel per row [30]. Since rows per ear is largely genetically controlled [31], variation in the number of kernels per ear would be expected to be largely due to differences in number of kernels per row. Earlier studies have shown that ear length is highly correlated with kernels per row studies [32]. However, a greater number of grains per ear observed under higher N rates might have resulted from the greater assimilates partitioning to the seeds as a result of a longer growth period and higher photosynthate availability during grain filling period [33]. A decrease in number of grains per ear under low N application might be attributed to poor development of the sinks and reduced translocation of the photosynthate [34].
Maize Grain Yields
The higher grain yields and 100-grain weight realized under Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and Calliandra (30 Kg N ha -1 ) combined with urea (45 kg N ha -1 ) or maize stover (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and maize stover at (30 kg N ha -1 ) combined with urea (45 Kg N ha -1 ), is mainly attributed to high mineral N levels in these treatments. Higher nitrogen content facilitates chlorophyll formation, photosynthesis, assimilate production and higher partitioning of dry matter to ears that resulted in optimum production of yield components which have direct bearing on the final grain yield. Hussaini, [29], reported similar response where they attributed this significant increase in yield to favourable effect of nitrogen on cob length and cob diameter, which all have direct bearing on the final grain yield. The increase in yield could be a result of good dry matter production for grain filling mostly attributed to a higher photosynthetic area provided by the leaves.
The low grain yields in sole maize stover treatments and the control could be attributed to low mineral nitrogen levels under these treatments (Table 3). Low quality organic materials such as maize stover with high C/N ratio (70) take long to decompose and release N leading to N immobilization [35]. Low maize yields may also be attributed to high levels of Striga infestation under these treatments (Table 3). Yield reduction from Striga infestation result from a series of physiological changes in the host plants following Striga parasitism. These include weakening of the host, wounding of its outer root tissues and absorption of moisture, photosynthates and minerals [36]. Apart from parasitism, Striga impairs photosynthetic efficiency [37] and exerts toxic or phytotoxic effects [38].
Relationship between Maize Grain Yield and Yield Components
The increase in grain yield results from beneficial influence of yield contributing characters and the positive interaction of nutrients in the blended fertilizer. The association of grain yield with number of kernels per row observed in this study agreed [39], who concluded that increasing the number of kernels per row contributes to an increase in maize grain yield. The strong relationships between grain yield and number of kernels per row and between grain yield and 100 kernels weight were also in agreement with the findings [40] indicating that these two yield attributes are the most important components directly related to grain yield in maize. Similarly, a strong positive association of maize grain yield with number of kernels per row [41].
Conclusions and Recommendations
All the three ear traits (ears per plant, ear length, and ears diameter) were significantly higher under Calliandra (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and Calliandra at (30 Kg N ha -1 ) combined with urea (45 kg N ha -1 ) or maize stover (45 kg N ha -1 ) combined with urea (30 kg N ha -1 ) and maize stover at (30 kg N ha -1 ) combined with urea (45 kg N ha -1 ). Early and consistent supply of nitrogen is critical for production of maize under Striga infested conditions. Number of ears per plant was a major yield component and accounted for (R 2 =0.74) of the variation in grain yield under Striga.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-10-29T09:08:55.725Z", "created": "2020-09-01T00:00:00.000", "id": "226354428", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.hrpub.org/download/20200930/UJAR4-10414585.pdf", "pdf_hash": "70b861e841559823862f26f2fdf4572e2190a87a", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:36959", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "321afbf4dd4aea40b93d6fb73a6ad30a250a933b", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of neem , betel leaf , devil ' s tree , jute and turmeric against gastrointestinal nematodes in sheep
Effects of neem, betel leaf, devil’s tree, jute and turmeric against natural gastrointestinal nematodes in sheep and on some hematological parameters (TEC, Hb and PCV) and body weight were studied. Thirty (30) naturally parasitized sheep were randomly divided into six groups(A, B, C, D, E and F), each consisting of five sheep. Ten percent water extract of leaves of neem, betel leaf, devil’s tree and jute were administered orally to the sheep of group A, B, C and D, respectively. Sheep of group E was treated orally with 10% water extract of rhizome of turmeric. Sheep of group F was kept as infected control group. Fecal samples, hematological parameters and body weight were examined before treatment and on 7, 14, 21 and 28 day. A significant (p<0.01) reduction of EPG count was found following administration of neem (37.60-47.03%), betel leaf (6.43-14.00%), devil’s tree (3.04-11.04%), jute (0.50-5.26%) and turmeric (0.46-8.30) in sheep. The EPG count of the control group (F) were significantly (p<0.01) increased up to the last day of experimental period. After treatment with neem, betel leaf, devil’s tree, jute and turmeric total erythrocyte count (TEC), hemoglobin (Hb) content and packed cell volume (PCV) were increased significantly (p<0.01 and p<0.05) in sheep. The body weight was increased significantly (p<0.01 and p<0.05) in neem, betel leaf, devil’s tree, jute and turmeric treated sheep. On the other hand, body weight was decreased in untreated control group. The present study reveal that 10% water extract of neem was moderately effective and betel leaf, devil’s tree, jute and turmeric were relatively less effective against gastrointestinal nematodes in sheep.
Introduction
Parasitism is an important limiting factor responsible for deteriorating the health and productivity of livestock. The agro-ecological and geo-climatic conditions of Bangladesh are highly favorable for the growth and multiplication of parasites. The greatest losses associated with nematode infections are subclinical and economic assessments have showed that financial costs of internal parasitism are enormous (Preston and Allonby, 1979;McLeod, 1995). Control of parasitic diseases has been mainly based on regular anthelmintic treatment in Bangladesh. However, as these are very expensive and unavailable to farmers in rural areas, livestock producers are not interested to use these anthelmintics. Furthermore, some serious disadvantages of using those anthelmintics, notably the development of resistance to helminth parasites (Waller and Prichard, 1985;Lans and Brown, 1998) against various anthelmintic compounds and classes, as well as their residues and toxicity problems (Kaemmerer and Butenkotter, 1973). Medicinal plants are one of the most important natural resources of a country. World Health Organization (WHO, 1993) has recognized the necessity for investigation and mobilization of ancient medicinal practices to fulfill the primary health care systems of the man and animals, and realizes that the traditional system of medicine may play an important role in the development of livestock of the third world countries. Plant remedies were also extensively used as anthelmintics in the developed world before the era of broad-spectrum synthetic drugs (British Veterinary Codex, 1953). Many currently available therapeutic compounds are plant derived and/or synthetic analogues derived from those compounds (Farnsworth et al., 1985). For these reasons, interest in the screening of medicinal plants for their anthelmintic activity has remained of great scientific interest despite extensive use of synthetic chemicals in modern clinical practices all over the world (Akhtar et al., 2000). The present study was undertaken to evaluate the efficacy of neem (Azadirachta indica), betel leaf (Piper betle), devil's tree (Alstonia scholaris), jute (Corchorus capsularis) and turmeric (Curcuma longa) against gastrointestinal nematodes in sheep. The effects of neem, betel leaf, devil's tree, jute and turmeric on hematological parameters (TEC, Hb and PCV) and body weight were also determined in this study.
Materials and Methods
The experiment was performed in the Department of Pharmacology, Faculty of Veterinary Science, Bangladesh Agricultural University (BAU), Mymensingh during the period between 1 st March, 2004 to 28 th March, 2004. Sixty sheep (approximately 2-3 years old) were selected for this study which were suspected to be suffering from natural gastrointestinal nematodes infection and they were marked at the ears by the numbered tag. Examination of fecal samples for gastrointestinal nematodes egg counts by floatation method (Rahman et al., 1996) were carried out over a week prior to commencement of treatment. On the basis of fecal sample examination results, 30 sheep of both sexes infected with gastrointestinal nematodes were selected for this study and randomly divided into six groups (A, B, C, D, E and F), each group consisting of five sheep. Ten Percent water extract of leaves of neem (Azadirachta indica), betel leaf ( Piper betle), devil's tree ( Alstonia scholaris ), jute (Corchorus capsularis ) and turmeric (Curcuma longa) were administered orally to the sheep of group A, B, C and D, respectively. Sheep of group E was treated orally with 10% water extract of rhizome of turmeric. Sheep of group F were kept as infected control without giving any treatment.
The fecal sample from all groups were examined by egg counting McMaster method as described by Soulsby (1986) before treatment (day 0) and at 7 th , 14 th , 21 st and 28 th day of post-treatment. Egg per gram (EPG) of feces were recorded. Blood samples were collected from the jugular vein of each sheep at different time intervals mentioned above. Various hematological parameters (TEC, Hb and PCV) were measured following the method of Coffin (1953). To determine the body weight gain or loss of treated and untreated control groups, body weight was taken on day 0 (pretreatment) and on 7 th , 14 th , 21 st , and 28 th day of experimental period. Collected data were statistically analyzed by the computer using statistical package programme MSTAT-C developed by Russel (1996).
Results and Discussion
The results of the effect of neem, betel leaf, devil's tree, jute and turmeric against gastrointestinal nematodes in sheep were shown in Table 1. A significant (p<0.01)reduction of EPG counts were found on 7 th , 14 th , 21 st and 28 th day following neem, betel leaf, devil's tree, jute and turmeric treated sheep of group A, B, C, D and E, respectively. Whereas, the EPG count of untreated control group (F) were significantly (p<0.01)increased up to last day of experimental period. In conformity to the present findings, Rob et al. (2004) observed that water extracts of neem was 53.72% effective against gastrointestinal nematodes (Haemonchus contortus) in sheep. Brelin (2002) found that fresh neem leaves significantly reduced H. contortus in the abomasum of the treated sheep. Arunachal et al. (2002) noted that neem leaves, seeds and bark were 53%, 49% and 38% infective against gastrointestinal helminths in sheep, respectively. Amin et al. (2008) reported that neem (10% water extract of leaves) reduced significantly (p<0.01)EPG count 62.23%, 65.77%, 56.70% and 48.05% on 3 rd , 10 th , 17 th and 28 th day, respectively in cattle. Rahman (2002) found the effects of water extract of neem, betel leaf and jute leaves were 62%, 58% and 42%, respectively in goat on 21 st day post-treatment.
The results of the effect of neem, betel leaf, devil's tree, jute and turmeric on hematological parameters (TEC, Hb and PCV) in sheep are shown in Table 2, 3 and 4. After treatment with neem, total erythrocyte count (TEC), hemoglobin (Hb) content and packed cell volume (PCV) were increased significantly (p<0.01 and p<0.05) at 7 th , 14 th , 21 st , and 28 th day post-treatment in sheep. Conversely, TEC, Hb and PCV were decreased significantly (p<0.01 and p<0.05) up to the last day of experimental period in untreated infected group. Rob et al. (2004) stated that water extracts of neem leaves increased TEC, Hb content, PCV in sheep on 28 day post-treatment. Likewise, Rahman (2002) observed water extract of neem, betel leaf and jute leaves increased TEC, Hb content on 21 st day of post-treatment in goat. Similarly, Amin et al. (2008) reported that neem (10% water extract of leaves) increased TEC, Hb content, PCV in cattle on 28 day post-treatment. Hossain et al. (1996) also reported that neem leaves increased Hb content in cattle. The above values represent the mean ± standard deviation (SD) of 5 sheep '+' = Increase ** = Significant at 1 per cent level (p<0.01)'-' = Decrease * = Significant at 5 per cent level (p<0.05)(2008) reported that body weight was increased significantly in neem treated cattle and decreased in untreated cattle. Hossain et al. (1996) also observed neem leaves and neem seed kernels increased body weight of cattle. The body weight was increased might be due to removal of parasitic load which facilitate the weight regain through proper digestion, absorption and metabolism of feed nutrients. It may be concluded that watery extracts of neem leaves was moderately effective against gastrointestinal nematodes in sheep and may be used as an alternative drugs in field condition of Bangladesh. The present study is a preliminary work on the medicinal plants in sheep in Bangladesh. However, further studies on its pharmacokinetic and toxic effects is needed before carrying out extensive field use in Bangladesh.
Table 4 . Effects of Neem, Betel leaf, Devil's tree, Jute and Turmeric on PCV (%) in sheep
Ahmed et al. (1994), betel leaf, devil's tree, jute and turmeric on body weight in sheep are shown in Table5. Neem, betel leaf, devil's tree, jute and turmeric significantly (p<0.01 and p<0.05) increased body weight in group A, B, C, D and E, respectively. On the other hand, body weight was decreased significantly (p<0.01 and p<0.05) in untreated controlled sheep of group F. These results were agreeable with the findings ofAhmed et al. (1994)in sheep. They observed body weight of neem seeds treated sheep was increased (6.74%)and decreased live weight value in untreated sheep. Similarly, Amin et al.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-09-10T03:54:50.183Z", "created": "2011-07-07T00:00:00.000", "id": "84510017", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://www.banglajol.info/index.php/JBAU/article/download/7935/5889", "pdf_hash": "b79e9a22445d03417eddb55700a59c6e34254ddb", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:39973", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "b79e9a22445d03417eddb55700a59c6e34254ddb", "year": 2011}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Variations in the Production , Qualitative Characteristics and Coagulation Parameters of the Milk of the Riverine Buffalo Determined by the Energy / Protein Content of the Diet
Sixteen Mediterranean pluriparous buffaloes were subdivided into two uniform groups of eight animals. The average weight of the two groups at the start of the trial was 671.2 and 656.7 kg. The number of days from calving were 33.4 and 33.3, and the average milk production was 12.73 and 12.33 kg/d. The trial lasted for 114 days, and was divided into two sub-periods of 58 and 56 days. The two diets, administered ad libitum, had the same forage/concentrate ratio (53/47) but in their formulation the percentage of the two forages varied. Diet 1: alfalfa hay = 10%, maize silage = 43%, concentrate 1 = 47% (6.63 MJ/kg DM of net energy; 179.5 g/kg DM of crude protein). Diet 2: alfalfa hay = 20%, maize silage = 33%, concentrate 2 = 47%, (5.99 MJ/kg DM of net energy; 155.4 g/kg DM of crude protein). For the overall trial period (33-146 days in milk), the intake of dry matter was 17.23 kg/d for Group 1 and 17.29 kg/d for Group 2 and corresponded to 2.50 and 2.58% (p<0.01) of live weight. There was no significant difference between the average weight (689.7 and 669.4 kg) and the body condition score (6.49 and 6.42) of the two groups of buffaloes. Group 1 produced a greater quantity of milk (11.89 vs. 10.90 kg/d, p<0.10) of better quality both for its higher fat content (82.32 vs. 77.29 g/kg, p<0.10) and its protein content (47.36 and 46.38 g/kg). The milk produced by the buffaloes receiving Diet 1 had a better clotting ability, lower values of r (15.98 and 16.42 min) and K20 (1.66 and 1.75 min) and a higher value of A30 (54.45 and 52.73 mm). Taking into consideration the two sub-periods, milk production was significantly different only in the first sub-period (33-90 DIM), in favour of Group 1 (13.08 vs. 11.56 kg/d, p<0.05), while the positive effect of Diet 1 was cancelled out (10.71 and 10.24 kg/d) in the second part of the trial (91-146 DIM). (
INTRODUCTION
Breeding of the Mediterranean buffalo (Bubalus bubalis L.) in Italy is undergoing continuous expansion, even outside the traditional areas located in the central and southern areas of the country. This phenomenon is principally determined by three factors: i) the milk quotas, set by the European Union, impose limits on the production of cows' milk; ii) the price of buffalo milk, which is used exclusively for transformation, is approximately three times higher than that for cows' milk; iii) the availability of mozzarella is inferior to the market demand, both nationally and internationally (USA, France, Germany, United Kingdom), and, according to de Stefano (1999), in order to satisfy this market demand it would be necessary to double the number of lactating buffaloes.
Concurrently with the increase in numbers being bred, modern stabling and milking techniques have been developed. Feeding management has been difficult due to the fact that only a few years ago the nutritional requirements were unknown since these animals were free to range on open pasture or were fed principally with industrial by-products while today the same foods are used as for the dairy cow. Now the Italian buffalo dairy farming industry has achieved economic importance, there has at the same time been an interest in increasing medium-level production. Definition of the energy and protein allowances of the lactating buffalo has become of fundamental importance since, with the optimization of feeding rations in the different productive phases, there is a greater economic advantage for businesses as well as control over environmental pollution.
The appropriate protein level at the different stages of lactation is still not sufficiently validated since only a limited literature exists, which provides contrasting indications (Verna et al., 1994;Campanile, 1998). A recent study (Bartocci et al., 2006) has demonstrated that through the administration of a diet with a high protein and energy level, a positive effect on the quantitative production of milk is achieved. Puppo et al. (2002) have demonstrated similar digestibility of crude protein by the Friesian cow and the Mediterranean buffalo. A higher ammonia level was found in the rumen of the buffalo (Kennedy et al., 1992;Bittante et al., 1994), which is more voluminous than that of the cow (Bartocci et al., 1997) with a greater capacity to degrade both crude protein and protein-free dry matter (Terramoccia et al., 2000). Bartocci and Terramoccia (2006) found that, in the buffalo species, the digestibility of the undegraded protein is higher than that of microbial origin; Nisa et al. (2008) have demonstrated that high ruminally undegradable protein diet increased nutrient intake and milk yield.
Recent research studies (Di Lella, 1998;Bartocci et al., 2002;Infascelli, 2003;Bartocci et al., 2006) have also sought to define the intake capacity of the lactating buffalo, but with conflicting results.
The aims of this study were to evaluate the effect of increased protein level on utilization of two diets, with energy concentration similar to that of previous work (Bartocci et al., 2006), in order to investigate the intake of dry matter and quantitative and qualitative milk production as well as to define the energy/protein allowances necessary for medium-high production in the first stage of lactation.
Animals and diets
Sixteen multiparous buffaloes were used in the trial, and were subdivided at the beginning of the study into two uniform groups of eight animals. The average production of the previous lactations was 2,147 and 2,388 kg, for 255 and 259 days; the number of lactations had been 2.66 and 3.16, and the distance from calving was 33.4 and 33.3 days. The average level of milk production of the two groups at the start of the trial, was 12.73 and 12.33 kg/d; the average weight was 671.2 and 656.7 kg. The animals were weighed monthly and at the end of the trial. The body condition score (BCS) of the two groups of buffaloes at the beginning of the experiment was 6.24 and 6.23; the degree of fattening was determined monthly by three estimators according to the method of Wagner et al. (1988) as modified by Campanile et al. (1998) for the lactating buffalo. This method provides for the use of a score which ranges from 0-9. The differences between the two groups, for all the above-mentioned parameters, were not significant. The trial lasted 114 days (33-146 days in milk, DIM) and was subdivided into two sub-periods of 58 (33-90 DIM) and 56 days (91-146 DIM). There were two milkings, with an interval of 12 h between the morning and the evening milking. The individual production of milk was established at the outset of the trial, subsequently at intervals of 14 days and then at the end of the trial (nine controls).
Chemical analysis
Samples of forages and concentrates as well as residues were collected and these underwent the following analytical determinations: dry matter (DM), crude protein (CP), crude fibre (CF), ether extract (EE) and ash according to AOAC (1995); neutral detergent fibre (NDF), acid detergent fibre (ADF) and lignin (ADL) according to the Goering and Van Soest method (1970). In addition, the non-structural carbohydrates (NSC) were calculated as reported by Van Soest et al. (1991). The individual milk samples, representative of the yield of the two milkings, underwent the following analytical determinations: fat, protein (Nx6.38),casein, pH and urea (ASPA, 1995). In addition, three parameters of milk coagulation were established: rennet clotting time (r), curd firming time (K 20 ), curd firmness (A 30 ) by means of the thromboelastograph Formagraph as reported by Zannoni and Annibaldi (1981). In order to estimate the quantity of mozzarella cheese produced daily, the equation proposed by Altiero et al. (1989) was used. The fluoro-opto-electronic method (Somacounter 300, Bentley Instruments, USA) was employed on monthly samples to determine the somatic cell counts.
Statistical analysis
For the intake of dry matter, net energy and chemical parameters, data from daily means of group were used whereas for live weight and milk quantitative and qualitative parameters, data from individual animals were utilized.
Diet composition
Table 1 records the chemical composition, the net energy of the foodstuffs and the two diets utilized. The energy/protein content of Diet 1 (6.63 MJ/kg DM; 179.5 g/kg DM) does not appear in the tables of indicative requirements (Zicarelli, 2000;Bartocci et al., 2002; Scientific Committee of the Consortium for the Protection of Campania Buffalo Mozzarella Cheese, 2002). Diet 2, based on knowledge acquired during a previous study (Bartocci et al., 2006), has an energy/protein ratio (5.99 MJ/kg DM; 155.4 g/kg DM) deemed adequate. Considering the low level of structural carbohydrates supplied in Diet 1, the animals never had any metabolic or physiological problems.
Intake capacity
Table 2 presents the average daily intake of DM, NE L , CP, structural and non-structural carbohydrates. Taking into consideration the entire period of the trial (33-146 DIM), Diet 1 did not modify the ingestion of dry matter which was similar for the two groups (17.23 and 17.29 kg/d), and these results concur with what was found by Bartocci et al. (2006); this research was undertaken under the same climatic conditions (spring ~ early summer) as the present trial utilizing pluriparous buffaloes with an average daily production for the previous year similar to that of the buffaloes used in the present study. The hypothesis can therefore be made that voluntary ingestion is not modified by an increase in the energy/protein level of the diet. According to Di Lella (1998), in the buffalo species, under the same physiological conditions, ingestion levels are almost always inferior to those detected for bovines. This result could be determined by the rate of transit of the foodstuffs with a slower ruminal transit (Bartocci et al., 1997) in the buffalo compared to the bovine. The difference between the ratio of ingestion of dry matter and the live weight (DMI/LW) of the two groups was significant (2.50 vs. 2.58%, p<0.01). These results concur with what was found in three experimental trials on lactating buffalo by Verna et al. (1994) where the ratio of dry matter/live weight varied from 2.2 to 2.6% depending on the type of forage utilized. Ranijhan (1992) also obtained the same variations while Bartocci et al. (2006) recorded a value of 2.43% for this parameter. On the contrary, Sarrubbi et al. (2000) and Infascelli et al. (2003) obtained an ingestion equal to 3.2% and 3.1% of live weight; those results were derived from the utilization of animals with a high productive capacity, not representative of the Italian buffalo population. The average daily intake of NE L , CP and NSC was significantly (p<0.01)higher for Group 1, while adversely the intake of NDF was higher for Group 2 (p<0.01).
When considering the groups within the sub-period, there was no significant difference in the intake of dry matter; on the contrary, this parameter increased significantly (p<0.01), as did the DMI/LW ratio, passing from the first to the second sub-period for both groups. The peak intake was recorded at 115 and 123 days of lactation with 18.38 and 18.50 kg/d for the first and the second group, respectively, thus confirming the findings of Bartocci et al. (2006) of peak intake (17.40 kg/d) at 120 days. The average daily intake of NE L , CP and NSC was significantly higher (p<0.01) in both the sub-periods for Group 1. On the contrary, NDF intake was always significantly (p<0.01)higher for Group 2. Passing from one sub-period to another there was a significant increase (p<0.01) in all the abovelisted parameters for both groups.
Milk yield and milk composition
Table 3 presents the live weight, average daily gain, body condition score, milk yield and quality. Taking into consideration the entire period, no significant difference was noted between the average weights (689.7 and 669.4 kg), and the body condition scores (6.49 and 6.42) of the two groups; likewise the average daily gain (416 and 283 g/d), that determined an average increase of body weight of 47.42 and 32.26 kg, did not display any significant difference. Group 1, which received the diet with the higher energy/protein concentration, produced a greater quantity of milk (11.89 vs. 10.90 kg/d, p<0.10) of better quality both for its higher fat content (82.32 vs. 77.29 g/kg, p<0.10) and its protein content (47.36 and 46.38 g/kg) but with a greater fattening capacity of the buffaloes of 15.16 kg. Therefore Diet 1, with a higher energy input, determined not only a higher protein level, as for dairy cows, but there was also a tendency to higher level of fat. The increase of lipids has also been confirmed by other authors (Usmani and Inskeep, 1989;Verna et al., 1994;Bartocci et al., 2002). Bartocci et al. (2006), with a diet similar to that of Diet 1 for NE L (6.69 MJ/kg DM) but with lower CP content (158.3 g/kg DM), achieved for the entire period of the trial (23-137 DIM) a milk yield similar to that of the group fed Diet 1 (11.66 kg/d) but of inferior quality with regard to fat content (73.20 g/kg), protein content (43.24 g/kg) and a lower average daily gain (307 g/d). Thus the greater quantity of protein of Diet 1 does not result in higher production but exclusively in better milk quality and an increased fattening of the animals. Comparing Diet 2 with that of previous work (Bartocci et al., 2006), there is a similar energy level (5.99 and 6.04 MJ/kg DM), but a different protein content (155.4 and 144.4 g/kg DM). This gives rise to a variation in the milk production of 10.90 and 9.42 kg/d, but a similar fat content (77.29 and 80.08 g/kg) and protein content (46.38 and 46.41 g/kg). Therefore, the difference in the protein level of the diet probably promoted galactopoiesis, increasing the quantity (+1.48 kg/d) but leaving the milk quality practically unchanged. This result reveals what was asserted by Campanile (1998) with regard to the onset of lactation and further conjectured by Bartocci et al. (2006) that, with a diet of 6.04 MJ/kg DM, the protein concentration must not be inferior to 150 g/kg DM. The casein content was similar for the two groups (41.36 and 40.37 g/kg) but higher than that obtained by Tripaldi (1994) and by Bartocci et al. (2006). This content was on average 87.16% of the protein content, while the previous authors noted 82.00 and 81.50%. The daily production of fat (0.97 vs. 0.83 kg/d, p<0.01), protein (0.56 vs. 0.49 kg/d, p<0.01) and casein (0.49 vs. 0.44 kg/d, p<0.05) was always statistically higher in favour of Group 1. With regard to milk urea, the highest value (39.47 mg/100 ml) was linked with the diet with the highest protein content, but the difference with Group 2 was not significant (38.12 mg/100 ml). Campanile et al. (1998), using a diet containing 120.0 g/kg of crude protein, noted a milk urea value equal to 35.00 mg/100 ml while Roy et al. (2005) obtained 44.82 and 42.53 mg/100 ml with diets containing 146.1 and 140.2 g/kg of crude protein. Bartocci et al. (2006), using diets containing 158.3 and 144.4 g/kg of crude protein, achieved similar milk urea values (39.09 and 39.05 mg/100 ml).
Within the two sub-periods, no significant difference emerged between the average weight, the average daily gain and the body condition score of the two groups. When considering the differences in the three above-listed parameters between the sub-periods, the only noteworthy finding to emerge from Diet 1, was that of average daily gain (252 vs. 580 g, p<0.05). Milk production, within the two sub-periods, was significantly different only for the first period in favour of Group 1 (13.08 vs. 11.56 kg/d, p<0.05). However, the positive effect of Diet 1 was neutralized (10.71 and 10.24 kg/d) in the second part of the trial. Consequently Diet 1, considering live weight, average daily gain and milk production, appears more appropriate only in the first ninety days of lactation. Figure 1 records the milk production of the two groups resulting from each control. The effect of Diet 1 was immediate with a peak of 14.00 kg/d after 47 DIM and with a successive production of approximately 13 kg/d, lasting up to 90 DIM. The same trend of milk production for Group 1, but at a lower level, was obtained by Bartocci et al. (2006) using 6.69 MJ/kg DM and 158.3 g/kg DM of crude protein, with a peak in production (13.13 kg/d) at 58 DIM. With Diet 2, there was a downturn (from 12.33 to 11.13 kg/d) with respect to the initial data and production was maintained almost steady at this level up to 90 DIM; therefore, in order to achieve an increase in production during this phase of lactation, it is essential to utilize diets with a high energy/protein concentration. The difference in milk production between the two groups at each control in the first sub-period (33-90 DIM), turned out to be statistically significant in favour of Group 1; in the second sub-period, this difference was not significant, and progressively decreased until it was erased at the end of the trial. Therefore Diet 2, taking into consideration milk production, live weight and average daily gain, proved to be the most suitable for buffalo feeding from day 91 to 146 of lactation. Within the two sub-periods (Table 3) there was no significant difference in the content of fat, protein or casein in the milk of the two groups, which was always higher in favour of Group 1. Comparing the first with the second subperiod, the content both in fat as well as in casein, increased significantly for both groups (fat: 76.58 vs. 88.08 g/kg, p<0.01, casein: 39.94 vs. 42.78g/kg, p<0.05 for Group 1; fat: 71.80 vs. 82.72 g/kg, p<0.05, casein: 39.06 vs. 41.68 g/kg, p<0.05 for Group 2) while the protein content increased significantly only in Group 1 (46.14 vs. 48.59g/kg, p<0.05). Considering the fat content (Figure 2) and protein content (Figure 3), control by control in both cases, these were always higher, although not significantly, in favour of Group 1. Therefore, the effect of the diet with the higher energy-protein concentration not only significantly increased the quantity of milk produced in the first part of the trial but also improved the quality for the entire duration of the research. Taking into consideration the average daily amount of fat, protein and casein within the two sub-periods, there was always a significant difference for all three parameters in the first sub-period, in favour of Group 1; in the second sub-period this applied only to the protein. In the two sub-periods there was no significant difference between the two groups with regard to the concentration of urea in the milk. With the passage from the first to the second subperiod, there was a decrease in this parameter which was significant only for Group 1 (42.21 vs. 36.73mg/100 ml, p<0.10). This reduction was probably due to the significant increase of casein content in both groups.
Coagulation parameters and estimated yield of Mozzarella cheese
Table 4 lists the acidity, the coagulation parameters and the estimated yield of mozzarella cheese for both milks produced. Taking into account the entire period of the trial, the milk pH, 6.70 and 6.80 respectively for Groups 1 and 2, was significantly different (p<0.01). The pH level was inversely related to the energy level of the diet, as observed by Tripaldi (1994), and also to the casein content (41.40 and 40.50 g/kg), as established by Alais (1984) and by Tripaldi et al. (1997). Bearing in mind the three coagulation parameters of milk: rennet clotting time (r), curd firming time (K 20 ) and curd firmness (A 30 ), no significant difference was found. Nonetheless, the values of r (15.98 and 16.42 min) and K 20 (1.66 and 1.75 min) were lower for the milk produced by Group 1. On the contrary, A 30 (54.45 and 52.73 mm) was higher. A positive link between curd firmness and casein content was observed as already highlighted in buffalo milk by Tripaldi et al. (1997) and by Bartocci et al. (2006). From these findings it can be deduced that milk produced from the buffaloes fed with Diet 1 was not only of better quality, but also had a better aptitude for coagulation as determined by a low value of r, K 20 and by a high value of A 30 . The estimated mozzarella production was significantly higher in favour of Group 1 (3.05 vs. 2.59 kg/d, p<0.01); an intermediate value (2.71 kg/d) was obtained over the same period of lactation by Bartocci et al. (2006) with a diet composed of 6.69 MJ/kg DM of net energy and 158.3 g/kg DM of crude protein. No significant difference was found in the number of somatic cells: 141.22 and 152.74 n×1,000/ml respectively for the milk produced by Group 1 and 2.
In the two sub-periods, the pH of the milk was constantly higher in favour of Group 2, but with a significant difference only in the second sub-period (6.57 vs. 6.71,p<0.01). Passing from the first to the second subperiod, for both milks a significant reduction (p<0.01) was found in both pH (6.83 vs. 6.57;6.89 vs. 6.71) and the values of r (18.44 vs. 13.54min; 19.09 vs. 13.76 min) but no significant reduction was observed for K 20. On the contrary, a significant increase (p<0.01) of A 30 (48.47 vs. 60.42 mm; 47.25 vs. 58.21mm) was detected. Therefore both milks had a better aptitude for coagulation in the second part of the trial (91-146 DIM) and of the two, that of Group 1 always had the best capacity for coagulation. Estimated mozzarella yield was, in both sub-periods, significantly higher for Group 1.
CONCLUSIONS
The higher energy-protein level did not affect the intake of dry matter which remained fairly constant. This diet could be used exclusively during the first ninety days of lactation, with a positive effect on milk production and without any negative symptoms of a metabolic or physiological nature. The diet with the lower energy-protein content could be used from the 90 th to at least the 150 th day of lactation.
The milk produced by both groups was of a good quality, but the best was from the buffaloes fed with the diet with the highest energy and protein content, which displayed a greater aptitude for coagulation.
DIM = Days in milk; DMI = Dry matter intake; LW = Live weight. Means in the same row followed by different superscripts are significantly different (A, B: p<0.01). Means in the same column preceded by different superscripts are significantly different (A, B: p<0.01).
Table 1 .
Dry matter (g/kg as fed), chemical composition (g/kg DM) and net energy (MJ/kg DM) of feedstuffs utilized in experimental diets 1
Table 2 .
Daily intake of dry matter, net energy, crude protein, structural and non-structural carbohydrates for entire period and interaction group×sub-period
Table 3 .
Live weight, average daily gain, body condition score, milk yield and quality for entire period and interaction group x subperiod
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-04-20T18:21:40.997Z", "created": "2010-08-20T00:00:00.000", "id": "17824354", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.animbiosci.org/upload/pdf/23-154.pdf", "pdf_hash": "3afbe01f35bee625d88b91116e49bedbb1776899", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:31292", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "d684804521852b80f9bcf9d796436403ce559d57", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Molecular characterization, biofilm production and sensitivity to the natural extract of Staphylococcus spp. from subclinical mastitis in goats
Mastitis is one of the most important diseases in dairy milk systems. Molecular studies of the principal agents that are involved, as well as tests of bacterial sensitivity to plant extracts, are currently the most common forms of research related to this disease. The aim of the present study was to molecularly characterize isolates from cases of subclinical mastitis and to investigate the production of biofilm associated with extracts from Hymenaea martiana. The phytochemical screening of the crude ethanolic extract (CEE) confirmed the presence of phenolic substances, flavonoids, steroids and terpenoids. The antimicrobial susceptibility profile of the isolates that were assessed exhibited variations, with a notably high sensitivity to gentamicin. After the addition of the extract, 77.3, 81.8 and 86.3% of the isolates exhibited a reduction in biofilm production in the ethyl acetate, chloroform and hexane fractions, respectively. An extensive production of biofilm can be observed in the isolates that were not in contact with the natural extract. However, when in contact with the plant extract, there was a reduction in biofilm production, thereby demonstrating the therapeutic potential of the natural extract of H. martiana in mastitis caused by Staphylococcus spp.
Key words: Mastitis, pathogens, resistance genes, sensitivity tests.
INTRODUCTION
Mastitis is one of the main diseases in dairy herds, considering the high costs resulting from falling milk production and quality, and also due to the costs that involve the control of the mastitis in dairy farms. This infirmity is characterized by an inflammatory process of the mammary gland, commonly caused by infectious microorganisms. Regarding the etiology in dairy goats, intramammary infections are mostly caused by several species of Coagulase-negative Staphylococcus (SCN) (McDougall et al., 2014;Gelasakis et al., 2016).
Among the several virulence factors, biofilm production is one of the main responsible factors for the persistence of intramammary infections. It is a structure formed by the organization of bacterial cells as a biopolymer matrix that surrounds the bacteria and attaches them to a surface. Bacterial adhesion to epithelial cells is usually the first stage of the infectious process and prevents the elimination of bacteria during milking. Biofilm formation occurs in successive stages, with its beginning characterized by the attachment of bacteria to a solid surface, where they multiply, and groups of cells accumulate on multilayers, resulting in the formation of a bacterial community. Biofilm protects the bacteria against the immune response of the host and the action of antimicrobial agents (Salaberry et al., 2015;Algharib et al., 2020).
Regarding the treatment of subclinical mastitis, besides the above-mentioned virulence factor, it is also important to take into account the cost, the drug elimination time, and the milk loss during the therapy period. The use of antimicrobials constitutes the main form of treatment for mastitis cases in the property. However, the high cost and the bacterial resistance to those compounds have led researchers to search for new alternatives to control this infirmity. The combination of natural compounds with commercial drugs has been a vast field for scientific research and may contribute to minimizing the resistance impact to pathogens (Mushtaq et al., 2018;Procópio et al., 2019).
Considering that conventional antimicrobial therapies present the disadvantage of selecting resistant bacteria, other mechanisms aiming at the control of infirmities have been sought. Therefore, controlling only the virulence of bacteria, but not their growth, seems to be a viable alternative. Plants might offer opportunities for the discovery of potentially bioactive molecules against the virulence factors of pathogenic organisms. Some studies have evaluated the ability of plant extracts in interfering with the mechanisms related to virulence factors, such as the potential of inhibiting or destructing biofilms or even interfering with quorum-sensing systems (QS) (Nazzaro et al., 2013;González-Ortiz et al., 2014).
According to Boniface et al. (2017), the plants of the genus Hymenaea are used in South America and Asia to treat several infirmities, among them, those caused by bacterial pathogens and parasites. The flavonoid astilbin has been reported as the main phenolic compound in this genus (Silva et al., 2012). Extracts containing high contents of phenolic compounds may be directly related to the antimicrobial activity (Dimech et al., 2013). In this perspective, this work aimed to perform the molecular characterization, evaluate the susceptibility to conventional antimicrobials and natural extracts of isolates that originated from cases of subclinical mastitis in dairy goats, as well as the biofilm production in interaction with the extract of Hymenaea martiana.
Location
This experiment was carried out in the Laboratory of Microbiology and Animal Immunology on the experimental farm of the Federal University of Vale do São Francisco -UNIVASF, Rodovia BR 407, km 12 -Lote 543 -Projeto de Irrigação Senador Nilo Coelho, s/nº "C1", Petrolina-PE.
Obtaining the ethanol extract of Hymenaea martiana Hayne
The plant material was collected in the district of Petrolina-PE and identified in the Reference Center for the Recovery of Degraded Areas (CRAD) of the Universidade Federal do Vale do São Francisco. One plant specimen (21,868) was deposited in the Herbarium of the Vale do São Francisco (HVASF). The crude and solvent fractions of the plant bark were prepared as previously described (Silva et al., 2012).
Phytochemical screening
For the phytochemical screening of the ethanolic crude extract (CEE), chromatography analysis was conducted using thin-layer chromatography on plates of silica gel 60, with aluminum support and fluorescence F254 in different solvent systems. This analysis was performed following previously described methodology (Wagner and Bladt, 1996) to investigate alkaloids, anthracene derivatives, coumarins, flavonoids, tannins, lignans, monoterpenes, diterpenes, naphthoquinones, triterpenes and steroids. HPLC-DAD analysis was performed to identify the profile of the phenolic compounds. The results obtained using HPLC-DAD indicated the presence of three compounds (flavonoid derivatives) (Peixoto et al., 2015).
Bacterial samples
In total, 33 isolates from the Laboratory of Microbiology and Animal Immunology were used. These isolates were first identified as belonging to the genus Staphylococcus based on their biochemical characteristics (Gram, catalase, coagulase, DNase and sugar fermentation). These isolates came from goats with subclinical mastitis in the district of Valente-BA. All of the isolates were tested for the presence of the nuc and rdr genes (Kateete et al., 2010;Shome et al., 2012), which would classify the microorganism as part of the Staphylococcus aureus or Staphylococcus epidermidis species, respectively. While investigating the nuc gene, DNA that was extracted from the bacterial isolates was used as template, with 4 l of the suspension added to 11 l of the mix containing 2 mM MgCl2, 0.4 pmol of the primers, 0.4 mM deoxyribonucleotides, 1× enzyme buffer and 1.5 U of Taq DNA polymerase. The amplification cycles of the nuc gene were conducted with a number of modifications: initial denaturation at 94°C for 5 min; 37 cycles, each at 94°C, for 1 min; hybridization of the primer at 55°C for 30 s; extension at 72°C for 1 min; and a final extension at 72°C for 7 min (Kateete et al., 2010).
Polymerase chain reaction (PCR) was used to study the rdr gene with a number of modifications: 5 l of the DNA suspension was added to 20 l of the mix containing 1.5 mM MgCl2, 0.5 pmol of the primers, 0.4 mM deoxyribonucleotides, 1× enzyme buffer and 1.5 U of Taq DNA polymerase. The amplification cycles of the rdr gene involved an initial denaturation at 94°C for 5 min; 30 cycles, each at 94°C, for 30 s; hybridization of the primer at 60°C for 30 s; extension at 72°C for 45 s; and a final extension at 72°C for 10 min (Shome et al., 2012).
The results of the PCR were confirmed with a 1.5% agarose gel that was stained with ethidium bromide (1.0 mg mL -1 ) and an UV transilluminator was used to see the bands. Table 1 describes the primers that were used.
Phenotypic test of the sensitivity to antimicrobials
The sensitivity profile of the microorganisms was determined using the Kirby-Bauer disc diffusion method. The isolates were seeded in Mueller Hinton broth and incubated at 37ºC until a turbidity of 0.5 on the McFarland scale was obtained. A swab was used to seed the isolates in Petri dishes containing a medium of Mueller Hinton agar culture. Subsequently, discs that were impregnated with the following drugs were applied: ampicillin (10 μg), doxycycline (30 μg), lincomycin (2 μg), erythromycin (15 μg), gentamicin (10 μg), rifampicin (30 μg), cephalothin (30 μg), amoxicillin (10 μg), nalidixic acid (30 μg) and oxacillin (1 μg). The plates were incubated in an oven for 24 h at 37ºC. The diameter of the zone of inhibition was measured to determine the sensitivity profile of the isolates (CLSI, 2019). Afterwards, the multiple antibiotic resistance index (MARI) was calculated.
Molecular analysis of the genetic potential for resistance to antimicrobials
The aim of the PCR was to assess the presence of genes of resistance. To achieve this aim, it was necessary to extract DNA from the isolates in advance. The DNA was extracted and purified based on modified versions of previously described protocols (Ausubel et al., 1989;Aldous et al., 2005). A loopful culture was placed in 300 µl of TE (Tris-EDTA), and the samples were vortexed for homogenization. Subsequently, 70 µl of 10% SDS was added, and the samples were homogenized once again. In the next step, 100 µl of 5 M NaCl2 and 80 µl of CTAB/NaCl were added. The samples were incubated at 65°C for 10 min. Then, 700 µl of chloroform/isoamyl alcohol (24:1) was added, and the samples were homogenized by inversion. The samples were then centrifuged at 11,750 g for 5 min. The first phase was transferred to another tube, and 450 µl of isopropanol was added. The tubes were inverted and left on ice for 20 min. Thereafter, the tubes were centrifuged at 11,750 g for 15 min; subsequently, the supernatant was discarded, and 500 µl of 70% ethanol was added. Next, the samples were centrifuged at 11,750 g for 10 min. The supernatant was then discarded, and the microtubes were inverted to facilitate drying. Finally, the samples were suspended in 50 µl of TE (pH 8.0) and left at 65°C for 10 min. Once completed, the samples were stored at -20°C.
The genetic potential for resistance to antimicrobials was analyzed by amplifying the mecA and blaZ genes for resistance to methicillin (oxacillin) and beta-lactam, respectively. The DNA that was extracted from the bacterial isolates was used as a template to amplify a fragment of 214 bp corresponding to the mecA gene and a fragment of 517 bp corresponding to the blaZ gene. In total, 4 l of this suspension was added to 11 l of a mix containing 2 mM MgCl2, 0.4 mM primers, 0.4 mM deoxyribonucleotides, 1x enzyme buffer and 1.5 U of Taq DNA polymerase. The amplification reaction of the mecA gene involved an initial denaturation at 94°C for 1 min, 15 cycles of denaturation at 94°C for 30 s, 68°C for 30 s, 72°C for 30 s, an additional 20 cycles of 30 s at 94°C, 30 s at 60°C, 30 s at 72°C and a final extension at 72°C for 2 min (Murakami et al., 1991). The amplification reaction of the blaZ gene involved an initial denaturation at 94°C for 4 min, 30 cycles at 94°C for 1 min, 50.5ºC for 30 s, 72°C for 30 s, and a final extension at 72°C for 5 min (Sawant et al., 2009). Table 1 describes the primers that were used.
Molecular analysis of the genetic potential for the formation of biofilm (icaD, icaA and bap)
The DNA that was extracted from the bacterial isolates was used as a template. For the icaD gene, 2 l of this suspension was added to 13 l of the mix containing 1.33 mM MgCl2, 1 mM primers, 0.2 mM deoxyribonucleotides, 1x enzyme buffer and 1.5 U of Taq DNA polymerase. The reaction was kept in the thermocycler and submitted to an initial denaturation at 94°C for 2 min, followed by 30 cycles of 45 s at 92°C, 49.8°C for 45 s and 1 min at 72°C, with a final extension at 72°C for 7 min (Vasudevan et al., 2003).
For the icaA gene, 5 µL of the suspension (template) was added to 20 µL of the mix containing 2.5 mM MgCl2, 1 pmol of the primers, 0.28 mM deoxyribonucleotides, 1× enzyme buffer and 2.5 U of Taq DNA polymerase. The amplification reaction of the icaA gene was conducted as previously described (Vasudevan et al., 2003) with a number of modifications: initial denaturation at 94°C for 2 min, 30 cycles of 45 s at 92°C, 58.6°C for 45 s and 1 min at 72°C, with a final extension at 72°C for 7 min.
To study the bap gene, 5 µL of the suspension (template) was added to 20 µL of the mix containing 2.0 mM MgCl2, 0.4 pmol of the primers, 0.4 mM deoxyribonucleotides, 1× enzyme buffer and 2.5 U of Taq DNA polymerase. The amplification reaction of the bap gene was conducted with a number of modifications: initial denaturation at 94°C for 2 min, followed by 35 cycles of 45 s at 9°C, 56.5°C for 45 s and 50 s at 72°C, with a final extension at 72°C for 5 min (Cucarella et al., 2001). Table 1 describes the primers that were used.
Phenotypic test to detect the presence of biofilm
The phenotype of the isolates, in relation to the formation of biofilm, was characterized using the plate adherence test. The colonies that were isolated were inoculated in 3 mL of Tryptone Soya Broth (TSB) with glucose (0.25%) and incubated at 37°C for 24 h. Next, 200 µl was inoculated in micro-dilution plates and incubated again at 37°C for 24 h. After this period had elapsed, the plates were washed three times with 200 µl of distilled water and left to dry at room temperature. The plates were stained with 100 µl of crystal violet 0.25% for 2 to 3 min at room temperature and washed three more times with distilled water. To dissolve the dye, 200 µl of ethyl alcohol was used (80:20). The absorbance was measured using an Elisa Easy® micro-plate reader and a filter of 620 nm. All of the samples were analyzed in triplicate, as were the positive and negative controls. A strain of S. aureus ATCC 25923, which is genotypically characterized as a biofilm producer, was used as the positive control while a strain of S. epidermidis was used as the negative control. It was possible to determine biofilm production using the following classification: no biofilm production (optical density (OD) sample ≤ OD negative control); weak biofilm production (OD negative control < OD sample ≤ 2 OD negative control); moderate biofilm production (OD negative control < OD sample ≤4 OD negative control) and strong biofilm production (OD sample <4 OD negative control) (Merino et al., 2009).
Test of sensitivity to the extract
The crude ethanolic extract of H. martiana and the fractions that were obtained in ethyl acetate, hexane and chloroform were used in this test. First, 0.25 g of each extract was weighed and diluted in 10 mL of 1% DMSO, obtaining a stock solution with a concentration of 25 mg mL -1 . The following concentrations were used: 12,500; 6,250; 3,125; 1,562.5; 781.2; 390.6; 195.3; and 97.6 µg/mL.
To prepare the inoculum, colonies that were obtained in Mueller-Hinton agar were used to create a bacterial suspension with turbidity of 0.5 on the McFarland scale. Of this suspension, 10 µL was inoculated in the wells of micro-plates containing the dilution of ethanolic extract. The material was incubated at 37°C for 24 h under aerobic conditions. The minimum inhibitory concentration (MIC) was determined to confirm the lowest concentration of the extract that was capable of inhibiting bacterial growth. When no bacterial growth was visible, one aliquot of 10 µl was withdrawn, seeded on the MH agar surface and incubated for 48 h at 37ºC. The following procedure was followed to determine the minimal bactericidal concentration (MBC): one aliquot of 10 μL was withdrawn from the wells without visible bacterial growth and seeded on the surface of the Mueller-Hinton agar. After 48 h of incubation at 35°C, the MBC was defined as the lowest concentration of the ethanolic extract that was required to cause the death of the inoculum. All of the trials were performed in triplicate.
Interaction of the extract with biofilm in the formation and consolidation of biofilm
Eleven (n=11) isolates that were identified as S. aureus were used to perform these tests. The biofilm in micro-plates was formed by incubating 100 µl of the bacterial suspension for 24 h at 37°C. Subsequently, the wells were washed three times with distilled water, and 100 µl of the extract was added, as described: 1/2, 1/4 and 1/8 of CBM were used for the tests. The optical density (OD) was determined immediately after the extract was added (0 h) and 24 h later.
The effect of the extract on the consolidated biofilm was defined by the following equation: OD0 h mean /OD24 h mean × 100 (Nostro et al., 2007).
Based on the MBC results that were obtained, three different concentrations (1/2, 1/4 and 1/8 of CBM) were used with biofilm in formation. This trial was conducted in micro-plates. The bacterial inocula were cultivated in 10 mL of TSB (1% glucose) for 24 h at 37°C. In total, 100 µl was added to the well plates, while 100 µl of vegetable extract and 100 µl of the culture media were previously added to the controls. After 24 h of incubation at 37°C, the plates were submitted to Gentian violet staining (Merino et al., 2009). The effectiveness of the extract in interfering with the formation of biofilm was defined by the following equation: (Nostro et al., 2007).
Scanning electron microscopy
To assess the effect of the H. martiana extract on biofilm formation, an isolate of S. aureus that had been classified as a strong producer of biofilm was used to test for plate adherence. The control sample was inoculated in TSB broth. The same isolate was also inoculated in TSB broth containing the plant extract. Both were incubated at 37°C for 24 h. After the inoculum had been withdrawn from the suspension, it was inoculated in cover slips, which were then washed with sterilized saline solution for one minute. The slips were then fixed in glutaraldehyde (1%) for a period of 12 h. After fixation, the slides were immersed in increasing concentrations of ethanol (50, 70, 80, 95 and 100%), with 20 min between each alteration. After dehydration, the samples were immersed in acetone 100% and subjected to gold coating (Freitas et al., 2010). The fragments were studied using the TM-1000 Hitachi scanning electron microscope.
Statistical analysis
Descriptive statistics were used, including the distribution of relative and absolute frequencies, for the microbiological and molecular findings. Friedman's analysis of variance was used to compare the MBC values in the three fractions of the extract. The Mann-Whitney test was used to assess the differences between the MBC values that were obtained for strong and moderate biofilm producers and for weak or non-producing isolates. The same test was used to compare the results of the MBC values between the isolates that were positive and negative for the blaZ gene (Thrusfield, 2007).
Phytochemical screening
The phytochemical screening of the CEE confirmed the presence of phenolic substances, flavonoids, steroids, and terpenoids (Table 2).
Bacterial samples
The molecular tests performed for the investigation of the nuc and rdr genes demonstrated that 97% (32/33) were positive for the nuc gene, being identified as S. aureus.
Only one isolate was positive in the investigation of the rdr gene, indicating S. epidermidis.
Phenotypic test for sensitivity to antimicrobials
The sensitivity profile to antimicrobials of the evaluated isolates presented variation, which can be seen in Figure 1. In the susceptibility testing to antimicrobials, those with the highest percentage of in vitro sensitivity were gentamicin (97%) and doxycycline (97%), followed by erythromycin, with 76%, rifampicin, lincomycin, and cephalothin, with 70%, and with no great variation between oxacillin, ampicillin, and amoxicillin, with 45, 52, and 48%, respectively. The drug with the lowest sensitivity percentage was the nalidixic acid, with 30%. Among the evaluated samples, the IRMA varied from 0 to 0.8, of which 51.5% (17 isolates) presented this index above 0.2; that is, were resistant to two or more antimicrobials.
Molecular analysis of the genetic potential for resistance to antimicrobials
The resistance to oxacillin was observed in 18 (54.5%) isolates; however, no amplification by PCR of the mecA gene was observed. Out of the 33 isolates evaluated, 11 (33.3%) were positive for the blaZ gene.
Molecular analysis of the genetic potential for biofilm formation
Out of the 33 evaluated samples, nine (27.2%), three (10.0%), and four (12.9%) presented the icaD, icaA, and bap genes, respectively. Only two samples were positive at the same time for the icaD and icaA genes.
Test of sensitivity to the extract
In the in vitro sensitivity tests to the crude ethanolic extract of H. martiana (EEBHm), the means obtained for the MBC of the crude ethanolic extract and its fractions are presented in Table 3, and no significant difference was observed (p >0.05). For the crude ethanolic extract, a median of 2864 µg/mL was observed among the isolates classified as moderate or strong biofilm producers. On the other hand, among the isolates classified as negative or weak biofilm producers, the mean MBC obtained for the crude extract of H. martiana was equal to 1563 µg/mL (p=0.07). Among the 11 positive isolates for the blaZ gene, a median of 781 µg/mL was observed. On the other hand, among the negative isolates for the presence of this gene, the median was equal to 2604 µg/mL, although without a significant statistical difference (p=0.14).
Interaction between the extract and biofilm in the formation and consolidation of biofilm
It was observed that, after the addition of the extract, 81.8, 90.0, and 81.83% of the isolates presented a reduction in biofilm production in the ethyl acetate, hexane, and chloroform fractions, respectively. Regarding the observation of biofilm after its establishment, no reduction in production was observed in any of the isolates (Table 4).
Scanning electron microscopy
In the control sample, which had no contact with the plant extract, a wide extracellular matrix production was observed. On the other hand, in the sample that remained in contact with the extract for 24 h, no biofilm production was observed (Figure 2).
DISCUSSION
Hymenaea is a genus that is distributed throughout tropical America, from Mexico to Paraguay, with one species located in coastal East Africa (Mackinder, 2005). The phytochemical screening of the CEE confirmed the presence of phenolic substances, flavonoids, steroids and terpenoids. Flavonoids are secondary metabolites and can be found in moderate concentrations in the extract that was used in the present study. The phytochemical screening of the crude ethanolic extract confirmed the presence of phenolic substances, flavonoids, steroids and terpenoids, as displayed in Peixoto et al. (2015).
In the test of susceptibility to antimicrobials, it is clear that there were only small variations between betalactams, with a similar resistance profile to the drugs in this group. Under similar experimental conditions, researchers recorded strong resistance among isolates from cases of subclinical mastitis in small ruminants (França et al., 2012). It is notable that a number of isolates exhibited resistance to oxacillin. A previous study expressed concern about the increasing occurrence of multi-drug resistance among microorganisms isolated from small ruminants with mastitis (Silva et al., 2004). Regarding the drugs in the group known as aminoglycosides, a high number of sensitive isolates were recorded, corroborating the findings of other studies (França et al., 2012;Kumar et al., 2009).
The mecA gene was not detected in the samples that were tested in the present study. A previous study reported that isolates that are resistant to oxacillin in several phenotypic tests cannot exhibit this gene given that the phenotype that produces beta-lactams, including class D oxacillinases, is resistant to penicillin (Soares et al., 2008). However, the assessments of the blaZ gene confirmed a positivity of 33.3% (n=11). Of this total, four (4) isolates were resistant to the beta-lactams that were tested. The results indicate the codified production of beta-lactam by the blaZ gene, a mechanism that has been reported in dairy cattle from other countries (Olsen et al., 2006;Taponen and Pyörälä, 2009). Although antimicrobial therapy focusing on mastitis in the sampled herds is not very common, different drugs are used for other infections in small ruminants. The generalized use of certain antimicrobials for clinical treatment could be associated with a high rate of resistance.
An analysis of the MARI scores indicated that more than half of the isolates exhibited indices that were greater than 0.2. This pattern of multi-resistance to antimicrobials has increased due to the exaggerated use of these drugs (Ferreira et al., 2006). The indices of resistance that were recorded in the present study could have been caused by the incorrect management methods that were used on the properties. Consequently, this pattern provides the lowest number of drugs available for use in the therapy of mastitis.
Among the isolates that were positive for the blaZ gene (n=11), nine exhibited mean MBC values less than 2,604 µg/mL, suggesting antimicrobial activity in the extract studied against the isolates that were carrying genes of resistance. However, it is essential to perform further studies to elucidate the mechanisms of the antibacterial action of the extract.
A molecular analysis of biofilm production confirmed the presence of the icaD, icaA and bap genes, although their frequencies were low. Only one of the isolates exhibited the concomitant presence of the icaD and icaA genes, which are responsible for the synthesis of biofilm. The genetic mechanisms that determine the production of biofilm are complex and involve other genes that were not investigated in the present study, including the IS257 gene (Tormo et al., 2005). A previous study demonstrated that molecular techniques to identify the ica genes that codify the synthesis of biofilm are important tools in the identification of virulent strains (Arciola et al., 2002). The implication of biofilm in chronic bacterial infections in many species has unleashed a growing interest in the characterization of the genes involved in its formation (Vautor et al., 2008). The bap gene was recently identified as the gene that is responsible for the codification of a protein associated with biofilm. However, currently, this gene has only been found in a small proportion of S. aureus strains for bovine mastitis in Spain (Cucarella et al., 2001). A study in France investigated the presence of the bap gene in S. aureus isolates from different species and regions using PCR and found negative results for all of the isolates, suggesting that the prevalence of this gene among S. aureus isolates must be very low (Vautor et al., 2008).
Among the strains that were classified as negative or weak producers of biofilm in the present study, the best antimicrobial activity was recorded for the crude ethanolic extract of H. martiana. Furthermore, the production of biofilm decreased after the addition of the plant extract, thereby demonstrating its potential in terms of affecting the production of this extracellular matrix. However, no reduction was recorded when the extract was added after the consolidation of biofilm. The reductions in biofilm production was also confirmed by scanning electron microscopy and suggest in vivo biological activity among the compounds present in the natural extract, mainly flavonoid derivatives. Biofilm affects the activity of antimicrobials due to low penetration and to the phenotypically protected state of the bacteria that compose the structure (Jacques et al., 2010;Altieri et al., 2013). Previous studies have listed the following strategies to reduce biofilm: prevention of microbial fixation; prevention of microbial growth; interruption of cell-cell communication; inhibition of the synthesis of the matrix; and disintegration of the matrix of biofilm. The inhibition of biofilm formation depends on factors that are associated with the enzymatic inhibition of proteases and quorum-sensing mechanisms (González-Ortiz et al., 2014). Because we are dealing with a natural antimicrobial, the reduction in the production of biofilm after interaction between the isolate and the H. martiana extract requires further research that will enable a better understanding of the action of this extract. A previous study revealed that the use of inhibitors could drastically alter the treatment of many infectious diseases (Cegelski et al., 2008). The basic strategy to discover biofilm inhibitors involves the screening of chemical compounds during trials that assess the effects of drugs or extracts on the formation of biofilm (Landini et al., 2010). The same study stated that till date, studies have focused on "quorum sensing" inhibitors, compounds that affect the metabolism of the c-di-GMP molecule or inhibitors of the biosynthesis of DNA and nucleotides.
Although studies of alternative therapy for diseases in veterinary medicine are common, they are characterized by discontinuity, with very small numbers of studies about the same plant. Thus, the minimal data that are required to use plants in veterinary medicine are not always obtained, as pre-clinical and clinical trials are not carried out.
Conclusion
The results that were obtained in the present study confirm the possibility of treating infections of Staphylococcus spp. among goats using gentamicin. The low frequency of genes that are involved in the production of biofilm suggests heterogeneity in the genetic origins of the isolates. The variation between the phenotypic and genotypic results demonstrates that the mechanisms involved in producing this matrix are quite complex. Future studies are fundamental in this molecular area, as well as researching other associated genes in the literature. A reduction in the production of this extracellular matrix was recorded upon contact with an ethanolic extract of H. martiana, demonstrating its antimicrobial potential in mastitis caused by Staphylococcus spp. Further studies are required to elucidate the mechanisms that cause the low production of biofilm when isolates come into contact with the extract of this plant, thereby ensuring continuity in this area of phytotherapy, particularly in relation to the H. martiana plant.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-07-23T09:09:28.690Z", "created": "2020-07-31T00:00:00.000", "id": "225517208", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://academicjournals.org/journal/AJMR/article-full-text-pdf/A7783D164227.pdf", "pdf_hash": "227619c9cd0b3c78c65dfe28e421353683fd643a", "pdf_src": "Adhoc", "provenance": "peS2o-0075.json.gz:31496", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "2814aec428c0f98f206e2f30d67dba9085a00291", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Weaning performance of beef Hungarian Fleckvieh calves : 3 . Genotype × environment interaction
The interaction of sire and population in Hungarian Fleckvieh beef cattle breed were examined in this study on data from the Hungarian Fleckvieh Breeders Association. Data of 2 345 progeny (1 260 male and 1 085 female), born between 1992 and 2003, of 35 sires from two populations were evaluated. Preweaning daily gain (PDG) and 205-day weight (205-dw) were analysed. Population, age of cows, year of birth, season of birth and sex of calves as fixed, sire and sire × population were treated as a random effect. Among the same performance data in the two populations (A, B) genetic correlation (rg), while by the gradiation of sires rank correlation (rrank), were evaluated. Data were analysed with HARVEY’S (1990) Least Square Maximum Likelihood Computer Program and SPSS 9.0 for Windows. Results were as follows: rg=PDGA−PDGB: 0.31(P<0.01); 205-dwA−205-dwB: 0.22(P<0.01) and rrank=PDG: −0.04(P>0.05); 205-dw: 0.078(P>0.05). According to the result of examination important and significant (P<0.001) sire × population interaction were found in case of the two traits in Hungarian Fleckvieh breed.
Introduction
In the former publications of this series of articles was reported about factors, which have an influence on the weaning performance of Hungarian Fleckvieh calves, about population-genetic parameters of the weaning traits and the estimated breeding value. In this work the experiences will be showed which were obtained in the examinations of genotype × environment interaction, appearing in weaning performances.
It is well-known for a long time that the traits of different populations, which can be measured phenotypically, not always change in the same way owing to the effect of the different environmental factors (WILSON 1974). It was proved by the result of researches that the different genotypes can react upon the different environmental factors in different way, too. It means that the animals with different genetic construction can react upon to the environmental factors in different way or the animals with the same genotype show different phenotypic value under different environmental conditions (HORN and DOHY 1970). The genotype × environment interaction can have several occurrences. DICKERSON (1962) mentioned some environmental factors, which can have different influences on the performances of populations with different genotype and can cause an interaction. For the examination of genetic × environmental interactions, for the estimation of interaction component can applied a two-or more-factor variance analysis (HORN 1978). FALCONER (1952) suggested determining the interactions in the different environments with the genetic correlations between the results of the production because any traits measured in two different environments can be treated from genetic point of view as two different traits (FALCONER and MACKAY 1996). If the estimated genetic correlation is close, the rank-line of the estimated breeding animals will be unrelated to the environment, whereas the loose genetic correlation or the total independence of traits means that the rank-line of the examined breeding animals evaluated in two different environments can considerably differ from each other. By calculation of rank-correlation coefficient informing data can be obtained referring to the type of genotype × environment interaction, where there are differences in the rank-line, which was built up according to the mean performances of the examined genotypes (HORN 1978). VOSTRY et al. (2009) found six Czech beef cattle races for weaning weight that environment interaction was not biologically important and can be ignored in the evaluation of beef cattle in the Czech Republic.
The aim of this work was to estimate the weaning performances of Hungarian Fleckvieh calves in a way, which wasn't applied so far and to obtain newer data about the performance of Hungarian Fleckvieh bulls in different environments, about the breeding value and the genotype × environment (sire × population) interaction.
Material and methods
The examinations were made on the database given by the Association of Hungarian Fleckvieh Breeders. In the evaluation the data of 2 345 calves (1 260 bulls and 1 085 heifers) born between 1992 and 2003 in two populations were used. The examined traits were the preweaning daily gain (PDG) and the 205-day weight (205-dw). From the evaluated factors age of cows at calving, year of birth and sex were treated as fixed effects, the sire and the sire × population interaction as random effects. The age of calves -from the birth to the weaning -was a covariant factor in case of preweaning daily gain. The Table 1 shows the models applied for the estimation of the effects of the several traits.
The database included the data of calves descending from 35 breeding bulls in two populations -population A and population B. Each bull had calves in both populations. The general form of the model applied for the preweaning daily gain is as follows: ( ) where Ŷijklmno is the weaning weight and gain/life day of the calf, whose age is o, sex is n, from the sire i, whose age is k, in the population j, in the season l, from a cow whose age is m; μ is the mean value of all observations, El is the fixed effect of the season of birth, Si is the random effect of sire, Cm is the fixed effect of age of cow at calving, Hj is the fixed effect of population, In is the fixed effect of sex, SHij the random effect of sire × population interaction, Yk is the fixed effect of the year of birth, b is the random effect of the regression coefficient and eijklmno is the random residual.
The method of evaluation of the 205-day weight differs from the former one so far as the age of the calves as covariant wasn't included by the model. The model was following: For the estimation of breeding value of bulls sire model was applied. The sire model is a mixed model, which takes into consideration the fixed and the random effects as well. It differs from the animal model so far as it is necessary to know the sire only; the other family relations of the animal aren't needed. The estimation was made by HARVEY'S (1990) Least Square Maximum Likelihood Computer Program.
The genetic correlations between the herds we calculated among the genetic values of the given trait with the following formula: where rg is the genetic correlation, σ 2 G1 is the variance of the given trait in one of the populations, σ 2 G2 is the the variance of the given trait in the other population and σG1G2 is the covariance of the two traits. The rank-line of the breeding bulls we calculated by a rank-correlation coefficient.
Data were arranged with Microsoft Excel XP program while variance analysis resp.rank-correlation coefficient calculation with SPSS 9.0 software.
Results and discussion
According to the results of the examination -as it can be seen in the Table 1 -the age of cows, year, season, sex, sire × population interaction and the age at weaning had a significant (P<0.001)influence on the preweaning daily gain and the 205-dw. These results are similar to the results of SZABÓ et al. (2006), LENGYEL et al. (2003b), and TŐZSÉR et al. (1996).
The contribution of examined factors to the total variance is shown by the Table 2. It can be seen that the sire and the population by itself didn't influence both traits, but they together influenced them significantly. In case of 205-dw the greatest effect had the sex (51 %). It is similar to the results of LENGYEL et al. (2003b), andKOVÁCS et al. (1993). In case of the preweaning daily gain the age of cows had the greatest effect (27.7 %). This value differs from the results of SZABÓ et al. (2006). The interaction component was in case of both traits the fifth most important source of variance, it gave not more than 5.23-3.08% of the total variance. A significant interaction was observed by MÜLLER (1991), NOTTER et al. (1992), FERREIRA et al. (2001), DE SOUZA et al. (2003), IBI et al. (2005). The Table 3 includes the genetic correlations calculated among the performance data in the two populations. According to ROBERTSON (1959) the genotype × environment interaction is important, when the genetic correlation between the same traits measured in the different populations is smaller than 0.8. In the results it can be seen that the genotype × environment interaction was of great importance in case of both traits, because small (rg=0.22-0.31)genetic correlation coefficients were obtained. It is in accordance with the results of SOTO-MURILLO et al. (1993), FERREIRA et al. (2001), DE SOUZA et al. (2003). The Table 4 shows the breeding value of the evaluated bulls. The Figure shows the rankline of bulls according to the estimated breeding values.
It can be seen in the figure that the sire × population interaction was so high that it caused change in the rank-line. The calculated rank-correlation coefficient ( The results are similar to the statements of MÜLLER (1991) andNOTTER et al. (1992), who found such sire × population interactions, which caused a change of the rank correlation of breeding bulls in the populations and regions.
Summing up the results it can be stated that an important sire × population interaction could be found relating to the preweaning daily weight and the 205-day weight in the breed Hungarian Fleckvieh. The genotype × environment interaction was found so high, that it caused the change of the rank-line of sires according to the weaning performances. This interaction calls the attention that it is to go about carefully in the evaluation of the Hungarian Fleckvieh sires if they aren't used in the same population in which they were ranked. The reliability of evaluation of the breeding value can be lower, if the interactions will be left out of consideration. To eliminate this, the mathematic model of several evaluating methods for the breeding value (e.g. BREEDPLAN) takes the sire × population interaction in account.
Table 2
The contribution of source of variance to total variance, % Das Verhältnis der Varianzquellen in der Gesamtvarianz, in %
Table 4
Breeding value of sires of populations A and BZuchtwerte der Bullen der Populationen A und B Table5) was rrank=PDG: −0.04; 205-dw: 0.07 and it wasn't significant. It means that you can‹t draw a conclusion from the measuring of the performance of the genotypes in one environment relating to the direction and nature of changes of performance, which can be expected in the other environment.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-03-30T13:11:57.534Z", "created": "2010-10-10T00:00:00.000", "id": "85730260", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://aab.copernicus.org/articles/53/123/2010/aab-53-123-2010.pdf", "pdf_hash": "21c70c7f746e55b2d4291fad3ca176ac0c36520a", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:30256", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "21c70c7f746e55b2d4291fad3ca176ac0c36520a", "year": 2010}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Apparent digestibility of nutrients , energy , and amino acid of nontoxic and detoxified physic nut cakes for Nile tilapia
The objective of this work was to evaluate the apparent digestibility coefficients of nutrients, energy, and amino acids of nontoxic and detoxified physic nut cakes treated with solvent plus posterior extrusion, for Nile tilapia. The apparent digestibility coefficients of crude protein and gross energy were higher for detoxified than for nontoxic physic nut cake. However, the apparent digestibility coefficient of ether extract of the nontoxic physic nut cake was higher than that of the detoxified one. The apparent digestibility coefficient of amino acids of both feed ingredients was superior to 80%, except for glycine, for the nontoxic psychic nut cake, and for threonine, for the detoxified one. Nontoxic and detoxified physic nut cakes show apparent digestibility coefficient values equivalent to those of the other evaluated oilseeds and potential for inclusion in Nile tilapia diets.
The physic nut (Jatropha curcas L.) crop is increasing in the tropic and subtropic regions due to its resistance to drought, high oil content (60%) and quality, which are suitable for biofuel production (Kumar et al., 2011). The cake/meal obtained after oil extraction contains high nutritional value, with 25-63% crude protein, 4.0-14.21%ether extract, 36.68-45%crude fiber, and 12.82 KJ g -1 gross energy, showing good amino acid profile, except for lysine (Martínez-Herrera et al., 2006).
Despite the potential of physic nut cake/meal as a protein source in animal feeding, its use is limited mainly due to the presence of phorbol ester, a toxic constituent, as well as of anti-nutritional factors, such as phytates, lectins, protease inhibitors, non-starch polysaccharides, saponins, and high fiber contents (Makkar et al., 1998b;Kumar et al., 2011). Therefore, further research is needed to recommend the use of physic nut cake/meal as an alternative feedstuff.
In aquafeeds, different physic nut varieties have been evaluated: detoxified (Kumar et al., 2011); and nontoxic, after roasting, from the Quintana Roo, Veracruz, and Morelos regions of Mexico (Makkar et al., 1998a). The knowledge of the digestibility of agro-industrial by-products is essential to allow their use in balanced diets for fish (Pezzato et al., 2002). Some studies have shown the potential of physic nut cake/meal in fish diets (Makkar & Becker, 1999;Kumar et al., 2011); however, information about the digestibility of nutrients and amino acids is limited.
The objective of this work was to evaluate the apparent digestibility coefficients of nutrients, energy, and amino acids of nontoxic and detoxified physic nut cakes treated with solvent plus posterior extrusion for Nile tilapia (Oreochromis niloticus).
The nontoxic physic nut seeds were obtained from a private farm and registered as CNPAE 169 and CNPAE 170 in the germplasm bank of Embrapa Agroenergia. The seeds were dried for 30 min, at 100ºC, in a rotary dryer and then mechanically processed in a ERT 60 expeller press (Scott Tech, Vinhedo, SP, Brazil) for oil extraction, to obtain the nontoxic cake, at Embrapa Agroenergia. The detoxified cake was produced using a mechanically-extracted toxic cake, subjected to a sequential solvent extraction with hexane at 45-50ºC (30 min; 1.6 L hexane per kg of cake), followed by three more wash-out extraction cycles with the same solvent at 45-50ºC (15 min each) and a new extraction with ethanol at 60-65ºC (30 min; 1.6 L ethanol per kg of cake), followed by three more wash-out cycles with the same solvent at 60-65ºC (15 min each). This process was performed at the Ercitec pilot plant unit of Instituto de Tecnologia de Alimentos, in the municipality of Campinas, in the state of São Paulo, Brazil. The cake was conditioned to 21,5% of moisture and then extruded in a Clextral twin-screw extruder (Clextral, Firminy, France), with the addition of 2% sodium hydroxide, at 160ºC, at Embrapa Agroindústria de Alimentos, in the municipality of Rio de Janeiro, in the state of Rio de Janeiro, Brazil.
The trial of apparent digestibility coefficients was conducted at the fisheries laboratory of Embrapa Agropecuária Oeste, in the municipality of Dourados, in the state of Mato Grosso do Sul, Brazil. Sixty fish with average weight of 55 g were stocked in three cylindrical 70-L plastic cages, totaling 20 fish per cage, allocated in a 1,000-L cylindrical tank with recirculating water system. In addition, three 250-L cylindrical tanks with conical bottom were used for feces collection.
Prior to feces collection, Nile tilapia juveniles were adapted to the experimental diets for 5 days. During the trial period, fish were fed ad libitum, and the procedure described by Pezzato et al. (2002) was adopted for feces collection, which was performed every 2 days in order to minimize stress. The following water quality variables were measured daily in all tanks and remained within acceptable values for tilapia: temperature, 27.41±0.75ºC;dissolved oxygen, 6.12±0.11mg L -1 ; total ammonia, 0.06±0.01mg L -1 ; and pH, 7.6±0.31.
The experimental diets corresponded to: nontoxic physic nut cake; detoxified physic nut cake treated with solvent plus extrusion; and a reference diet formulated according to the National Research Council (Nutrients…, 1983), with soybean meal as a major protein source, in order to calculate the digestibility of the test ingredients (Pezzato et al., 2002). The nontoxic and detoxified physic nut cake diets consisted of the replacement of 30% of the reference diet by a test ingredient. Chromium oxide (Cr 2 O 3 ) 0.1% was used as an inert marker.
The ingredients were ground to <0.5 mm, moistened with approximately 20% water, and processed into a 3.0-5.0-mmdiameter in a laboratory pellet mill. Crude protein, ether extract, gross energy, neutral detergent fiber, acid detergent fiber, chromium, and amino acids were determined according to the AOAC methods of analysis (Horwitz & Latimer Jr., 2010).
The following anti-nutritional compounds were also determined: trypsin inhibitors, phytates (Horwitz & Latimer Jr., 2010), and the toxic compound phorbol ester (Makkar et al., 1998a). A completely randomized design was used, with two treatments (nontoxic and detoxified physic nut cakes) and four replicates. The data were analyzed by the SPSS statistical software, version 13.0 (IBM, Armonk, NY, USA), by Student's test, at 5% probability. The means of the nutritional composition of nontoxic and detoxified physic nut cakes corresponded to the average of three replicates, and the means of amino acids (values and apparent digestibility coefficients) to a polled sample.
The crude protein values obtained in the present study for nontoxic and detoxified physic nut cakes (Table 1) were similar to those reported by Guedes et al. (2014), of 12.55-18.07%,but lower than those found by Aderibigbe et al. (1997), of 45.8-61.2%, and by Makkar & Becker (1999), of 35.3%. In these studies, the cake was processed from the kernel (dehulled seed), where most protein is located, which increased protein content.
The values of the evaluated anti-nutritional factors, for the nontoxic and detoxified physic nut cakes, were, respectively: 16.75 (1.67%) and 21.49 mg g -1 (2.15%) phytates; and 10,302.41and 1,153.89g -1 units of trypsin inhibitor activity. Phorbol ester, however, was not detected for the nontoxic physic nut cake, whereas 0.023 mg g -1 phorbol was obtained for the detoxified physic nut cake. Makkar et al. (1998a) analyzed the previously mentioned anti-nutritional factors of physic nut meals and obtained higher values than those found in the present study. According to these authors, variations in anti-nutritional contents can be explained by differences among genotypes.
Soybean (Glycine max L.) meal is the most widely used plant protein source in fish diets, due to its high digestibility of nutrients, energy, and amino acids. In the present study, soybean meal, as well as some cakes/ meals from other oilseeds with potential for biofuel production, was used as a comparative reference to physic nut cakes. The apparent digestibility coefficient of crude protein differed between nontoxic and detoxified physic nut cakes (p<0.05),corresponding to 77.51 and 81.11%, respectively (Table 1). Riche et al. (2001) found similar results (84.40%) for soybean meal fed to tilapia. However, while evaluating Nile tilapia, Pezzato et al. (2002), Guo et al. (2011), and Zhou & Yue (2012) observed higher values of apparent digestibility coefficients of crude protein for soybean meal, with values ranging from 91.56 to 94.90%. Compared to cakes/meals derived from oilseeds with potential for biofuel production, the apparent digestibility of crude protein of nontoxic and detoxified physic nut cakes was similar to those reported for: radish (Raphanus sativus L.) forage meal (Santos et al., 2010); palm kernel (Elaeis guineensis Jacq.) meal (Braga et al., 2010); cottonseed (Gossypium hirsutum L.), rapeseed (Brassica napus L.), and peanut (Arachis hypogaea L.) meals (Zhou & Yue, 2012), with values of 82.10, 75.87, 76.70, 77.60, and 77.60%, highlighting the biological value of protein of the nontoxic and detoxified physic nut cakes for Nile tilapia. Furthermore, the nontoxic physic nut cake showed higher values of acid detergent fiber and neutral detergent fiber than the detoxified one, which can explain the reduction in crude protein digestibility of the nontoxic physic nut cake (p<0.05). According to Guimarães et al. (2008), high fiber increases the digesta flow rate, reducing the nutrient utilization by the short time of contact with endogenous digestive enzymes.
The apparent digestibility coefficient of ether extract was superior (p<0.05) for the nontoxic physic nut cake (90.48%), in comparison to the detoxified physic nut cake (52.10%) (Table 1). Pezzato et al. (2002) and Guo et al. (2011), while evaluating soybean meal, observed similar values of 82.67 and 87.23%, respectively, for apparent digestibility of ether extract of the nontoxic physic nut cake for Nile tilapia. However, the results obtained in the present study were lower than those recorded by Zhou & Yue (2012), of 92.60%, for the apparent digestibility coefficient of ether extract. The lower results for the detoxified physic nut cake can be interfered by residual phorbol ester, which is lipophilic and found principally in physic nut oil and kernel (Makkar et al., 1998b), reducing the apparent digestibility of ether extract. Moreover, in the present study, a large difference was observed between nontoxic and detoxified physic nut cakes regarding ether extract, due to the adopted solvent extraction (solvent plus extrusion).
Other co-products (cakes and meals) derived from oilseeds with potential for biofuel production showed similar results to those found for rapeseed meal, of 88.19% (Furuya et al., 2001), andcottonseed meal, of 81.80% (Guimarães et al., 2008). However, higher apparent digestibility values of 97.2 and 98.37%, respectively, were observed for cottonseed and rapeseed meals (Zhou & Yue, 2012). (1Mean values of the nutritional composition of nontoxic and detoxified physic nut cakes correspond to the average of three replicates. (2) Means of apparent digestibility coefficients followed by different letters in the same lines differ by Student's test, at 5% probability.
In a study with Nile tilapia, Santos et al. ( 2010) also observed similar efficiency for the apparent digestibility of gross energy (75.26%) of radish forage meal. In addition, Braga et al. (2010) reported 75.87% apparent digestibility of gross energy of palm kernel meal for Nile tilapia, whereas Zhou & Yue (2012) found 73.70, 78.9, and for rapeseed, peanut, and cottonseed meals, respectively, for hybrid tilapia.
In general, the different responses of the apparent digestibility coefficient of detoxified and nontoxic physic nut cakes can be related to anti-nutritional factors. There is also evidence of interaction between phytates and lipids, known as lipophytins, which are complexes of Ca/Mg-phytate with lipids and peptides (Leeson, 1993), hindering the absorption and possibly contributing to the lower digestibility of ether extract, gross energy, and crude protein.
The amino acids of detoxified and nontoxic physic nut cakes showed apparent digestibility coefficient up to 80%, except for threonine (79.66%), for the detoxified physic nut cake, and for glycine (79.80%), for the nontoxic physic nut cake (Table 2). Compared to other amino acids, methionine presented the highest apparent digestibility coefficients, of 95.08 and 92.97%, respectively, for detoxified and nontoxic physic nut cakes. The lowest values for the apparent digestibility coefficient of amino acids were obtained for threonine (79.66%), for the detoxified physic nut cake, and for glycine (79.80%), for the nontoxic one.
In comparison to the apparent digestibility coefficients of soybean meal obtained by Guimarães et al. (2008) and Guimarães et al. (2012), the apparent digestibility of physic nut amino acids was superior only for methionine (95.08%) and glycine (84.75%), for the detoxified physic nut cake, and for glycine (79.80%), for the nontoxic physic nut cake. The high digestibility of methionine in the physic nut cake is relevant because this amino acid limits weight gain, is incorporated into proteins, and is also used for the synthesis of other essential compounds (Nutrient…, 1983).
The digestibility of the amino acids of some oilseeds used for biofuel production was evaluated by Zhou & Yue (2012) for hybrid tilapia. The authors reported apparent digestibility coefficients of peanut, rapeseed, and cotton meals of 74.10, 63.90, and 73.30% for methionine; 89, 84.50, and 79.40% for threonine; and 82.60, 85.10, 80.50% for glycine, respectively.
Furthermore, the imbalance of the amino acid profile and the high level of physic nut anti-nutrients contributed to the lowest apparent digestibility. The detoxification process was efficient to inactivate some anti-nutrients, because, in general, the apparent digestibility of amino acids showed the highest values for the detoxified physic nut cake, except for tryptophan and threonine. The apparent digestibility coefficients of nutrients, amino acids, and energy of nontoxic and detoxified physic nut cakes is equivalent to those reported for most oilseeds and shows potential for inclusion in Nile tilapia diets.
Table 1 .
Total values and apparent digestibility coefficients (ADC) of dry matter, crude protein, ether extract, and energy of nontoxic and detoxified physic nut (Jatropha curcas) cakes for Nile tilapia (Oreochromis niloticus).
(1) Means of amino acid values and ADC correspond to a polled sample.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-04-01T13:12:53.844Z", "created": "2015-10-09T00:00:00.000", "id": "89425972", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.scielo.br/pdf/pab/v50n9/1678-3921-pab-50-09-00849.pdf", "pdf_hash": "73909e53091ad966957ce2d38c27a2d951fe0cc8", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:34242", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "73909e53091ad966957ce2d38c27a2d951fe0cc8", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Analysis of differentially expressed genes in soybean leaf tissue of tolerant and susceptible cultivars under flooding stress revealed by RNA sequencing
Flooding stress causes severe yield reduction in soybean worldwide. The development of stress-tolerant cultivars could be an effective measure to reduce the negative effects of flooding stress. Molecular information on the gene expression pattern of tolerant and susceptible genotypes under flooding stress could be valuable to improve the flooding tolerance in soybean. The objective of this study was to analyze the differentially expressed genes (DEGs) revealed by RNA sequencing in the soybean leaf tissues of tolerant (‘Paldalkong’ and ‘Danbaekkong’) and susceptible (‘NTS1116’) cultivars under flooding stress. Seedlings were grown in a well-watered condition up to the V1–V2 stage and flood-stressed by inundating ~ 10-cm water for 14 days. A total of 22,468 genes were differentially expressed in flood-stressed condition compared to the well-watered control condition, out of which 13,729, 13,405, and 13,160 were differentially expressed in ‘Paldalkong’, ‘Danbaekkong’, and ‘NTS1116’, respectively. A higher number of some of the flooding tolerance-related genes such as lipoxygenase, expansin, glutathione S-transferase, and sugar efflux transporter were up-regulated in the tolerant cultivars than in the susceptible cultivar. The number of some abscisic acid-related transcription factors of basic leucine zipper domain and myeloblastosis families was also higher in the tolerant cultivars than in the susceptible cultivar. The molecular information about the DEGs of tolerant and susceptible cultivars obtained in the present study could be valuable to improve the flooding tolerance in soybeans.
Introduction
Considering the multiple uses of soybean as sources of food, feed, biodiesel, and other industrial products, extensive efforts have been made to increase its yield worldwide. The rise in commercial value of and demand for soybean in national and international markets has increased the soybean cultivation area. Although soybean is an upland crop, the cultivation of soybean in converted paddy fields has been increased for a few years owing to comparative economic benefit, nutrient management, and government policy (Nishida et al. 2013;Singh 2010). Soybean crop is generally sensitive to flooding stress which hampers crop growth and causes significant yield reduction (Ahmed et al. 2013;Komatsu et al. 2009). Flooding stress results from heavy rainfall and/or excessive irrigation coupled with poor drainage. This situation is further worsened under paddy soils which are specially developed to hold water for longer durations.
Plants, under flooding stress, make efforts to overcome the stress by adopting various mechanisms, including flooding avoidance, flooding tolerance, and flooding adaptation (Bailey-Serres et al. 2012;Mustroph 2018;Yamauchi et al. 2018). Various adaptive mechanisms, as well as physiological, biochemical, and molecular responses under drought and flooding stresses, help to survive and retain normal growth in plants ( Dat et al. 2004). Plants perceive environmental signals that are subsequently transmitted to molecular signaling as the initial step in a stress response. Among various signal compounds in several pathways of stress responses, abscisic acid (ABA) dependent and independent responses include transcription factors (TFs). Regular transcription and its regulation rely on specific protein factors, TFs, which bind to certain DNA sequences in gene regulatory regions and control their transcription (Latchman 1993). A number of TFs of diverse families account for approximately 10% of the genes in soybean that are participated in many biotic and abiotic stress responses ( [URL] tfdb.cbi.pku.edu.cn/ index .php?sp=Gma).
A number of efforts have been made to unravel the complex molecular mechanisms underlying various biotic and abiotic stresses in soybean. Molecular techniques such as microarray profiling have been conducted to produce gene expression data of soybeans and stored in a public database ( https ://www.ncbi.nlm.nih.gov/geo/). The microarray platforms have several drawbacks, such as limited sensitivity, cross-and non-specific hybridization, and limited usefulness under scarce gene information, especially for soybean in which gene models have not been well characterized. RNA sequencing (RNA-Seq), a sequencing-based recently advanced techniques, can overcome the limitations of microarray techniques (Trapnell et al. 2013). Several RNA-Seq studies have been conducted to investigate the gene expression in numerous tissues of soybean under biotic and abiotic stresses, such as drought (Vidal et al. 2012), drought and flooding (Chen et al. 2016), common cutworm attack (Du et al. 2019), salt (Zeng et al. 2019), and seed-flooding (Sharmin et al. 2020). The objective of this study was to analyze the differentially expressed genes (DEGs) revealed by RNA-Seq in soybean leaf tissue of tolerant and susceptible cultivars under flooding stress.
Plant material and growing condition
Two flood-tolerant ('Paldalkong' and 'Danbaekkong') and one flood-susceptible ('NTS1116') soybean cultivars were selected on the basis of a previous study (Koo et al. 2014).
Plants were grown in round-bottomed pots of 20 × 30 cm dimension (top and bottom diameters × height), kept in the greenhouse of Department of Southern Area Crop Science, National Institute of Crop Science, Miryang, Republic of Korea in April of 2019. The pots were filled with the soils prepared by mixing upland soil, compost, and nursery soil at 1:1:0.75 ratio. Five seeds were sown in three replicates for each cultivar and treatment condition. Seedlings were thinned by the first trifoliate (V1) stage to keep three plants in each pot. The plants were grown in a well-watered condition up to the V1-V2 stage and flood-stressed by inundating ~ 10-cm water for 14 days. The control-designated pots were grown in the well-watered condition during the period. Leaf samples were collected at 14 days after flooding (DAF) in 1.5-mL tubes, immediately kept into liquid nitrogen, and stored at -80 °C until RNA extraction.
Measurement of flooding tolerance
Leaf chlorophyll content (CC) was measured on the second trifoliate leaves of the control and flooded plants at 2-3-day intervals using a chlorophyll meter (SPAD-502Plus, Minolta Camera Co., Osaka, Japan). Leaf chlorophyll index (CCI) was calculated as the ratio of mean CC values obtained under flooded to control conditions (Dhungana et al. 2019). Three replications were maintained for each cultivar and treatment condition and the mean values of three replicates were reported.
RNA extraction and library preparation
Total RNA was extracted using an RNA extraction kit (RNeasy Plant Mini Kit, Qiagen, Hilden, Germany) following the manufacturer's instructions. The concentration of RNA was measured with a NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA). Pooled samples of three biological replicates were prepared for individual cultivar and control/flooding treatment and thus six RNA samples were applied for RNA-Seq and DEGs analysis. The possible contaminated DNAs in the RNA samples were removed using DNase. Total RNA was further purified using a ribo-zero rRNA removal kit. The cDNA libraries were constructed using a TruSeq RNA Sample Preparation v2 Guide, Part # 15026495 Rev. F.
Quality filtering and mapping of RNA-Seq reads
The paired-end RNA-Seq reads were generated on the Illumina Genome Analyzer platform and the quality of the trimmed reads was evaluated with FastQC (Bolger et al. 2014). The RNA-Seq reads were aligned to the Glycine max reference genome (Gmax2.1version) using Bowtie2 aligner.
The read mapping was performed using the HISAT2 program (Kim et al. 2015).
Sequence assembly and differential counting
After read mapping, transcript assembly was performed through the StringTie program (Pertea et al. 2015). The estimated gene abundance of the samples was determined in terms of the fragment per kilobase of transcript per million mapped reads (FPKM). The DEGs between the floodstressed and control samples of three cultivars were identified using the StringTie-e option. The significantly up-and down-regulated genes of each cultivar were obtained for the flooding stress. The genes with a log 2 fold change ≥ + 1.5 and ≤ − 1.5, and a false discovery rate adjusted P ≤ 0.05 were identified as DEGs.
Functional annotation and gene ontology enrichment
The DEGs were annotated for gene ontology (GO) terms and grouped into three categories: molecular function (MF), cellular component (CC), and biological process (BP). The GO enrichment analysis was done using gProfileR (Raudvere et al. 2019). The DEGs were also mapped to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database of soybean to find the genes associated with different KEGG pathways. The TFs and transcription-related genes were identified using the plant transcription factor database ( [URL] tfdb.cbi.pku.edu.cn/index .php?sp=Gma) to analyze their potential roles in flooding tolerance.
Screening of DEGs using the QTL results for flooding tolerance
To identify the DEGs for their functions related to flooding stress response, the physical location of some DEGs in the genome was compared with the former mapping results of quantitative trait loci (QTL) associated with flooding tolerance in soybean. The recombinant inbred line (RIL) populations for QTL were developed using the tolerant ('Paldalkong' and 'Danbaekkong') and susceptible ('NTS1116') cultivars as parents which were also used for RNA-Seq in this study. The QTL result of one of the RIL populations was published (Dhungana et al. 2020). QTL map-based identification of DEGs may allow us to find the putative genes related to flooding tolerance and to verify the results of RNA-Seq study.
Mapping and analysis of DEGs under flooding stress
The average number of reads was 65.3, 67.5, 69.5, 60.1, 58.5, and 69.5 million for 'Paldalkong' control, 'Paldalkong' flooding, 'Danbaekkong' control, 'Danbaekkong' flooding, 'NTS1116' control, and 'NTS1116' flooding, respectively. Over 98.8% of the raw reads from each library retained after quality control and approximately 97% of reads in each library were mapped (Table 2). A summary of the workflow for RNA-Seq analysis is shown in Supplementary Fig. S1.
In total, 22,468 genes were differentially expressed under flooding stress as compared to the control condition (Supplementary Table S1). The core set of DEGs was analyzed in three combinations, namely 'Paldalkong' flooding vs. 'Paldalkong' control, 'Danbaekkong' flooding vs. 'Danbaekkong' control, and 'NTS1116' flooding vs. 'NTS1116' control. The DEGs specific to and common between the three combinations were also determined ( Fig. 1; Supplementary Table S2).
Some of the stress-related DEGs common to both tolerant cultivars 'Paldalkong' and 'Danbaekkong' such as lipoxygenase (LOC100785480) were significantly upregulated only in the tolerant cultivars but not in the susceptible cultivar, 'NTS1116'. Two phosphofructokinases (LOC100795344 and LOC100798766) were up-regulated and other two (LOC100782166 and LOC100820495) were down-regulated in the tolerant cultivars but were not significantly regulated in the susceptible cultivar. Four glutathione S-transferase (GST) (LOC100306119, LOC100796296, LOC106796199, and LOC100793025) were up-regulated and one LOC100775492 was down-regulated in the tolerant cultivars; however, none of these five genes were significantly regulated in the susceptible cultivar. Similarly, two sugar efflux transporter (SWEET) genes (GMSWEET30 and GMSWEET40) were up-regulated in the tolerant cultivars but not in the susceptible cultivar. The analysis of stress-related DEGs common to tolerant and susceptible cultivars showed mixed results, that is some of the DEGs were significantly up-regulated and others were down-regulated in the tolerant as well as in the susceptible cultivars. Lipoxygenase LOC100811820 was up-regulated in 'Paldalkong' and down-regulated in 'NTS1116'; LOC100793614 was down-regulated in 'Danbaekkong' but up-regulated in 'NTS1116'; whereas, LOC100802887 and LOC100800451 were down-regulated in 'Paldalkong' and 'NTS1116'. Expansin-coding gene LOC100799370 was up-regulated in 'Paldalkong' but down-regulated in 'NTS1116'; LOC100807807 was up-regulated in 'Danbaekkong' but down-regulated in 'Paldalkong' and 'NTS1116'; whereas LOC100807183, EXP2, and EXPB8 were downregulated in 'Paldalkong' and 'NTS1116'. Phosphofructokinase LOC100815278 was up-regulated in 'Paldalkong' but down-regulated in 'NTS1116'; LOC100818755 and LOC100803139 were down-regulated in 'Paldalkong' but up-regulated in 'NTS1116'; whereas, LOC100780164 was up-regulated in 'Danbaekkong' and 'NTS1116'. Three GST were up-regulated in 'Paldalkong', out of which only one GSTU58 was up-regulated and the other two LOC100805827 and LOC548030 were down-regulated in 'NTS1116'; whereas, EF1BGAMMA3 and GSTZ3 were down-regulated in the tolerant as well as susceptible cultivars. Three SWEET genes GMSWEET21, GMSWEET29, and GMSWEET42 were down-regulated in 'Danbaekkong' and 'NTS1116'.
Functional annotation and gene ontology (GO) enrichment
The GO analysis was performed to observe the gene expression pattern based on three categories: BP, MF, and CC. The DEGs were functionally annotated to investigate their GO ( Supplementary Fig. S2) and were also mapped to KEGG pathways to analyze their functional enrichment. The GO functional annotations indicated that the DEGs were involved in a series of biological processes, such as photosynthesis, light reaction, biosynthetic process, metabolic process, and cell wall biogenesis. The DEGs were also related to the following molecular functions: binding, transporter, and signal receptor. In addition, the DEGs were components of photosystem, membranes, cells, and organelles. The KEGG pathway analysis revealed that some of the major pathways were biosynthesis of secondary metabolites, carbon metabolism, biosynthesis of amino acids, glycolysis/ gluconeoprotein, starch and sucrose metabolism, and oxidative phosphorylation (Supplementary Fig. S3). A total of 462, 464, and 437 differentially expressed TFs were identified in 'Paldalkong', 'Danbaekkong', and 'NTS1116', respectively. The RNA-Seq analysis showed that 308 and 154; 244 and 220; and 286 and 151 TFs were up-and down-regulated in 'Paldalkong', 'Danbaekkong', and 'NTS1116', respectively (Fig. 3). The number of the up-regulated TFs was higher than that of the down-regulated TFs in all cultivars. The bHLH, bZIP, ERF, MYB, and WRKY families, some of the TF families related to stress tolerance, represented the majority of the differentially expressed TFs in all cultivars. The number of up-regulated TFs in bHLH, ERF, and WRKY was higher in the susceptible cultivar; whereas that of bZIP and MYB was higher in the tolerant cultivars. Another stress-related TF family G2-like also consisted of a higher number of up-regulated TF in the tolerant cultivars than in the susceptible cultivar.
Differentially expressed genes in the QTL region
Several abiotic stress-related DEGs including GST (GLYMA_11G216400), xyloglucan endotransglucosylase/hydrolase (LOC100799976), polygalacturonase (GLYMA_07G066900), calmodulin-binding p rot e i n ( G LY M A _ 0 7 G 0 0 8 4 0 0 ) , c yst e i n e p rotease (GLYMA_12G039400), soybean NAC gene (GLYMA_07G050600), DEAD-box RNA helicase gene (GLYMA_07G056600), and dehydration-responsive element-binding (GLYMA_07G017300) reside in the QTL associated with flooding tolerance in soybean. Moreover, all of these genes (except GST) were found in the QTL hotspots which were the regions containing relatively more stable QTL with higher phenotypic variation. Some of these DEGs, which were found in the QTL and revealed by the RNA-Seq analysis, were up-regulated in the tolerant and down-regulated in the susceptible cultivar (Supplementary Table S1).
Discussion
The results indicated that the flooding tolerance level of susceptible cultivar reduces as the flooding period extends. Several genes are found to be differentially expressed in various abiotic stresses including flooding, drought, salinity, and low temperature (Deshmukh et al. 2014;Patil et al. 2016). Although several physiological and molecular mechanisms are involved in specific stress responses, the majority of genes and pathways are common across various stress conditions (Deshmukh et al. 2014). The RNA-Seq expression analysis in this study revealed the DEGs in flood-tolerant and flood-susceptible cultivars under flooding stress conditions. The results of the up-and down-regulated genes in the tolerant and susceptible cultivars could provide useful information for the comparative transcriptional analysis of cultivars with differential stress tolerance. Such transcriptional analyses are useful in collecting the information about candidate genes associated with flooding tolerance that would help develop flooding tolerant cultivars.
Plants undergo various adaptive mechanisms, including molecular responses by regulating several genes, to survive and retain normal growth under stress conditions. In this study, several genes were also found to be differentially expressed in tolerant and susceptible cultivars under flooding stress. The analysis of DEGs was basically focused on the genes that were related to stress tolerance mechanism. Relative expressions of genes such as lipoxygenase, expansin, phosphofructokinase, GST, and SWEET between the tolerant and susceptible cultivars were analyzed. Although some of the stress-related genes were differently expressed across cultivars, the overall analysis showed that a relatively higher number of genes were up-regulated in the tolerant than in the susceptible cultivars. Lipoxygenase was reported to play roles in the adaption of radish to biotic and abiotic stresses . Results with higher numbers of up-regulated expansin-coding genes were also found in a seed-flooding tolerant soybean (Sharmin et al. 2020). The roles of expansin on germination, root length, and the number of lateral roots under abiotic stresses have been reported (Lü et al. 2013;Marowa et al. 2016). Increased expression of phosphofructokinase, a glycolytic enzyme involved in the glycolytic pathway, is associated with adaptation to flooding and drought stresses in soybean (Oh and Komatsu 2015). The GST is believed to play a protective role against flooding and drought stresses in soybean (Chen et al. 2016;Oh and Komatsu 2015). The GSTs scavenge the reactive oxygen species, which may be produced in high concentration due to stresses (Klok et al. 2002) leading to cell death (Clement et al. 2008;Wrzaczek et al. 2011), and protect plants from oxidative damage. Up-regulations of several SWEET genes were also observed in tolerant soybean lines under drought and flooding stresses . It is implied that the regulation of SWEET genes affects sugar balancing under flooding stress condition . The role of SWEET gene on plant development and stress tolerance has also been reported in Arabidopsis (Klemens et al. 2013).
As in a previous study (Sharmin et al. 2020), some cell wall-related genes such as GLYMA_08G249800 and GLYMA_18G272100 associated with COBRA-like protein were significantly up-and down-regulated in tolerant and susceptible cultivars, respectively. However, a higher number of the genes related to proline-rich cell wall protein, which were up-regulated in tolerant genotype (Sharmin et al. 2020), were down-regulated in the present study. Sharmin et al. (2020) reported that these cell wall-related genes had an important function in retaining the cell wall flexibility under flooding stress condition.
Abscisic acid (ABA) and other upstream signals may regulate the downstream pathways in abiotic stress responses (Tuteja 2007). A few number of the ABA-related TFs such as AP2, bZIP, and MYB were up-regulated in 'Paldalkong' and/or 'Danbaekkong' than in 'NTS1116'. Soybean AP2 TFs are believed to play a vital role in enhancing abiotic stress tolerance in Arabidopsis and soybean (Zhao et al. 2019). The TF families of bZIP and MYB had been also reported to have an important role in abiotic stress response in soybean, tobacco, and Arabidopsis (Cai et al. 2015;Shukla et al. 2015;Xu et al. 2016). The bZIP TFs may adjust regulation of a few stress-related genes that result into accumulation of proline (Hoang et al. 2017). Proline is supposed to contribute in stabilizing sub-cellular structures, scavenging free radicals, and exhibiting signals for cross tolerance to many stresses (Kaur and Asthir 2015). The numbers of up-regulated G2-like TF were higher in 'Paldalkong' (16) and 'Danbaekkong' (12) than in 'NTS1116' (7). Tahmasebi et al. (2019) implied that G2-like TFs contributed to enhance abiotic tolerance in tobacco since they participated in the formation and development of chloroplast (Liu et al. 2016).
The results of some stress-related DEGs by RNA-Seq in this study were comparatively analyzed with putative genes in the QTL associated with flooding tolerance (Dhungana et al. 2020). Several genes located in the QTL regions were also found by RNA-Seq to contribute to flooding and other abiotic tolerance. For example, the cell wall remodeling enzymes such as xyloglucan endotransglucosylase/hydrolase maintains cell wall stiffness and helps overcome stress (Tenhaken 2015). Polygalacturonase has been found responsive to flooding stress in soybean (Nanjo et al. 2013). The contribution of calmodulin-binding protein in biotic and abiotic stresses tolerance has also been reported . Cysteine protease shows a positive impact on abiotic stress tolerance in soybean (Mangena 2020). Overexpression of the soybean NAC gene enhanced root formation and abiotic stress tolerance in transgenic Arabidopsis (Yang et al. 2019). Tomato plant with the overexpressed DEAD-box RNA helicase gene showed drought and salt tolerance . The dehydration-responsive element-binding gene improved heat and drought stress tolerance in Arabidopsis (Mizoi et al. 2013). These reports further support the contribution of DEGs to flooding tolerance and increase the usefulness of their genetic information for soybean breeding programs.
Conclusion
Flood-tolerant and -susceptible soybean cultivars were grown under flooding stress and control conditions to investigate the gene expression patterns in leaf tissues using RNA-Seq. The flooding tolerance level, calculated as the leaf chlorophyll index, of susceptible cultivar reduced as flooding period prolonged. The number of DEGs in the tolerant cultivars was higher than that in the susceptible cultivar. A higher number of various genes and TFs related to stresses, including flooding, were up-regulated in the tolerant cultivars than in the susceptible cultivar. The results of QTL associated with flooding tolerance that were obtained with the same genetic background of tolerant parents, 'Paldalkong' and 'Danbaekkong' could support the findings of putative genes by RNA-Seq in this study. The variation in expression of some genes in the tolerant and susceptible cultivars could provide useful information about candidate genes associated with flooding tolerance that may be valuable to develop flooding tolerant soybean cultivars.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-07-26T13:36:32.140Z", "created": "2020-07-26T00:00:00.000", "id": "220746500", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://link.springer.com/content/pdf/10.1007/s12892-020-00061-6.pdf", "pdf_hash": "d3632dcb397e05090d800aba15e78198364ce769", "pdf_src": "Springer", "provenance": "peS2o-0075.json.gz:30667", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "d3632dcb397e05090d800aba15e78198364ce769", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effect of Defoliation on the Defense Reactions of Silver Birch ( Betula pendula ) Infected with Phytophthora plurivora
: In natural environments, plants develop adaptive mechanisms at the cellular and molecular levels to cope with many external factors, e.g., insects and soil pathogens. We studied physiological stress induced by different levels of foliage removal (defoliation 30% and 60%) and by infection of root rot pathogen Phytophthora plurivora on the common Polish tree species, Betula pendula , grown in an open greenhouse. This study showed that P. plurivora damaged the root system which negatively impacted all morphological parameters. However the loss of 30% of the leaves had a positive effect on defense responses. Chlorophyll-a fluorescence parameters indicated a decrease in photosynthetic efficiency in defoliated plants, but plants inoculated with the pathogen had a higher performance index showing increased vigor of the infected plant than birches. The study demonstrated the intense immunity response of birch to P. plurivora through the expression of Hsp 90 and Hsp 83 genes. The trees weakened by P. plurivora became more susceptible to infection by Erysiphe ornata var. ornata .
Introduction
Silver birch (Betula pendula Roth.) is a common deciduous tree species in the temperate and boreal climates of the Northern Hemisphere [1]. The varied ecotype conditions impact Silver birch trees growing in forests by exposing them to different biotic and abiotic stresses that negatively affect tree growth and development [2].
Infections caused by pathogenic fungi are significant biotic factors leading to the phenomenon of birch dieback [3]. Phytophthora species are important plant pathogens that infect many forest, ornamental, and crop plants [4]. Phytophthora pathogens cause over 66% of all fine root diseases and 90% of all collar rot [5]. One of the most common species of pathogens in the genus Phytophthora is Phytophthora plurivora (T. Jung & T. I. Burgess) [6].
Its zoospores are easily dispersed along aquatic systems in agriculture and forestry and transmitted by birds during their migration [7]. The first reports of P. plurivora infecting birch in the wild came from Finland [8]. In Poland, P. plurivora has caused significant damage to oak stands in the east and south of the country, mainly at Krotoszyn Plateau [9]. In addition, there are reports on the negative impact of P. plurivora on the loss of infected plant roots [10,11]. At the same time, there is no information on defense responses that occur at the level of gene expression, photosynthetic activity, and susceptibility to other pathogens.
There is also insufficient data on the effect of defoliation on the above mechanisms. Insect feeding weakens trees and may increase the possibility of pathogen invasion. Various primary insects often attack birch trees, the most common being Eriocrania spp, Lymantria dispar (L.), Deporaus betulae (L.), Phyllobius betulae F., Ph. arborator, and Scolioneura betuleti Zdd [12]. Defoliation of birch crowns by Lymantria dispar (L.) can cause a reduction in immunity and is the most common cause of infection with the pathogen Erwinia multivora (Scz.-Parf.) [13].
Plants routinely face various biotic (e.g., nematodes, herbivores, and pathogens) stresses, sometimes simultaneously. Plant responses to the cumulative impact of several different biotic stressors are often synergistic and cannot be extrapolated directly from combining the responses to each individual stressor, so it is important to characterize plant acclimation responses to a combination of biotic stressors [14]. Examining the cumulative impact of stressors and their effect on survival and productivity is important because the plant has a network of interconnected cellular stress response systems [15]. A thorough understanding of this network is important for developing new methods to improve plant stress tolerance. Overall, pathogen infection and defoliation cause a number of changes in morphology, physiology, biochemical and molecular processes. Heat shock proteins (Hsps) play an extensive role in many cellular processes, giving them a general role in tolerance to different environmental stress treatments. Maintaining proteins in their functional conformation and preventing aggregation of non-native proteins are particularly important for cell survival under stress [16][17][18]. Under stress, these stress-responsive biomolecules act as molecular chaperones through up-or down-regulation [19,20]. Stress has a significant impact on the expression of Hsp [20][21][22]. Increased expression of the Hsp genes has been observed in birch (Betula platyphylla (Sukaczev) exposed to drought stress [23]. To assess the molecular response to plant damage caused by phytopathogens, we examined the expression of two selected Hsp genes, Hsp90 and Hsp83.
To evaluate the level of damage caused by defoliation and infection, we analyzed сhlorophyll fluorescence to detect changes in the photosynthetic apparatus. The chlorophyll fluorescence technique is a rapid and economical method for measurements of photosynthetic activity of plants without disturbing or destroying the structure of the studied object [24].
To examine the presence or absence of synergies or cumulative effects of multiple biotic stressors, we tested the impact of one soil-borne pathogen, P. plurivora, on seedlings of silver birch (B. pendula) subjected to stress caused by mechanical defoliation, simulating primary insect feeding. We compared expectations of synergy in the impact on seedling growth and physiological performance (H 1 ) against the absence of any multiplicative, countervailing, or otherwise nonlinear influences, i.e., the total effects were a linear summation of the individual, single-factor effects (H 0 ).
General Experiment Design
Two-year-old seedlings of silver birch (Betula pendula Roth.) served as study materials. The seedlings were planted in pots and grown under semi-natural conditions, in an open greenhouse in the Forest Research Institute in Sękocin Stary (Poland) (52 • 06 03.4 N 20 • 52 56.5 E). In spring 2018, 60 trees were infected with P. plurivora NCBI KX618501; the inoculum was identified according to the methodology of Jung et al. [25]. We manually removed leaves to mimic 30% and 60% insect infestation. In the second growing season (2019), to stimulate sporangia, we flooded the pots with seedlings with water for 72 h and repeated the defoliation. The experiment has six treatments with 20 replicates each: Control-uninfected and non-defoliated trees P. plurivora-plants infected with pathogen P. plurivora without defoliation; P. plurivora + defoliation 30%-pathogen P. plurivora infected plants slightly stressed caused by removing 30% of the leaves; P. plurivora + defoliation 60%-pathogen P. plurivora infected plants with severe stress caused by removing 60% of the leaves; Defoliation 30%-uninfected plants and slightly stressed caused by removing 30% of the leaves, and Defoliation 60%-uninfected plants and severely stressed caused by removing 60% of the leaves.
Verification of the Infection of the Seedlings
In October 2019, soil samples were collected from each variant, and qPCR assays were performed to verify the presence of the pathogen P. plurivora. We extracted DNA from rhizospheric soil and root tissues using NucleoSpin Soil Mini Kit (Macherey-Nagel, Duren, Germany) according to the manufacturer's instructions. DNA was eluted in 50 µL H 2 O for each sample, and the extracted DNA was stored at −20 • C.
Biometric Parameters
At the end of the experiment, in October 2019, all plants were removed from the pots. We measured the stem length and the diameter at the root collar. The stems were separated from the roots and collected in labeled envelopes. The stems were dried in the Termaks Series 2000 cabinet at a temperature of 60 • C for 72 h.
Measurement of the Root Systems
To thoroughly clean the roots, they were first floated (about 60 min) in large containers (50 L) filled with 3/4 of their volume with tap water. When the adhered soil to the fine roots was softened and dissolved, they were washed under a light stream of water. The clean roots were placed in EPSON Perfection V700 scanner Photo Scanner, and the image files generated were processed using WinRHIZO 32-bit software (Regent Instruments Inc., Ltd., Quebec, QC, Canada). It was necessary to spread the roots properly in the scanner to avoid overlapping and crossing. We replaced the water in the scanner's water pool after each measurement to avoid contamination of the subsequent samples.
We measured the following parameters: TRL-total root length; MRL-length of the mother roots (2-5 mm); FRL-length of fine roots (0-2 mm), and FRSA-area of fine roots.
Chlorophyll Fluorescence
Photosynthetic efficiency was assessed by measuring chlorophyll fluorescence parameters, which can provide detailed information about the state of the photosystem II (PSII) [28]. Measurements were performed using a Handy PEA (Handy Plant Efficiency Analyzer) fluorimeter, following instructions from Hansatech Instruments Ltd. (King's Lynn, Norfolk, UK). Measurements were performed twice: immediately after defoliation (1 May 2019) and five months later (27 September 2019). Five randomly selected leaves from each tree and in each treatment were used for chlorophyll fluorescence measurements. The first step was to adapt the leaves to darkness with light leaf clamps for 30 min. After dark adaptation, a single strong 1 s light pulse (3500 µmol/m 2 /s) was applied to the leaf using three light-emitting diodes (650 nm). All elements were normalized to the control treatment. We examined several photosynthetic parameters: the total performance index of photosystem II (PI total), the time at which the maximum fluorescence value was reached (Tfm), the initial fluorescence representing the emission by excited chlorophyll molecules in the antenna structure of the photosystem II (Fo), the maximum fluorescence value (Fm), variable fluorescence (Fv), the maximum photochemical quantum yield of PSII (Fv/Fm), the area across the fluorescence curve between Fo and Fm (Area), ABS/RC-absorption flux (of antenna Chls) per RC, TR0/RC-trapped energy flux (leading to QA reduction) per RC, ET0/RC-electron transport flux (further than QA−) per RC, DI0/RC-total energy dissipated per reaction center [28].
The maximum quantum yield of PSII (Fv/Fm) shows the probability that a trapped photon will end up in the reaction center and cause a photochemical event. The value of Fv/Fm for unstressed leaves is about 0.83 [29]. Variable fluorescence Fv was calculated by subtracting the Fo value from the Fm value. The lower value of the Fv parameter, the lower the PSII performance [30].
Parameter Area is the area above the Chl fluorescence curve between Fo and Fm, and it quantifies the pool of electron transporters in the electron transport chain. This parameter displays any change in the shape of the induction kinetic between Fo and Fm, which would not be evident from the other parameters. A decrease in the parameter Area reveals blocking the electron transfer from the reaction centers to the quinone pool (for example, in herbicide action) [29].
Healthy plants channel part of the flux of absorbed photons (ABS) as trapping flux (TR) to the reaction center (RC), and other photons of this excitation energy are dissipated (DI), mainly as heat. An electron transport (ET) will be created within the reaction centers (RC) and further to PSI (RE), ultimately leading to CO 2 fixation. These specific fluxes refer to the beginning of fluorescence induction (time zero): ABS/RC, TR0/RC, DI0/RC, and ET0/RC [31]. We focused on the total performance index (PI total) parameter because it essentially is an indicator of sample vitality. It indicates an internal force of the sample to resist constraints from outside [32].
Analysis of Heat Shock Proteins Gene Expression
To study the plant's response to stress, two genes were selected-Hsp90 (KP245816.1) and Hsp83 (KP245815.1). A previous analysis of the cellular location of their encoded proteins using DeepLock software [33] found that they are in different areas in the cells. The protein Hsp90 is located in the cytoplasm, while Hsp83 in mitochondria or chloroplasts. Total RNA was extracted using Plant RNA Mini Kit (Syngen Biotech, Wrocław, Poland), following the manufacturer's protocol. The total RNA extracted and its purification from protein and polysaccharides were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). RNA integrity was checked electrophoretically in a 1.5% agarose gel stained with ethidium bromide. Only samples that met both quality and integrity requirements were used in subsequent experiments. Three highquality RNA samples (i.e., biological replicates) were obtained for each condition. Reverse transcription was performed using GoScriptTM Reverse Transcription System (Promega GmbH, Walldorf, Germany) according to the manufacturer's instructions. Primers for analysis of Hsps gene expression were designed using Primer-BLAST [34] under default parameters. α-Tubulin (FJ228477.1) was used as a reporter gene in our experiment. Realtime PCR reactions were performed in 20 µL volume: 10 µL 2× qPCR SYBR Master Mix (Sigma Aldrich, Milwaukee, WI, USA), 2 µL cDNA, 2 µL each primers (5 µM forward and reverse) (Genomed, Poland), 6 µL H 2 O. Thermocycling conditions consisted of the initial denaturation step at 95 • C for 3 min and 40 cycles at 95 • C for 10 s, annealing at 55 • C for 20 s, and elongation at 72 • C for 20 s. Real-time PCRs were performed in Mic Real-Time PCR Cycler (Bio Molecular Systems, Australia). Based on obtained Ct values, the 2∆∆Ct method was used to calculate the relative ratio of Hsps' expression, but the correct amplification efficiency was used instead of the value 2 [35]. We used a noise-resistant iterative nonlinear regression algorithm (Real-time PCR miner; www.miner.ewindup.info, accessed on 1 January 2005) to determine the efficiency of the PCR reaction [36].
Observation and Identification of Parasitic Organisms on Birch Leaves
As the plants grew, they were monitored for changes in health status like infections of leaves by unwanted parasitic microorganisms. The degree of fungal infection on leaves was assessed twice: in August (3 August 2019) and in September (15 September 2019) using a five-point powdery mildew damage scale [37] as follows: 0-healthy leaves (no symptoms), 1-little disease (25% of leaf surface), 2-medium disease (26% to 50% of leaf surface), 3-strong disease (51% to 75% of the leaf surface), 4-severe disease (more than 76% of the leaf surface).
At the same dates as the plant health observations (the appearance of fungal on the leaves), we removed samples of symptomatic birch leaves to identify the species of pathogens by Sanger sequencing using forward ITS1 and reverse ITS4 primers [27]. Samples taken on 3 August 2019 were marked M1, and those taken on 15 September 2019 marked as M2 and M3.
Genomic DNA extraction was performed using NucleoSpin Plant II Kit (Macherey-Nagel, Düren, Germany) following the manufacturer's instructions. The 20 µL PCR mixture consisted of: buffer 2 µL, primers 2 µL, DNA 2 µL, dNTPs 4 µL, polymerase 1 µL, H 2 O 9 µL. Thermocycling conditions consisted of an initial denaturation step of 95 • C for 30 min, followed by 30 cycles of 95 • C for 30 s, 55 • C for 30 s, elongation at 72 • C for 30 s, and a final extension step. After purification of PCR products by Clean-Up (A&A Biotechnology, Poland), the obtained amplicons were sequenced with an ABI 3500 × L genetic analyzer (ThermoFisher Scientific, Waltham, MA, USA) following the manufacturer's procedure. The species affiliation of the analyzed samples was determined using a BLAST algorithm.
Chemical Analysis of Birch Shoots Extracts
Birch shoots up to 5 mm in thickness were selected and into a 0.5 mm fraction to facilitate chemical analysis. Raw material (5 g) was extracted three times with 50 mL of diethyl ether and filtered through a paper filter. The solvent was removed using a rotary evaporator. Ether extract (10 mg) from birch shoots was dissolved with 1 mL of pyridine, 100 µL of BSTFA was added, and the sample was heated for 30 min at 60 • C, according to a previously developed method of silylation [38,39].
Silylated extract from birch shoots was analyzed with an Agilent 7890A gas chromatograph with an Agilent 5975C mass spectrometer. Injection of 1 µL sample was done using an autosampler. Chromatographic separation was performed on a capillary column HP-5MS (30 m, 0.25 mm, 0.25 µm) at a helium flow rate of 1 mL min −1 . The injector worked in a split (1:10) mode at a temperature of 300 • C. The initial column temperature was 50 • C, rising to 325 • C, at 3 • C min −1 , and the final temperature was held for 10 min. The detection was performed in a full scan mode from 41 to 800 amu, according to a previously developed method of GC-MS analysis [40][41][42].
For the identification of extracted compounds, both mass spectra and retention indices were used. After integration, the percentage content of each component in the total ion current (% of TIC) was calculated. All measurements for shoots extracts were performed in 3 replicates to determine the detection error.
Statistical Analysis
All responses were modeled as two-way factorial designs using Generalized Linear Mixed Models (GLMMs) that included the main effects of P. plurivora infection (yes/no) and defoliation level (0%, 30%, and 60%) and infection x defoliation interaction. A random heteroscedasticity component for each infection x defoliation combination was also included in all of the models. Each model utilized the appropriate conditional distribution of the response based on the response range. Specifically, TRL, MRL, FRL, FRSA, Root Collar, and Stem Dry Mass were modeled as Gamma with a log link, and the mildew damage scale was modeled as a cumulative logit link response or ordered logit [43].
All model prediction confidence bounds were controlled for family-wise experiment rate using the Sidak adjustment method with α = 0.05 [44]. Multiple comparisons were adjusted for the family-wise experiment rate using Tukey's HSD method with α = 0.05 [45]. Data were analyzed using the R statistical environment version 4.0.5 (R Core Team, 2021) using R libraries glmmTMB [46], MASS [47], and emmeans [48].
Detection of Phytophthora plurivora
A year after inoculation, the real-time PCR reaction confirmed the successful infection of plants by P. plurivora in the treatments: P. plurivora, P. plurivora + defoliation 30%, and P. plurivora + defoliation 60% (Ct for values ranged from 26.53 to 27.38). As expected, the Control, Defoliation 30%, and Defoliation 60% treatments yielded negative results (Ct > 40).
Biometric Parameters
The height of trees in infected and defoliated treatments was low compared to the control, but the differences were not statistically significant (p = 0.15) (Table S1 and Figure S1). The pathogen alone (p < 0.01) or in combination with both levels of defoliation (p = 0.04 and p < 0.01) negatively affected the diameters of the root collar, significantly decreasing its width compared to the control (Figure 1, Tables S2 and S3). Diameters of the root collar and dry biomass of stems. The lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentiles), and the center line corresponds to the median. The upper whisker extends from the upper hinge to the largest value no further than 1.5 * IQR from the hinge (where IQR is the inter-quartile range or distance between the first and third quartiles). Likewise, the lower whisker extends from the lower hinge to the smallest value no further than 1.5 * IQR from the hinge. Treatments with different letters in the notation are significantly different. Points were jittered in order to avoid overplotting of identical values.
The fresh biomass of non-leafy stems ranged from 12.96 to 18.1 g. After drying, the biomass decreased on average 1.25 g, compared to fresh biomass (data not shown). Stem dry biomass of severely defoliated trees (60%) without pathogen was significantly lower than the biomass of the P. plurivora infected trees with defoliation 30% (p = 0.01). However, treatments were not statistically different from the control.
Morphological Parameters of the Root System
Analysis of the 5 morphological roots parameters revealed a significantly negative impact of P. plurivora soil inoculation on the root system health (Figure 2, Figures S2-S5, Tables S4-S8). However, Defoliation 30% (without pathogen) did not cause significant changes of all monitored parameters compared to the control. The upper whisker extends from the upper hinge to the largest value no further than 1.5 * IQR from the hinge (where IQR is the inter-quartile range or distance between the first and third quartiles). Likewise, the lower whisker extends from the lower hinge to the smallest value no further than 1.5 * IQR from the hinge. Treatments with different letters in the notation are significantly different. Points were jittered in order to avoid overplotting of identical values.
Total root length (TRL) significantly decreased in treatments P. plurivora compared to the Defoliation 30% (p < 0.01) and control (p < 0.01). Graphs and tables with sample statistics for the length of the mother roots, length of fine roots, and fine root area are presented in the Supplementary Materials. The comparison of mother roots' length (MRL) between all P. plurivora inoculated plants and treatment Defoliation 30% indicated statistically significant differences (p < 0.01) ( Figure S1). The most severe effects of the pathogen were found on the fine roots. P. plurivora decreases fine root length ( Figure S2) and fine root surface areas (FRSA) by 59% ( Figure S3). Nevertheless, the combined impact of the pathogen and defoliation resulted in higher values of these growth parameters than the pathogen itself. Half of the treatment combinations-Defoliation 60% (p < 0.01), P. plurivora (p < 0.01), and P. plurivora + defoliation 60% (p < 0.01)-had significantly less root biomass than the control.
Chlorophyll-a Fluorescence Measurements
Measurement of chlorophyll fluorescence signals on birch trees after treatments of defoliation and root infection with P. plurivora pathogen were conducted in May, right after defoliation, and in September 2019. The results are displayed in Figures 3 and 4 and expressed as a percentage of the control, with higher values indicating more efficient photosynthesis. Estimates and statistical comparisons are presented in Tables S9 and S10.
The measurement results obtained in May (Figure 3) showed the negative effects of defoliation by the values of the total performance index (PI total) in comparison to the control plants.
Total performance index parameter decreased by 35% for treatment Defoliation 30% and 17% for Defoliation 60%. However, treatments of P. plurivora with 30% defoliation and without defoliation demonstrated a positive response of the photosynthetic activity. PI total increased by 18% for P. plurivora itself and by 31% for combination P. plurivora + defoliation 30%. The difference between P. plurivora + defoliation 30% and Defoliation 30% was 66% (p < 0.01). The treatments of defoliation 30% (p < 0.01) and defoliation 60% decreased their Fv/Fm by 2%, whereas P. plurivora and P. plurivora + defoliation 60% increased it by 2%. While Tfm for all treatments decreased compared to the control, Tfm for defoliation 60% increased by 17%. The rise in Fo was observed in all treatments, except for the treatment with P. plurivora alone. The Fm and Fv were higher than the control in all treatments. The value of the Area parameter increased in all P. plurivora infected trees, with the greatest effect for P. plurivora + defoliation 60%, increased by 57% in comparison to the control (p = 0.02). The specific energy fluxes in one of the active reaction centers ABS/RC, TR0/RC, ET0/RC, DI0/RC were decreased for treatments with P. plurivora. Parameter ABS/RC for P. plurivora treatment ABS/RC was 10% less than the control (p = 0.04), ET0/RC decreased by 20% for P. plurivora + defoliation 30% (p < 0.01) and for P. plurivora + defoliation 60% (p < 0.01).
The results of measurements carried out in September (Figure 4) point to an increase in the total performance index compared to May's results.
The largest change was observed in the P. plurivora + defoliation 30%, increasing 158% over the control (p < 0.01). Moreover, in May, the total PI for defoliated trees without P. plurivora was lower than in the control group, while in September, they had higher results than the control group.
September's results of the Tfm value decreased for the Defoliation 30% and Defoliation 60% (p = 0.01) groups but increased for P. plurivora, P. plurivora + defoliation 30%, and P. plurivora + defoliation 60% groups. The Fo slightly decreased in all groups. On the other hand, Fv/Fm increased compared to the results obtained five months ago. The parameters ABS/RC, TR0/RC, ET0/RC, DI0/RC for treatments with P. plurivora were still less than the values for the control group.
Heat Shock Protein Gene Expression Analysis
For the analysis of Hsp90, BpHsp90-f and BpHsp90-r were used, and for the analysis of Hsp83, BpHsp83-f and BpHsp83-r primer pairs were used. The primer pair for amplification of the tubulin gene was BpTub-f and BpTub-r. Each amplicon was found only in a single peak in melt curves indicating no dimer or multiple products.
Analysis of the expression of the Hsp90 gene in birch leaves showed that P. plurivora infection combined with 60% defoliation increased Hsp90 expression about seven times ( Figure 5), though other combinations with P. plurivora were not higher than the control. Birch root infection by P. plurivora in combination with partial loss of leaves (30% defoliation) led to a two-fold increase in the expression of the Hsp83 gene. Other treatments with P. plurivora also exhibited greater Hsp83 than the control. Stress caused by defoliation without the pathogen responded by down-regulation of both Hsp90 and Hsp83.
Interactions with Other Species of Pathogens on Birch Leaves
At the beginning of August 2019, symptoms of infestation by leaf parasites appeared on the leaves for the first time ( Figure 6). Sequencing of ITS from sample M1 revealed that this sequence has 100% identity to Erysiphe ornata var. ornata (U. Braun) (HM057441.1). Erysiphe ornata var. ornata is a species of powdery mildew. The ITS sequence from sample M2, taken in September, also has 100% identity to Erysiphe ornata var. ornata. However, the ITS sequence from the sample M3 has 98.24% identity to Ampelomyces quisqualis (Ces.) (JX681067.1). A. quisqualis is a mycoparasite and the natural antagonist of powdery mildew fungi. Although the seedlings in all variants were infected by powdery mildew, birch trees inoculated with P. plurivora showed a much greater degree of damage (Figure 7, Table S11).
Discussion
This work reveals silver birch fine root damage by P. plurivora and confirms earlier observations [8]. Necrosis of fine roots' length and surface area are typical symptoms of Phytophthora root rot [49]. Tissues infected by Phytophthora show root rot which reduces proper absorption of water and nutrients [50]. Scars formed on maternal roots after the death of fine roots raise the risk of penetration of other microorganisms causing secondary infections [51]. There was a visible reduction of the diameters of the root collar in all the treatments with P. plurivora which may be explained by lesions spreading from the roots up to the root collar. The results of the fine root length and the fine root area suggest slight mitigation of the P. plurivora effect on the roots under 30% defoliation.
Moreover, non-infected trees with defoliation 30% had the same or higher length of the root parameters than the control. One of the possibilities is that moderate stress induces plant resistance [40]. Considering the increasing aboveground biomass and the damaging effect on the underground mother root, the cumulative effect of both stressors could lead to a higher risk of tree failure. The attenuating effect of P. plurivora infection on growing stem biomass under defoliation could be explained by the observed activation of defense gene expression and increasing photosynthetic activity due to pathogen treatment.
Analysis of photosynthetic activity after 72 h immersion in water and defoliation revealed negative photosynthetic reactions to the defoliation and positive to the pathogen. The PI total parameter, which is essentially an indicator of tree vitality, was the lowest in defoliated trees without pathogens, but treatments with P. plurivora had higher values, which suggests some sort of compensation mechanism [31,52]. The presence of the pathogen significantly increases the photosynthetic activity of the plant compared to the control. Considering that P. plurivora consumes sugars produced by the tree in its parasitic activity, this is a positive response of the tree to the pathogen. On the other hand, defoliation reduces sugar production, which could mitigate the effects of the pathogen.
The high Tfm values in Defoliation 60% mean that a longer period of time is needed to reach a full reduction state of the primary electron acceptors that indicate a low tolerance to stress caused by loss of many leaves [53]. Besides, the high values of F0 of defoliated trees point to low efficiency of excitation energy transfer between chlorophyll molecules. They could be the result of the partial dissociation of chlorophyll in PSII [54]. The area above the fluorescence transient (Area) reflects the size of the plastoquinone pool. This Area variable for the plants which were infected by P. plurivora increased, indicating the presence of a larger quinone pool. Though this parameter was decreased in the case of the 60% defoliation plants, this might be involved in blocking the transport of electrons from reaction centers to the pool of plastoquinone [55]. However, September's performance index results were higher in all treatments than May's, indicating the adaptation of plants to stress over time. Additionally, plants have complex adaptive mechanisms at the cellular and molecular levels. Under stress, plants transcribe and translate heat shock proteins. Hsps' genes from distinct plant species respond differently to various types of stress [22]. While their response to heat, cold, osmotic, and salt stress is relatively well studied [56,57], information about their response to pathogen infection is rather scarce. This study revealed an intense response of silver birch immune mechanisms to P. plurivora infections and defoliation by activating the expression of the Hsp90 and Hsp83 genes. The maximum level of expression for these genes as the response to the pathogen was observed in different treatments, Hsp90 dramatically increased in the group P. plurivora + defoliation 60%, and Hsp83 in the group P. plurivora + defoliation 30%. Phytopathogenic fungi of the genus Phytophthora have an impact on the expression of other Hsp genes. Phytophthora infestans ((Mont.) de Bary) led to up-regulation of Hsp70 genes and increased synthesis of Hsp70 proteins in tomatoes [58] and Solanum tuberosum (L.) [59], which indicates that it could participate in mediating the disease resistance in plants in response to biotic stresses. Lee et al. [60] investigated the mode of action of a P. infestans effector, Pi23226, which induces cell death in Nicotiana benthamiana (Domin), and found that the effector interacts with two isoforms of heat shock protein 70. Infection of avocado (Persea americana (Mill.)) by Phytophthora cinnamomi (Rands) downregulated 17.3-kDa small heat-shock protein [61]. Hsp90 also participates in plant disease resistance; together with the co-chaperones RAR1 and SGT1, Hsp90 modulates many cytosolic R proteins, such as MLA, RPM1, RPS2, and RPS4. Hsp90 activates cytosolic R protein, which contains a nucleotide-binding site (NB) and leucine-rich repeats (LRRs), mediating the defense of plants against many pathogens [62]. It was shown that the RXLR cytoplasmic effector AVR3a of P. infestans confers avirulence on potato plants carrying the R3a gene. R3a activation by AVR3aKI is dependent on the ubiquitin ligase-associated protein SGT1 and heat-shock protein Hsp90 [63]. Hsp90 also interacts with necrosis-inducing Nep1-like proteins (NLPs) [64].
The synthesis of biologically active substances by plants is an effective method to reduce the attack of insects and the spread of pathogens. Biochemical compounds, such as volatile substances (belonging to secondary metabolites produced by the synthesis of terpenes and phenylpropanoid compounds), as well as growth inhibitors and some hormones, are involved in plant defense responses to stress. Plants under the stress of defoliation generally reduce the production of sugars, proteins, starch, lignin, and hemicellulose while increasing the synthesis of secondary metabolites, including phenols, terpenes, and sterols [65]. Defoliation 60% increased the emission of sterols, fatty alcohols, fatty acids, and fatty acid esters; however, the contents of these substances in plants with a lower degree of defoliation (defoliation 30%) were similar to the control group. The combination of P. plurivora infection and defoliation increased the production of triterpenes, which was a defensive stress response. In previous research [66][67][68], we found that birch tissues contain different groups of biologically active substances, e.g., flavonoids, triterpenes, and sterols. We proved that birch tissues secrete VOC, such as 2-phenylethanol or linalool [69,70]. We detected that P. plurivora emits volatiles also [71]. Moreover, in previous investigations, we studied the effect of other species from the Phytophthora genus, i.e., P. cactorum [72,73], on the composition of bioactive compounds in birch shoots.
Additionally, weakening of birch immunity caused by infection of the roots by P. plurivora predisposes them to other diseases. The seedlings grew in the forest in the open air, in conditions close to natural, which made their leaves vulnerable to infections of other pathogens. In August, we observed the parasitic fungus Erysiphe ornata var. ornata on the leaves. This leaf parasite belongs to the group of biotrophic pathogens, which use nutrients obtained from the host plant, leading to the loss of carbohydrates and other substances in the plant [74]. As the level of damage by powdery mildew was significantly higher in the group plants with P. plurivora than control and defoliated-only trees, this may mean that P. plurivora increases susceptibility to other. Moreover, a month later, hyperparasite Ampelomyces quisqualis appeared, a natural biocontrol agent for powdery mildew.
Conclusions
This study demonstrates different responses to P. plurivora depending on degrees of defoliation. Defoliation of 60% leads to damage of the assimilatory apparatus and promotes pathogenic infection, while a lower degree of defoliation (defoliation 30%) seems to counteract the pathogen's influence. The oomycete P. plurivora is a dangerous pathogen for silver birch, causing considerable damage to the root system. However, the combination of stress from P. plurivora and defoliation 30% results in a much healthier root morphology than trees infected with P. plurivora without defoliation or trees infected with P. plurivora in combination with 60% defoliation. Although defoliation significantly reduced total photosynthetic activity, the P. plurivora treatments still showed high vitality of the photosynthetic apparatus, as expressed by an increase in the total performance index (PI total).
We examined the molecular response to plant injury and found that defoliationinduced stress down-regulates, whereas P. plurivora infection caused up-regulates of Hsps' gene expression. Exploring Hsp gene expression under a combination of different stresses was also one of our targets. We discovered that Hsps responded differently to various combinations of stressors. The highest expression of Hsp83 was obtained with P. plurivora + defoliation 30% treatment, and the highest expression of Hsp90 was obtained with P. plurivora + defoliation 60%. This outcome may also result from Hsps localized in the cytoplasm (Hsp90) and chloroplast (Hsp83) responding differently to stressors.
Moreover, defoliation 60% increased the production of sterols and fatty acids, fatty acid esters by 2.5-fold compared to the control group. The combination of defoliation with inoculation of P. plurivora increased the content of triterpenes. The highest content of flavonoids was found in the treatment P. plurivora + defoliation 60%.
In addition, we found the parasitic fungus Erysiphe ornata var. ornata on the leaves of all trees. Moreover, we also detected the presence of the hyperparasite Ampelomyces quisqualis on Erysiphe ornata var. ornata. To our knowledge, this is the first record of a host-parasite relationship between Erysiphe ornata var. ornata on Betula pendula and Ampelomyces quisqualis in Poland. However, we do not yet know whether this natural biological control agent can maintain the hyperparasite population at an adequate level.
This study documented the combined outcomes at the biochemical, physiological, and biochemical levels of simultaneous defoliation and fungal infection upon birch tree seedlings, a first of its kind. We confirmed our hypothesis that combined forest pest attacks created an effect that is not simply the linear summation of the individual stressors. Some of these harmful influences had immediate negative impacts, whereas others initially counteracted these negative impacts until the combined physiological toll overwhelmed the plant's natural defenses. As more and more ecosystems face attacks from various biological organisms and abiotic influences such as warming climate and increasing drought, integrated studies such as this one which parses out the constituent influences upon plant health will become more and more valuable.
Supplementary Materials: The following are available online at [URL]/10 .3390/f12070910/s1, Table S1. Height of the seedlings of silver birch, Table S2: Root collar diameter sample statistics, Table S3: Stem dry mass sample statistics, Table S4: Total root length sample statistics, Table S5: Mother roots length sample statistics, Table S6: Fine roots length sample statistics, Table S7: Fine root area sample statistics, Table S8: Root dry mass sample statistics, Table S9: Measurement values for parameters of chlorophyll fluorescence in May, Table S10: Measurement values for parameters of chlorophyll fluorescence in September, Table S11: Sample statistic of the degree of leaf damage according to the Khvasko, Table S12: Detailed chemical composition of extracts from birch seedling. Figure S1. Height of the seedlings of silver birch, Figure S2. Root morphology parameters depending on the treatment: MRL -length of the mother roots 2-5 mm), Figure S3. Root morphology parameters depending on the treatment: FRL-length of fine roots (0-2 mm), Figure S4. Root morphology parameters depending on the treatment: FRSA-fine root area, Figure S5. Root dry mass.
Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the study's design, in the collection, analyses, or interpretation of data, in the writing of the manuscript, or in the decision to publish the results.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-09-28T01:09:06.947Z", "created": "2021-07-13T00:00:00.000", "id": "237743709", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.mdpi.com/1999-4907/12/7/910/pdf?version=1637056942", "pdf_hash": "4c0b8d87c2b801c5142caf079503205c8c482136", "pdf_src": "Adhoc", "provenance": "peS2o-0075.json.gz:32164", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "77a5167fe11bb29e5e6acbc7826003c20f6ac3a7", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
This below document has 2 sentences that start with 'Treatments included low-forage diet',
2 sentences that start with 'Means ± standard deviation.ab -values',
2 sentences that end with 'two treatment groups',
2 sentences that end with '(French et al., 2000)',
3 sentences that end with '2015;Ghassemi Nejad et al., 2016)',
2 sentences that end with 'al., 1988;Zinn et al., 1994)',
2 paragraphs that start with 'Fatty acid profile of meat',
2 paragraphs that end with 'letters differ significantly at P<0.05'. It has approximately 2803 words, 120 sentences, and 34 paragraph(s).
===
Fatty acid profile of meat , diurnal changes in volatile fatty acids , rumen fluid parameters , and growth performance in Korean native ( Hanwoo ) steers fed high-and low-forage diets supplemented with chromium-methionine
The objective of this study was to determine the effects of forage level in diets supplemented with chromium-methionine (Cr-Met) on rumen fluid parameters, meat fatty acid composition, and performance of Korean beef (Hanwoo) steers. Twenty-three Hanwoo steers were used in this experiment. A completely randomized design and repeated measurements were used to analyze the data set. Beef steers were fed diets containing high (10 head; average body weight (BW) = 525.1±27.5; forage:concentrate (F:C) = 60:40) (60F) and low (13 head; average BW = 531.8±32.4; F:C = 40:60 ratio) (40F) forage diets supplemented with Cr-Met for 60 d. Dry matter intake, BW, and feed efficiency were not different between the two treatment groups. Fatty acid composition of meat including myristate, stearate, and gamma linoleate was not different between the two groups; however, palmitate, palimtoleate, and linoleate were higher in 60F group than 40F group. Ammonia-N showed a higher trend in 40F group, whereas pH demonstrated higher values in 60F group. Ruminal acetate was higher in 60F group than 40F group and maintained a high trend throughout the sampling time, whereas no differences were found in ruminal propionate, butyrate, and valerate between two groups. A high-forage diet (60%) improves meat quality and has no adverse effects on performance of Hanwoo steers.
Introduction
Previous studies in our laboratory have reported improvements in carcass characteristics in Korean native (Hanwoo) steers (Sung et al., 2015), meat fatty acid profile, and performance of Holstein steers (Song, et al., 2013;Ghassemi Nejad et al., 2016) receiving chromiummethionine (Cr-Met) supplementation. In those studies, 400 ppb/d of Cr-Met supplementation was reported as the optimum level. On the other hand, Sung et al. (2015) reported higher marbling score, rib-eye area, and meat yield index in Hanwoo steers fed a high-forage diet supplemented with 400 ppb/d of Cr-Met compared with the low-forage diet group. However, to the best knowledge of the authors, the effects of Cr-Met supplementation on daily pattern of rumen fluid parameters including volatile fatty acids (VFA), pH, ammonia-N, and fatty acid composition of meat from Hanwoo steers fed high-and low-forage diets is yet to be investigated. Monounsaturated fatty acids (MUFA) and saturated fatty acids (SFA) are the main components of the great majority of fatty acids in beef fat. Furthermore, beef fats are among the richest natural sources of conjugated linoleic acid (CLA) and trans-vaccenic acid (Chin et al., 1992), which have been shown to have health benefits for humans (Belury, 2002). Any increase in ruminal starch digestion should be followed by an increase in organic acids that are later converted to glucose, which is a precursor for marbling (Kitchalong et al., 1995;Lindemann et al., 2008). Any discussion of marbling levels should also include the genetic influence. Hence, there might be differences between Holstein and Hanwoo steers. Previous research has shown that forage-finished cattle produce beef with more CLA and n-3 fatty acids compared with grain-finished cattle (French et al., 2000). Melton et al. (1982) found that grassfed cattle beef had a decreased concentration of MUFA and a greater concentration of SFA compared with grain-fed cattle; however, one study found that grass-fed cattle beef had less SFA and more MUFA than grain-fed cattle beef (French et al., 2000). Moreover, it has been reported that Cr supplementation can be employed to manipulate the quality of meat due to its biological function on body fat and muscle metabolism (Kitchalong et al., 1995;Sung et al., 2015); however, factors such as the level of Cr supplementation and its source, nutrients, chromium levels in the basal diet, breed and species may intervene with these functions (Sung et al., 2015;Ghassemi Nejad et al., 2016). Decreasing diet energy density through increasing forage level may also increase ruminal acetate:propionate ratio and methane energy loss (Merchen et al., 1986;Kinser et al., 1988;Zinn et al., 1994). Therefore, the effects of high-and low-forage diet when supplemented with Cr-Met in diet of Hanwoo steers on rumen fluid characteristics and fatty acid profile of beef were investigated in this study.
Material and Methods
Animal procedures were approved by the Ethical Committee for Animal Research of Kangwon National University. The experiment site was the Gangwon province (37°55'03.64"N,127°46'22.71"E) in Republic of Korea. Twenty-three Korean native (Hanwoo) steers were used in this experiment and assigned to two treatment groups. The feeding amount of Cr-Met (Innobio Co., Ltd., Shiheung, Korea) to animals was limited to 400 ppb/cow/d. The duration for the study was two months. Feed was total mixed ration (TMR) including commercial concentrate and forage (alfalfa hay, bermudagrass hay, and rice straw) comprising 16±0.05%crude protein (CP) and 78±12.3total digestible nutrients (TDN). Treatments included what follows: steers fed diets that contained 60% concentrate and 40% forage supplemented with 400 ppb/d of Cr-Met (average body weight (BW) = 531.8±32.4;low-forage diet) (40F); and steers fed diets that contained 40% concentrate and 60% forage supplemented with 400 ppb/d of Cr-Met (10 head; average BW = 525.1±27.5;high-forage diet) (60F). Forages and concentrate were mixed and offered twice daily at 09.00 h and 18.00 h, according to nutrient requirements of beef cattle (NRC, 2000). Water was available ad libitum throughout the experiment.
For the period of two months, feed was collected twice a month to analyze the common feed ingredients according to methods of AOAC (1990; Table 1). Neutral detergent fiber (NDF) and acid detergent fiber (ADF) contents were analyzed following Goering and Van Soest (1991).
Each day, at 08.30 h, feed intake was measured by subtracting the amount of residue and was calculated by dividing the two values. Body weights were measured prior to the experiment and monthly. Average daily weight gain was measured subsequently. After 21 d of adaptation to the diet, for 3 d nonconsecutively (24, 26, and 28 d of each month), rumen contents (1 kg) were sampled through a rumen cannula (2 treatments × 2 animals) to describe the rumen pH, ammonia-N, and VFA content. Sampling was carried out before the morning feeding (defined as 0 h) and 0.5, 1.5, 2, 4, and 7 h after feeding on each of three nonconsecutive days (days 24, 26, and 28 of each experimental period). The rumen fluid was collected through the cannula and then filtered with four layers of cheese cloth within 1 min and measured using a digital pH meter (Model 420, S/N 049686, Orion Research, Inc. USA). Ammonia-N (ammonia nitrogen) in the rumen fluid was mensured by disconnecting the upper supernatant of the fluid and then diluting 10 times with automatic water quality analyzer (Quikchem, 8000). Finally, 12 mL of the diluted fluid was inserted into the centrifuge tube and then centrifuged at 3000 rpm for 15 min.
Volatile fatty acids (VFA; acetate, propionate, butyrate and valerate) of rumen were measured using a gas chromatograph (GC, shimadzu GC-17A, 30 m × 0.25 mm × 0.25 µm column). After filtration of the rumen fluid, metaphosphoric acid was added to precipitate proteins, and formic acid (5%) was used to eliminate the disturbing "ghosting" effect. The clear supernatant obtained after centrifugation (3000 rpm for 10 min) was injected directly into the column. The analysis of one rumen sample requires about 8 min, in which the temperatures of injector, column, and detector were -20 °C, 100 °C, and 230 °C, respectively. Depending on the implementation of the column temperature programing, helium (He) as a carrier gas with a flow rate of 7 mL/min and hydrogen (H 2 ) gas and air oxygen with a flow rate of 15 mL were used. The split ratio was 1:30, and 1 µL of the sample was used. For the quantitative determination, a nonautomatic integrator was used. The individual VFA were calculated by comparing with a standard.
Upon completion of the field experiment, five steers per group were slaughtered to measure the fatty acid composition of beef from the loin side. Samples from each steer were frozen at -20 °C for 12 h and thawed prior to analysis. According to the method of lipid extraction (Folch et al., 1957), 6 g of sample and chloroform/methanol (2:1) solution were homogenized in a 25 mL homogenizer (Diax 6000, Heidolph, Germany) at 1100 × g for 30 s. Next 6 mL of 0.88% KCl solution were added to the homogenate, followed by centrifugation at 2500 × g (GS-6R Centrifuge, Beckman, Ramsey, MN, USA) for 10 min. The fluid was filtered through filter paper and lipid was concentrated using a nitrogen gas concentrator (MGS-2200, Eyelaa Tokyo Rikakikai Co., Ltd, Tokyo, Japan) following the method of AOAC (1990). Each of the fatty acid methyl ester standards (Sigma-Aldrich Co., Saint Louis, MO, USA) was qualitatively compared with retention time, and the analytic conditions were used for gas chromatography (Agilent 6890N, Agilent Technologies, Santa Clara, CA, USA). For this, a sample of beef and tissue was kept for the split ratio of 1:10. The oven injector was heated with 220 °C.
A carrier of gas 1 mL/min was heated at 150 °C for one min.
A column HP-Innowax (30 m length × 0.32 id × 0.25 µm thickness) was kept for a detector temperature of 275 °C.
The oven maintained the temperature of 200 °C to 250 °C at 3 °C/min and 250 °C for 5 min.
Statistical analysis was carried out using the GLM procedure of SAS (Statistical Analysis System, version 9.1) for a completely randomized design. Duncan's multiple range test was used for ranking treatment means within a significant F test and means were considered significantly different at P<0.05. Means with probabilities between 0.05<P<0.10reflected a tendency to difference. All the data are reported as the sample mean±the standard deviation. Pairwise comparisons between means of different groups were performed using a t-test. The difference between two subsets of data is considered statistically significant if the t-test gives a significance level P (P-value) less than 0.05.
Rumen fluid parameters including pH, ammonia-N, acetate, propionate, butyrate and valerate were analyzed using PROC MIXED of SAS (Statistical Analysis System, version 9.1) for repeated measurement analysis (Littell et al., 1998) and the means were compared for significance by Tukey's test. The statistical model used for analyses was as follows: y ijk = µ + α i + d ij + τ k + (ατ) ik + e ijk , in which y ijk is the response at time k on animal j in treatment group i; µ is the overall mean; α i is the fixed effect of treatment; d ij is the random effect of animal j in treatment group i; τ k is a fixed effect of time k; (ατ) ik is a fixed interaction effect of treatment i with time k; and e ijk is random error at time k on animal j in treatment i. Variance and covariance assumption structures (AR(1), UN, CS, etc.) were tested, then AR(1) was selected as the best covariance structure for final analysis.
Results
Dry matter intake (DMI), body weight, ADG, and FE were not different (P>0.05) between the two groups (Table 2).
Rumen ammonia-N increased for the first 3 h of the post-feeding period and then decreased; however, no difference (P>0.05) was observed between the two treatment groups (Figure 1). Ruminal propionate, butyrate, and valerate showed no differences (P>0.05) between the two treatment groups (Figure 2).
Discussion
The similar DM intake in both groups may explain the lack of differences in body weight of steers and consequently ADG and FE between 40F and 60F groups (Table 2). It appears that the 20% difference in forage ratio between the two groups could not affect DM intake. Ohh and Lee (2005) concluded that Cr-Met supplementation may not improve body weight of animals. However, by contrast, it has been reported that a higher forage ratio may lead to lower DM intake and lower weight gain in steers in a longer period of growth (Song et al., 2013;Sung et al., 2015).
A timely decrease in pH post-feeding was predictable in both groups; however, in 60F group, these changes were slighter than in 40F group (Figure 1). Because of the higher concentrate ratio in 40F, a higher amount of soluble carbohydrate may result in lower pH. Ghassemi Nejad et al. (2012Nejad et al. ( , 2016) ) reported lower pH in Brown Swiss heifer and Holstein steers fed a higher concentrate ratio. Normally, pH begins to decline post-feeding and continues to fall down after 3 h. Other factors can influence pH decreasing trend, including but not limited to synchronization between carbohydrate and protein, rate of passage of rumen flora, digesta flow rate, and microbial protein production (Sung et al., 2015;Ghassemi Nejad et al., 2013).
A slight increase in ammonia-N content of rumen 3 h post-feeding and then a decline were expected due to the availability of nitrogen for both feeding groups. Microbial protein synthesis depends on the availability of nitrogen (or amino acids) and on the degradable carbohydrate content, or on the content of organic matter, and synchronization in rumen degradation between protein and carbohydrates. If there are high amounts of degradable nitrogen or if the content and rate of carbohydrate degradation in the rumen is reduced or not synchronized with the degradation of protein, losses of nitrogen and/or energy in the rumen could occur (Ghassemi Nejad et al., 2014, 2015).
Ruminal acetate was higher (P<0.05) in 60F compared with 40F group, and remained higher post-feeding (Figure 2). The higher acetate in rumen of 60F group can be explained by the higher forage ratio resulting in higher pH values. Higher pH values in 60F group favor rumen fiber digestibility and maintain stability of pH in due time (Sung et al., 2015). Moreover, a higher forage ratio in the diet causes higher secretion of saliva, which fosters higher and stable rumen pH. The content of volatile fatty acids and the acetate/propionate ratio may change due to different forage: concentrate ratios. Higher acetic acid may be observed in the rumen fluid of steers fed diets with higher forage ratios, as described by several researchers (Merchen et al., 1986;Kinser et al., 1988;Zinn et al., 1994). Therefore, a large amount of forage in 60F, which may have resulted in more salivary flow to rumen, provides favorable conditions for the better microbial growth. This may explain the higher acetate production in rumen fluid in 60F compared with 40F group. In 40F group, a higher concentrate ratio is a major factor for lower ruminal pH and consequently lower acetate production. Zinn and Plascencia (1996) found no significant differences in pH, acetate, propionate, and butyrate between feedlot cattle fed various levels of alfalfa (10% compared with 30%).
The meat fatty acid composition has a promising role in the quality of beef for consumers (Lindemann et al., 2008;Sung et al., 2015;Ghassemi Nejad et al., 2016). Chromium can influence fat mobilization from body stores to meet nutrient requirements. Hence, alterations in some fatty acid compositions can be expected. Chromium-methionine has a high potential to be absorbed through increased digestion in the gastro-intestinal tract (Ohh et al., 2004;Sung et al., 2015;Ghassemi Nejad et al., 2016). The finishing diet strongly influences the fatty acid composition of beef (Smith et al., 2009;Ghassemi Nejad et al., 2016). Zea et al. (2007) reported higher saturated fatty acids (SFA) in animals fed concentrates. Furthermore, Smith et al. (2009) reported that grain feeding stimulates the activity of adipose tissue stearoyl-CoA desaturase in marbling adipose tissue and lowers ruminal isomerization/hydrogenation of dietary polyunsaturated fatty acids (PUFA), resulting in a noticeable increase in monounsaturated fatty acids (MUFA) in beef over time. However, the current study found no differences in the SFA, unsaturated fatty acids (UFA), or UFA/SFA values, which were previously reported in Holstein steers by Ghassemi Nejad et al. (2016).
Conclusions
Chromium-methionine supplementation in high-forage (60%) diets can improve meat quality regarding fatty acid composition by increasing the palmitic, palmitoleic, and linoleic acid contents.
Figure 1 -
Figure 1 -Diurnal changes in ruminal pH and ammonia-N in Korean beef steers fed low-and high-forage diets supplemented with chromiummethionine (hourly basis).
Figure 2 -
Figure 2 -Diurnal changes in ruminal acetate, propionate, butyrate, and valerate acids in Korean beef steers fed low-and high-forage diets supplemented with chromium-methionine (hourly basis).
Table 2 -
Growth performance in Hanwoo steers fed high-and lowforage diets supplemented with chromium-methionine BW -body weight; DM -dry matter. Treatments included low-forage diet (40% forage; 40F) and high-forage diet (60% forage; 60F) supplemented with 400 ppb/d of chromium-methionine. Means ± standard deviation.ab -values within a row with different letters differ significantly at P<0.05.
Table 3 -
Fatty acid profile of meat from Hanwoo steers fed high-and low-forage diets supplemented with chromium-methionine SFA -saturated fatty acids; USA -unsaturated fatty acids. Treatments included low-forage diet (40F), and high-forage diet (60F) supplemented with chromium-methionine. Means ± standard deviation.ab -values within a row with different letters differ significantly at P<0.05.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-04-01T13:15:14.815Z", "created": "2016-08-01T00:00:00.000", "id": "88880761", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.scielo.br/pdf/rbz/v45n8/1806-9290-rbz-45-08-00483.pdf", "pdf_hash": "7558712f0866c95189f3f945d2ecaaf79d974500", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:34704", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "7558712f0866c95189f3f945d2ecaaf79d974500", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Volatiles play an important role in the attractiveness of food for trapping the stored grain pests Oryzaephilus surinamensis L. (Coleoptera: Silvanidae)
Insects pests of stored products can cause serious damage to stored grains and have threatened global food security for centuries. It is clear that new approaches and strategies are required including insects population monitoring technology and green control measures when insect resistance to pesticides is increasing worldwide, and increasing awareness of environmental stresses and human health concerns. Oryzaephilus surinamensis L. (Coleoptera: Silvanidae) (ORSU) is a major worldwide pest of stored products. In this study, six types of foods that require storage (oatmeal, figs, sucrose, hawthorn, cashew, millet, and wolfberry) were selected from a total of 38 different food products as the most attractive to these beetles. The mixture (M17) from mixtures of these six individual foods at different ratios that was the most attractive to ORSU and its attractiveness varied at different population densities of insects and trapping rates were highest at an environmental temperature of 25 °C. Interestingly, M17 remained effective as a trap bait for at least 40 days and also attracted other stored product insects (Tribolium castaneum, Cryptolestes ferrugineus, and Sitophilus oryzae). M17 volatiles were found to be perceived by O. surinamensis, in which three of these, nonanal, dodecane, and ß-caryophyllene were also behaviorally active; The chemical, tridecane, had no obvious EAG activity but was effective as a trap bait for O. surinamensis, and conversely, dibutyl phthalate (probably a contaminant), which had obvious EAG activity was not significantly attractive. Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 7 April 2021 doi:10.20944/preprints202104.0214.v1 © 2021 by the author(s). Distributed under a Creative Commons CC BY license. Our results show that food volatiles play an essential role in attracting stored-product pests and suggest that specific food mixtures could be used as effective lures to trap the pests. The use of food-based attractants could be an effective and environmentally friendly part of integrated pest management programs as trapping and monitoring tools.
Introduction
Pests have been harming natural resources, including trees, fruits, and other human foods, since before the rise of agriculture [1,2]. However, in addition to these insect pests in the field, a group of herbivorous insects, known as stored-product insects, began to appear as humans started to produce excess agricultural produce and store this as provisions over longer periods of time [2,3]. About a third of the total potential harvest of plant food crops are damaged or destroyed by pests during growth, harvest and storage [4]. Stored grain pests can cause major losses, and it is estimated that up to 9% stored grains in developed countries and 20% or more in developing countries can be lost through these pests [5]. A total of 1,663 known insect species have been recorded to damage post-harvest agricultural commodities and about a hundred are major, economically important pests [3]. How stored-product insects are to be effectively and safely controlled in order to decrease the losses of food during storage in warehouses is a huge challenge for humans, especially with the increasing human population.
In recent decades, because of its simplicity and effectiveness, chemical control, including fumigants and residual insecticides, has been the most common method to control stored product insects [6,7]. Unfortunately, because of the widespread use of pesticides, the environmental impact of these chemicals has become a significant problem, and resistance of the insect pests to the chemicals has also increased over the last five decades [8]. Resistance to fumigant insecticides in stored-product insects is now common [3,7]. In a global survey investigating resistance to the fumigant gas phosphine at 250 localities in 60 countries, six out of eight beetle species were reported to have developed resistance to the gas [7] and phosphine resistance has been reported in many types of insects including Oryzaephilus surinamensis (Silvanidae), Rhyzopertha dominica (Bostrichidae), Tribolium castaneum (Tenebrionidae) and Cryptolestes ferrugineus (Laemophloeidae) [9][10][11]. The most resistant T. castaneum population was found to be 119-fold more resistant than the susceptible strain, and the most resistant R. dominica population was over 1,500-fold more resistant [12]. A different study, also in Oklahoma, detected resistance to phosphine in four out of 14 populations of O. surinamensis adults and nine out of 14 populations of eggs, with resistance frequencies between 2 to 100% [13]. Some stored-grain insect fumigants have been banned. Methyl bromide has been used globally since the 1930s as a quarantine treatment for plants and to control insects in buildings and commodities. Methyl bromine, however, was found to be an ozone-depleting substance, and the bulk of its consumption has been banned since 2005 under the Montreal Protocol [6]. Therefore, with the increasing resistance of the insect pests to chemicals, and with environmental protection in mind, it is necessary to find new chemicals or new alternative methods that can effectively control pests while leaving humans and the environment unharmed [7,14,15]. system [3,5,17]. The use of attractants can lead to earlier detection of infestations, can accurately detect insect population levels, and can also help managers to take reasonable control measures and minimize pest damage before any chemical residues are deposited in the environment or in the food products [18,19]. Much progress has been made in controlling stored-product insects through food attractants and pheromones [15,20]. Experiments using closely-spaced pheromone traps for moths including Plodia interpunctella, Ephestia spp., and other species in different warehouses of grain showed that traps attracting males and disrupting mating were able to significantly reduce the population density of P. interpunctella and Ephestia cautella in the second year of trap use [15,21]. Interestingly, another study showed that traps baited with agave tissue as well as pheromones were able to capture more Scyphophorus acupunctatus weevils (Coleoptera: Cuculionidae) than could traps baited with either pheromones or agave alone [22].
There has been significant research into food lures, and many products have been tested. There has been significant scientific interest in behaviors and mechanisms of responses of stored-product insects towards traps baited with food attractants [5]. As early as the 1980s it was reported that adult O. surinamensis beetles could be attracted in laboratory assays using Y-tube olfactometers baited with rolled oats and pentane extracts of oats [23]. A previous report suggested that the "secondary pests" O. surinamensis, T. castaneum and Tribolium confusum were primarily attracted to kernels which had been previously damaged by "primary pests" [24]. Results of experiments investigating the attraction of Lasioderma serricorne (F.) towards 11 stored products, demonstrated that tobacco, cocoa, soybean flour, black tea, and wheat flour significantly attracted the beetles [25]. Crawling insects Trogoderma granarium and O. surinamansis tend to be attracted to traps with food lures such as walnuts, oats or wheat germ oils [26]. Mee oil (Madhuca longifolia) and coconut (Cocos nucifera) oil are also able to attract T. castaneum adults under laboratory conditions [27]. Furthermore, food based attractants are commonly used to capture sawtoothed grain beetles, merchant grain beetles, rice weevils, granary weevils, and rusty grain beetles [28]. Several commercial attraction trap products have also appeared on the market, including the multi-species trap for stored-product pests Xlue MST, which consists of two specialist food lures and three pheromones, is advertised as being able to trap twelve different stored-product pest species, and which is widely used throughout the world ( [URL]/). The volatile chemicals in the food may play a role in determining the attractiveness of food baits, and food volatiles have been well studied as potential attractants for storedgrain pests [29,30]. A two-choice pitfall olfactometer has been used to examine the behavioral responses of granary weevils to twenty individual food volatiles, including aliphatic alcohols, aldehydes, ketones, and aromatics, showing that the beetles found eight of the tested compounds attractive, at at least one concentration [31]. Volatiles from kibbled carob have also been shown to attract three species, Sitophilus zeamais, Sitophilus oryzae, and Sitophilus granarius [32]. A great deal of literature also exists addressing the attractiveness of food volatiles to other pest species [33][34][35][36].
The stored-product insect Oryzaephilus surinamensis L. has achieved a worldwide distribution and is regarded as a major stored grain pest, and not only attacks stored grains and their products, but also poses a potential threat to public health as it can cause allergies in humans [37,38]. The presence of this insect causes global damage to grains and grain products with an estimated cost of seventy billion dollars annually [39].
In the present study, the relative attractiveness to O. surinamensis (ORSU) of thirtyeight different foods was estimated. A series of food mixtures was then designed, each with six selected foods, and the attractiveness of these mixtures was assessed both in the laboratory and in a large flat granary. The function of the food volatiles and bioactive compounds in those volatiles in attracting stored-product insects was also analyzed using both bioassays and electrophysiological methods. We found that some food mixtures, in-cluding the food mixture M17, are effective as trap attractants to ORSU, and several chemicals in the eluents of volatiles from food mixtures may play important roles in the attractiveness of foods to these beetles.
Insects
Adult Oryzaephilus surinamensis L. (ORSU) beetles were collected from Yuezhou Provincial Grain Reserve Storage (Shaoxing, China). Four other insect species, Tribolium castaneum, Sitophilus oryzae, Cryptolestes ferrugineus, and Rhyzopertha dominica were collected from Zhongsui Provincial Grain Reserve Storage (Xiaoshan, China). The insect populations were then reared on a mixture of whole wheat flour, oatmeal and yeast (whole wheat flour : oatmeal : yeast = 6 : 3 : 1), and grown in an incubator at 30 ± 1 °C and 70% -75% relative humidity (RH), in darkness [40]. Between 20-30 days after emergence, adult beetles of both sexes were collected held in centrifuge tubes without food for 24 h in darkness at 31 °C, before behavioral bioassays were conducted.
Foods
Thirty-eight types of foods, including oatmeal and dried figs, were chosen as potential attractants. These foods are listed in Table S1. All of these foods were purchased from a Wall-Mart supermarket, Lin'an district, Hangzhou, China. A series of food mixtures consisting of six foods (oatmeal, dried hawthorn, millet, sucrose, cashew, and dried fig) were assembled using an orthogonal design based on three weights of each food (Table 1).
Determination of relative attractive activity of single food
The six-armed olfactometer consisted of a glass cylinder (diameter 5 cm, height 11.8 cm). The bottom of the chamber had six arms (diameter 1.5 cm, length 23 cm in length, with two arms extended at 45°), with a 50 mL pear-shaped flask connected to the end of each arm. For each experiment the flasks were prepared as follows: four of the flasks were each filled with 2.5 g of a single kind of attractant, one flask was filled with the same weight of flour for a standard control and one was left empty as a blank. 50 adult O. surinamensis beetles were placed in the center of the olfactometer cylinder. The olfactometer together with the insects was then placed in the climate incubator in the dark, and at 31 ± 1 °C and 70% RH. After 24 h, the number of insects preferring each food was counted. A beetle was deemed to have chosen a particular food if it had moved down the arm of the olfactometer by more than 5 cm. The experiment was repeated three times and the relative attractiveness of each single food was calculated as the ratio of the average number of beetles preferring this food over other foods or flour.
Attractiveness of food mixtures in the lab
To measure the effectiveness of the eighteen food mixtures, M1-M18, as trapping baits, a plastic box (38 cm × 26 cm × 18.5 cm) was filled with rice to a depth of about 13 cm, and 200 adult insects were uniformly distributed throughout the box 2 h before the experiment. Twenty 50 mL type traps (diameter 3 cm, length 11.5 cm) with nine gaps (length 1.5 cm, wide 0.2 cm) ( Figure S1) were evenly inserted in four rows and five columns into the rice, keeping the top gape circle of the trap (Figure S1 d) parallel with the rice surface. A food mixture was randomly assigned to eighteen of the twenty traps, and the other two traps were left empty as a control. The box was placed in the dark at 28 ± 2 °C and 70% RH for 24h. After 24 h, the number of insects in each tube was recorded and the attractiveness of each food mixture was calculated as the number of insects trapped in tube divided by the total number of insects in all the tubes. These experiments were repeated three times.
Evaluation of the effect of temperature and insect density condition on the attractiveness of food mixture M17
A plastic box (24 cm × 34 cm × 18.5 cm) was filled with rice to two thirds of the height of the box height. Two 50 mL type traps (diameter 3 cm, height 11.5 cm) were each baited with 7.6 g food mixture M17 as described above, or with nothing as a control. All the trap tubes were arranged in a ring, and adjacent tubes were kept 13 cm apart ( Figure S2). Each tube was inserted into the rice so that the top gape circle of the trap was flush with the rice surface. 20 adults of O. surinamensis were uniformly distributed over the surface of the rice, and the box was placed in the dark, with 70% RH at 28 ± 2 °C for 24 hours. Experiments were performed with an insect density of 1, 2.5, 5, 15, and 30 insects per kg rice and at temperatures of 20, 25, and 30 °C. The number of insects trapped in each tube was recorded after 24 h, and the experiment was repeated three times.
To evaluate the attractiveness of food mixture M17 on different species of storedgrain pests, a 50 mL type trap was used, with most flasks baited with 7.6 g food mixture M17 and one left empty as a control. Four species of stored-grain pests, Tribolium castaneum, Sitophilus oryzae, Cryptolestes ferrugineus, and Rhizopertha dominica were investigated, and 30 insects from one species were used in each experiment. The experiments were kept in an incubator at 30 ± 1 °C and 75% RH in the dark.
Determination of the time for which food mixture M17 remains effective
In the lab, 140 insects were dispersed equally through rice in a plastic box (24 cm × 34 cm × 18.5 cm). Two 50 mL traps were inserted into rice 20 cm apart, and the traps were baited with either 7.6 g of food mixture M17 or nothing, as a control. The number of insects in each tube was recorded regularly after 1, 3, 7, 14, 25, 35, and 40 days. After each observation, the trapped insects were removed from the traps, and an equal number were added to the rice box; the positions of the two traps were also swapped each day. The experiment was conducted in darkness at 28 ± 2 °C and 70% RH, and was repeated three times.
Further experiments were conducted in a granary in Zhejiang Yuezhou Industrial Co., Ltd., (Zhejiang, China) from June 21 to August 1 in 2019. The granary was 25.2 m × 18 m, and contained about 2000 T rice. For the granary experiments, 100 mL type tube traps (diameter 4 cm, height 11 cm) were used. In each experiment, three tubes were baited with 7.6 g, 50 g, or 100 g of food mixture M17 respectively, and a fourth was left empty as a control. The four tubes were inserted into the rice and arranged in a 40 × 40 cm square, and the places of the four tubes were exchanged randomly after each insect count had taken place.
The trapping rate and the time for which the commercial attractant Xlure MT ( [URL]/) and food mixture M17 remained effective were determined using the methods described above with slight modifications. One Xlure MST and three Xlure MST trap sets ( Figure S3, [URL]/) baited with 7.6 g, or 22.5 g of M17 or left empty as a control were placed on the surface of the rice arranged in a 20 × 20 cm square. Eighty adult insects were released in this experiment.
Collection of food mixture volatiles and GC-MS analysis
Volatiles released from the food mixture M17 were collected as previously described [40]. 3 g of food mixture M17 was placed in a 35 mL glass collecting bottle, and a vacuum pump was used to extract gas at 400 mL/min. An adsorption tube containing 40 mg of absorbent PoraPak TM Q (80-100 mesh) was connected between the bottle and the vacuum pump to absorb the volatile matter from food mixtures M17 ( Figure S4). After 8 h of adsorption, each adsorption tube was eluted with 200 μL dichloromethane, and subsequently stored at -40 °C.
Chemical analysis was conducted using a GCMS-QP2010 Plus equipped with an HP-5 capillary column (30 m × 0.25 mm id, 0.25 μm film). The carrier gas was helium, with a flow rate of 1 mL/min. Samples (2 μL) were injected into the column with a split ratio of 1:1. The component separation temperature was programmed to rise from 100°C to 250°C at a rate of 8°C/min and was then held at 250°C for 10 min 25s. The percentage composition was calculated by integrating the GC peak area normalization. The identification and quantification of compounds in the food mixture volatiles has been described in [41] based on standard samples and mass spectrum Library NIST11. LIB analysis. The retention index under temperature programmed conditions was calculated for each selected compound [42].
The attractiveness of food volatiles and chemicals as estimated using an olfactometer.
Six-armed or four-armed olfactometer assays ( Figure S5) were carried out as previously described [41] with little modification. The air flow in each arm of the olfactometer was 200 mL/min, and in each experiment, 90 insects (six-armed olfactometer) or 60 insects (four-armed olfactometer) were released. After 25 min, the number of insects that had moved down an olfactometer arm at least 5 cm were recorded. Each assays was repeated four times. For volatile eluents and chemicals, 20 µL solution at different concentrations was applied to a natural rubber septa (for diameter 1 cm tubes), and the solvent was allowed to evaporate into the air for 1 min before the olfactometer experiment took place.
GC-EAD and EAG Assays
To determine whether the chemicals from food volatiles were able to stimulate O. surinamensis antennae, a GC-EAD analysis was conducted. Nitrogen with a flow rate of 1 mL/min was used as a carrier gas [41]. Samples (1 μL) were injected (splitless mode) into the column and the oven temperature was programmed to hold initially at 100 °C for 1 min, and was then ramped to 250 °C at 8 °C/min, with a final hold for 10 min 25 s at 250 °C. The active components were identified by retention index and GC−MS technology. At least five O. surinamensis antennae were measured for each treatment.
For EAG recordings, an adult O. surinamensis head with antennae was excised from its body using a scalpel. The tips of the antennae were cut slightly and inserted into one glass capillary, and another glass capillary was connected to the severed head neck to form a closed electrical circuit [41]. Different concentrations of volatiles or chemicals were prepared in dichloromethane, and in each experiment, 10 μL of one preparation was added to a piece of folded filter paper (3 cm × 1 cm), and allowed to evaporate for one minute before being placed into a glass Pasteur pipette as the stimulus for the EAG recordings. Dichloromethane alone was used as a control at the start and end of each recording. Average antenna EAG values of tested volatiles or chemical (EV), solvent di-chloromethane (Ec) and 1000 mg/L benzaldehyde (ES) were recorded, and the EAG responsive index EA of antennae to volatiles or chemicals was calculated according to the formula: EA= (EV-EC)/ (ES-EC) [43]. For each EAG assay, at least six replicates of treatment were carried out.
Data analysis
The experimental data were analyzed using the DPS© Data Processing System 18.1 [44] ( [URL]/). The relative attractiveness of single foods, the attractiveness of food mixtures, the effect of the food mixtures under different temperatures and insect densities, and the EAG response data were analyzed using a one-way ANOVA. A total of 18 food combinations were designed using orthogonal experimental design. For the olfactometer data, we used generalized linear mixed models (GLMMs) with Poisson distribution of error, and the replicates were treated as the random factor. The data of trapping rate of the individual foods and the food volatiles were statistically analyzed after an arcsine transformation.
Relative attractiveness of foods to O. surinamensis and efficiency of attraction of food mixtures
Thirty-eight foods showed different relative attractiveness, ranging from 0.02 to 13.70 based on the attractiveness of flour to ORSU (Figure 1). The seven foods with higher relative attractiveness from 13.70 to 4.32 were oatmeal, dried fig, sucrose, hawthorn, cashew, millet, and wolfberry in order (Figure 1). Because the soft dried wolfberry fruits go bad easily, wolfberry was not selected as a component of any of the food mixtures (Table1). The six foods with the highest relative attractiveness were combined into a "cocktail" food attractant, and 18 food mixtures were designed using an orthogonal design (Table1). The trap bioassay results suggested that the food mixtures M17, M16 and M2 have the highest trapping rates, 10.55, 6.63, and 6.62%, respectively ( Figure 2A). Bioassay data suggested that the six food mixtures, including M17, were able to trap ORSU effectively, and food mixture M17 had the highest attractiveness, about 1.24 -1.66 times that of the other five food mixtures and significantly higher than the attractive activity of either M6 or the blank control treatment ( Figure 2B). Compared to the commercial attractant trap Xlure MST, the trapping rate of food mixture M17 was about 2.66 -10.72 times as high ( Figure 2C). The trapping rate of traps baited with 1 g food mixture M17 were all significantly higher than those baited with Xlure MST or the control treatments over the total 21-day duration of the experiments, and at the 21st day, the trapping rate of traps baited with M17 was 66.70%, which was 5.63 and 33.86 times higher than those of Xlure MST and the control, respectively ( Figure 2C). Food mixtures M17 was therefore used for further analysis.
Temperature and insect density of ORSU affect trapping rate
Both insect population density and environmental temperature were able to influence how many ORSU were trapped by traps baited with food mixture M17. At an insect population density of one insect per kg grain, 19.05% of ORSU could be trapped in 24 hours. At a population density of 2.5 ORSU per kg grain, traps baited with M17 reached a maximum trapping effectiveness of 58.10%, but with further increases in insect population density, the trapping rate of traps baited with M17 decreased to minimum of 17.70% at 30 insects per kg grain ( Figure 3A). Trapping rate of M17 traps were higher given insect population densities of 2.5-15 insects per kg grain at 1 and 30 insects per kg grain ( Figure 3A). It is interesting that all three studied temperatures, there were significant differences between attractiveness of M17 and the control group, and the trapping rates of M17 traps were 12.31, 4.88, and 9.76 times those of the blank control treatments at the same temperature. The highest trapping rate observed for the M17 traps was 65.00% at temperature 25 °C, which was obviously higher than the 18.99% and 19.44% observed at temperatures of 20 °C and 30 °C, respectively.
Time over which food mixture M17 remained attractive
In the lab, with an insect density of 10 insects per kg grain, food mixture M17 remained attractive for a significant time. After 40 days, the trapping rates of M17 traps over 24 h were 69.44% -88.15%, and all were significantly higher than those of the control treatments ( Figure 4A). On the first day of trapping, the M17 baited tubes were able to trap 88.15% of the ORSU, with the average 24 h trapping rate slowing to 69.44% after seven days, then maintaining a rate of between 86.99% and 77.38% to forty days. After forty days, the attractant M17 under laboratory conditions was still effective, with a trapping rate of 77.38%, which was significantly higher than the control ( Figure 4A). Moreover, both 7.6 g and 22.8 g of M17 could attract significantly more insects at all time points after seven days and 38-47 days compared to both the the positive control (the commercial attractant Xlure) and the blank control ( Figure 4B). It is interesting that no differences were found between the trapping rates of traps baited with 7.6 g and 22.8 g M17 at any time points ( Figure 4B) in the lab. In addition, both the attractiveness and the time over which food mixture M17 remained attractive were measured in a granary containing 2000 T rice. Measured over a 1 -40 day period showed significant differences in the trapping rate between the blank controls and the treatment traps (F3, 54=3.5790, P=0.0196) were visible each day ( Figure 4C). In the first three days, the traps baited with 100 g M17 always had the highest trapping rate, which was significantly higher than that of the blank treatment. At seven days, all three M17 treatments obviously trapped more ORSU than did the blank treatment, and at 40 days, the average trapping rate of traps baited with 7.6 g M17, were 50.67%, higher than any other treatment and apparently also higher than that of the positive control ( Figure 4C). Attractiveness of food mixture M17 in the large-scale granary over 40 days. CK, control. The letters represent significant differences between treatments at the same time points (P < 0.05) and stars indicate significant differences between treatment and control as tested with t-tests (*, P<0.05; **, P<0.01).
Food mixture M17 was able to effectively attract other stored grain pests
Food mixture M17 was obviously attractive not only to ORSU, but also to four other stored-grain pests (T. castaneum, C. ferrugineus, S. oryzae and R. dominica), with trapping rates of 41.11%, 46.67%, 68.89%, and 8.89 %, respectively, which are about 12.35, 14.01, 4.13 and 8.01 times those of the control treatment ( Figure 5A). The volatiles from M17 were collected and the antenna EAG responses to these volatiles were then assessed. Consistent with the M17 trapping rates of the four pests ( Figure 5A), the responses of the antennae from all four pests to M17 volatiles were markedly higher than to the control treatment, with the EAG values ranging from 0.09 to 1.17 mV ( Figure 5B). However, in comparison to the control, the M17 volatiles were only successful in trapping more C. ferrugineus or S. oryzae ( Figure 5C). Behavioral responses of different stored grain pests to eluent of the volatiles from food mixture M17. B, blank treatment control; SC, control treatment with 10 μL dichloromethane; M17, food mixture M17. Stars represent significant differences between the treatments at the same time as tested with t-tests (*, P < 0.05; **, P<0.01); The letters represent significant differences between treatments at the same time points (P < 0.05).
The volatiles from food mixture M17 were obviously attractive to ORSU
In order to find out whether the volatiles from food mixture M17 played an important role in its attractiveness to O. surinamensis, we performed a series of six-arm and four-arm olfactometer assays. The results show that ORSU preference for M17 volatiles was significantly (27.24%) higher than for any of the other five treatments, including the flour treatment and the blank control ( Figure 6A). Nevertheless, no preference was found between volatiles from sucrose, cashew, fig, millet, or oatmeal singly, five of the six components of M17, compared to the blank control ( Figure 6B and 6C). The volatiles from the sixth component of M17, hawthorn, were, however, significantly more attractive to ORSU than the other treatments and the blank control (36.67%), (C), which suggests that hawthorn volatiles may play a major role in the attraction of ORSU to the food mixtures. We then compared the attractiveness of volatiles released by the hawthorn slices with those from food mixtures M17, and found that the number of ORSU preferring food mixture M17 was about 4.73 times that preferring hawthorn alone ( Figure 6D). The volatiles released by food mixture M17 were collected by headspace dynamic adsorption and eluted with dichloromethane. We ran an experiment comparing ORSU attraction to the eluent from the M17 volatiles to dichloromethane and to a blank control. The results indicated that compared with the blank, and dichloromethane, eluent from the volatiles from food mixture M17 was significantly more attractive to ORSU, with trapping rates of 40.08% with ( Figure 7A) and 31.25% without ( Figure 7B) airflow in the tubes of a four-arm olfactometer. There were no significant differences between the electrophysiological responses of ORSU antennae to volatiles collected from 3 g or 300 g M17, however, the antenna EAG values of volatiles eluent from 3 g and 300 g food mixture M17 was 1.96 and 2.24 times as high as the EAG values from the solvent control, respectively, both with significant differences ( Figure 7C). The antenna EAG value of the response to M17 volatiles eluent was 0.157 mV, which was apparently higher than the 0.032 mV response of the antennae to the solvent control ( Figure 7D), however there were no differences between the antenna electrophysiological responses to volatiles eluent when the eluent was diluted 10-10 000 times ( Figure 7D).
The bioactive compounds in food mixture M17 volatile eluent
GC-EAD analysis indicated that eight compounds in the volatiles eluent from the food mixture had obvious antenna electrophysiological activity (Figure 8). We were able to identify six of these compounds using GC-MS and comparisons with chemical standards and retention index confirmation. The chemicals were identified as nonanal (peak 2, with a retention time (RT) of 5.32 min); dodecane (peak 3, RT 6.51 min); tridecane (peak 4, RT 8.07 min); β-caryophyllene (peak 5, RT 10.23 min); methyl palmitate (peak 6, RT 16.98 min), and dibutyl phthalate (peak 7, RT 17.52 min) (Figure 8 and Table 2). Table2. Chemicals in food mixture M17 volatiles that resulted in ORSU antenna electrophysiological activity.
a Retention index (RI) retrieved from the NIST08 mass spectrometry database; b Retention index calculated in relation to the series of n-alkanes (C7-C40) in lab; c Relative percentage of the identified volatiles based on GC-FID; d Retention index obtained from the literature [40].
Discussion
Food based lures or attractants have been successfully used to control a variety of different pests, including tephritid fruit flies [45], Drosophila suzukii [46], Mediterranean fruit fly [47], cockroach [45,48] and others. A great deal of work on the food selection behaviors of stored-product pests and attractants has also been carried out. The attractiveness of 30 different plant-based foods, most of which are used as condiments, including paprika (Capsicum annuum L., Solanaceae), cinnamon (Cinnamomum spp., Lauraceae), and turmeric (Curcuma longa L., Zingiberaceae) as potential food attractants for Lasioderma serricorne (F.) has been tested, and it was discovered that mulberry leaf tea was the strongest attractant of the tested materials [49]. In our assay, of 38 kinds of food, six foods (oatmeal, dried figs, sucrose, dried hawthorn, cashews, and millet) were particularly attractive to adult O. surinamensis, and we found that oatmeal was the most attractive to these beetles ( Figure 1). Oatmeal was also reported to be the grain most attractive to adults of the the stored-grain pest Cryptolestes turcicus (Grouvelle), with selection rates as high as 41.1% in a bioassay experiment [50]. Our results also showed that hawthorn played an important role, but also that the trapping activity of food mixture M17 cannot be attributed solely to hawthorn ( Figure 6C and D). In a laboratory, flight tunnel studies have confirmed that chocolate products and volatiles isolated from chocolate were attractive to Ephestia.cautella and Plodia interpunctella [51]. However, we found that chocolate was not particularly attractive to ORSU (Figure 1). This all suggests that the attractiveness of a certain food is very different to different insect species, and also that traps baited with food mixtures may be able to trap more insects than those with single foods alone. Furthermore, the proportions of different foods in food mixtures are critical. In our experiments involving 18 food mixtures, M17 was the mixture with the highest and steadiest attractiveness to ORSU (Figure 2).
Stored-product pests are primarily thermophilic in nature, and their growth and survival is greatly influenced by temperature [52]. Our data suggest that temperature is able to influence the attractiveness of food attractants. Food mixture M17 worked well at all three temperatures (20, 25, and 30 °C), but at 25 °C, M17 seemed to be most attractive to ORSU, and trapping rates reached about 65%, while at under 20 °C and 30 °C, trapping rates of traps baited with M17 were about 21% ( Figure 3). The different trapping rates of food attractants at different temperatures may be due to beetle behaviors such foraging and crawling varying at different temperatures, and temperature can greatly impact the amount and speed of volatiles released from food attractants, factors that we were unable to exclude from our experiments. Previous reports have indicated that between 11 °C and 35 °C, ORSU walking speed increases with increasing temperature [53]. [54] verified that at 2 ºC, emissions of total volatiles and specific monoterpenes, mainly limonene, but also linalool and α-terpineol from star ruby red grapefruit were enhanced, while storage at 12 ºC resulted in higher emissions and diversity of cyclic sesquiterpenes and aliphatic esters.
As well as the environmental temperature, the insect population density in the grain influenced the effectiveness of the traps baited with food attractants. At an insect density of 2.5 insects/kg, M17-baited traps reached their highest trapping rates of 58.10%, however, interestingly, with an increase in insect density from 2.5/kg to 30/kg, the trapping rate decreased ( Figure 3A). A previous study demonstrated that insect population density profoundly affected the olfactory responses of O. surinamensis, and that low populations of beetles in culture (including when under starvation conditions) decreased the length of the refractory period to beetle and frass volatiles and advanced the onset of the period of positive olfactory response [55]. Moreover, the trapping percentages of pheromone traps have been found to increase with increasing T. castaneum population sizes, with the highest trapping percentage observed when beetles were released in population sizes 300, 350 or 400, and the lowest trapping percentage observed when the population size was 100 [56].
In this study, we tested the attractiveness of food mixtures not only in the laboratory, but also in a large, flat granary ( Figure 4). Our data suggested that the food mixture M17 was the most attractive to ORSU, with the maximum trapping rate of 75.56%, and that this mixture remained attractive for at least 40 d. Low quantities (7.6 g) of M17 as well as larger amounts (to 100 g), were able to trap insects ( Figure 4C). We also found the attractiveness of both 1 g or 7.6 g of M17 was stronger than that of the commercial attractant Xlure MST trap after 21 days ( Figure 2C) or 47 days ( Figure 4B), respectively. The price of the food attractants tested here is much lower than that of the commercial product Xlure MST. Furthermore, the Xlure MST trap could attract more than one species of stored-product insects, and our food mixture M17 was also attractive to three (T. castaneum, C. ferrugineus, S. oryzae) of four tested insects ( Figure 5A). We therefore believe that food mixture M17 has commercial potential and may be worth popularizing as a broad spectrum insect attractants, particularly once its attractiveness to further species of stored-product insect has been tested.
The behavioral and electrophysiological assays show that the volatiles from food mixtures may play an important role in the attractiveness of foods to insects (Figure 6-9). The air flowing over M17 attracted significantly more ORSU than did that flowing over the control, flour, or other three food mixtures ( Figure 6). Furthermore, eluents of volatiles showed effective trapping activity ( Figure 7A-C). [57] also reported that volatile compounds were a factor involved in location and selection of foods by Liposcelis bostrychophila when in proximity to a food. Insects, like many other animals, rely on chemical signals to find food and mates [58].
In general, insects are able to recognize volatile odors swiftly through their antennae. Using GC-EAD technology, we identified eight chemicals as potentially bioactive compounds in the volatile extracts of the food attractant. We were able to identify six of these chemicals as nonanal, dodicane, tridicane, β-caryophyllene, methyl ester, and dibutyl phthalate using GC-MS (Figure 8 and 9, Table 2). Interestingly, nonanal has been identified from both oatmeal [59] and brown sugar [59], and dibutyl phthalate has also been identified from brown sugar [60]. We found that β-caryophyllene prompted significant electrophysiological activity in the test antennae as well as ORSU behavior, and β-caryophyllene is well known as a chemical attractive to several insect species, including Trigonotylus caelestialium (Kirkaldy) (Heteroptera: Miridae) [61], and Asian honey bees (Apis cerana) [62]. However, only four of the six compounds isolated from the food volatiles in our experiment were able to elicit a significant antennal EAG response from ORSU, at a series of different chemical concentrations ( Figure 9A). Moreover, not all compounds with significant EAG activity were attractive to the living ORSU insects. For example, dibutyl phthalate was no more attractive to the insewcts than the control ( Figure 9A and B), and tridecane, which had no significant EAG activity compared to the control, showed apparent trap activity ( Figure 9A and B). [63] investigated the GC-EAD responses of the antennae of the kudzu bug Megacopta cribraria (Hemiptera: Plataspidae). Six GC-EAD active compounds were identified from volatiles collected from undamaged potted plants, but the antennal responses of the bug, to farnesene and ocimene differed from the behavioral responses. Differences between antennal and behavioral responses also have been reported in Lygus hesperus Knight (Heteroptera: Miridae) [64], and Maruca vitrata Fabricius (Lepidoptera: Crambidae) [65]. Under natural conditions, insects are confronted with complex mixtures of compounds and their responses to these odor blends can only be evaluated in behavioral studies [66].
In our experiments, four chemicals, nonanal, dodecane, tridicane and β-caryophyllene were attractive to ORSU, while the attractiveness of the other two chemicals, methyl palmitate, and dibutyl phthalate, were low and may be detected as background odors. In many cases it appears that the volatile organic components of a target are not all always essential for attraction [67]. We infer that these two compounds may also be used as background odor substances to enhance the recognition of other important active compounds, or they perhaps play a role in the mixture [68,69]. However, we found that the background odors also have some role in the attractiveness of the food mixture, and that ORSU was able to more easily distinguish mixtures of all six compounds than of mixtures comprising fewer chemicals compared to the control ( Figure 9D). [31] showed that adults of the stored-grain pest species Sitophilus granarius (L.) (Coleoptera: Curculionidae) were attracted by three compounds at lower concentrations, but that these chemicals acted as repellents at higher concentrations. They also mention that the adult beetles have the ability to respond behaviorally to a wide range of cereal volatiles, and that the responses may change as a function of the chemical concentration. Interestingly, the chemical mixtures of the six compounds we isolated from eluents of M17 volatiles did not attract ORSU at 100 times concentration more effectively than the control (Figure 9 C), but at 1 times concentration, this mixture worked well (Figure 9 D). Unlike Sitophilus granarius, which is able to respond effectively to the three test compounds at a wide range of concentrations, it is possible that ORSU may only be attracted to chemical mixtures at a narrow range of concentrations.
Conclusions
In conclusion, this study provides insight into the possibility that food mixtures could be developed as effective attractants and the volatiles used by stored-product insects to locate them also could be utilized as lures of chemical mixtures as one form of semiochemical-based management of ORSU. Moreover, the food volatiles may have potential as the basis for the development of semiochemical-based IPM approaches as well as pheromones for the control of stored-grain pests. The food mixtures or food volatiles may work as lures not only to control populations of stored-product insects, but also to monitor their population dynamics in a granary. Much remains to be studied in the field of interactions between stored-product insects, food based attractants and environmental factors such as temperature and humidity, but we believe that food-based attractants may have a future as one part of an IMP program in the effective control of stored-grain insects.
Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Figure S1: Schematic diagram of insect trap. a, gape holes around the insect trap; b, plan of 50mL trap decvice; c, the top circle of gape holes around the trap is kept parallel to the surface of the grain, Figure S2: Diagram of attraction experiment in plastic frame. a, insect trap containing food mixtures; b, plastic frame; Figure S3: Schematic diagram of the commercial trap Xlure, Figure S4: Schematic diagram of volatile collection device. a, air pump; b, e, flowmeters; c, adsorption tube; d, collection bottle; f, activated carbon. Figure S5: Schematic diagram of four-armed olfactometer.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-05-04T22:05:18.371Z", "created": "2021-04-07T00:00:00.000", "id": "233566359", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://www.preprints.org/manuscript/202104.0214/v1/download", "pdf_hash": "8318fbc96e131820b920e4491d3a71de0674818d", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:32929", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "e3f5798ea0d1c8d1be1d5ae62f92bac2330dbbab", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Development and reproductive performance of beef heifers supplemented with brown rice meal and / or protected fat on temperate grasslands
The response of energy supplementation was evaluated on the development and reproductive performance of beef heifers on temperate grasslands. Twenty-eight Charolais × Nellore heifers, with initial average age of 18 months and initial live weight of 274.9 kg were utilized. The animals were maintained on oat + ryegrass pasture and distributed in the following treatments: no supplementation (NS): heifers kept exclusively on pasture; MEG: supplementation with protected fat Megalac®; BRM: supplementation with brown rice meal; BRM+MEG: supplementation with BRM + protected fat. The average final weight of the heifers was of 403.4 kg and corresponded to 89.5% of the adult weight. The body condition of heifers increased linearly with daily increase of 0.012 points, correlating positively with the final weight. There was interaction between treatment and period for average daily weight gain. The highest daily weight gain, 1.395 kg, occurred in the first period when the animals were supplemented with BRM+MEG. In the last period, the NS animals presented the lowest daily weight gain, 0.888 kg. Supplementation with brown rice meal and/or protected fat does not interfere in the intake of pasture by heifers or increase the total intake of dry matter, not changing, therefore, the average daily weight gain at the end of the period of grazing. The daily weight gain does not change during supplementation. The use of temperate pasture with and without supplementation promotes the proper development of the structure and reproductive tract of heifers, benefiting the animal performance indexes in the first mating at 25-27 months of age.
Introduction
Rearing replacement heifers is certainly one of the most important stages within the reproductive system aimed at breeding, where improving the reproduction of young females, through the additional inclusion of food in the process of growth is important for increasing the productivity of this system (Barcellos et al., 2003). According to ANUALPEC (2010), in the last 10 years, the productivity of the Brazilian breeding system was almost stagnant, still with a low birth rate, near 60%.
Much of the low animal productivity of most Brazilian properties happens due to the deficiency in pasture management and lack of forage and property planning (Lupatini, 2000). In this context, the incorporation of temperate grasslands exactly during the period of greatest forage need has crucial role for beef cattle. According to Semmelmann et al. (2001), the main objective of a replacement rearing system is to develop heifers that have reached puberty and cycle regularly before the beginning of the first mating season. Di Marco et al. (2006) highlight that a nutrient management that enables the females to reach their minimum weight of mating with a certain advance is necessary, since it is important to obtain higher pregnancy rate in heifers, in addition to the good development until the mating has a long-term effect on their reproductive performance.
The use of supplements rich in fat such as brown rice meal and protected fat may improve the use of temperate grasslands, because, according to Pascoal et al. (2000), these pastures present high crude protein content, and the energy level is the limiting factor for the animal performance. However, the types of lipids used in the diet can influence the fermentation and the ruminal digestibility of the fibre, through the suppression of cellulolytic and methanogenic bacteria, and saturated lipids behave in a less harmful way to the microbial flora (Duarte et al., 2005). Besides the possible benefit of energy supplementation, the intake of extra lipids, 44 g/kg in the diet, may be beneficial to the reproductive development, resulting in an increase in the percentage of pubertal heifers at the beginning of the breeding season (Lammoglia et al., 2000).
Thus, the response of the supplementation with brown rice meal and/or protected fat was evaluated on the development and reproductive performance of beef heifers on temperate grassland.
Material and Methods
The experiment was carried out in the Beef Cattle Laboratory in the Department of Animal Science of Universidade Federal de Santa Maria. This area is located in the Central Depression of Rio Grande do Sul, at 95 m of altitude, latitude 29º 43' South and longitude 53º 42' West. The soil of the experimental area is owned by the São Pedro mapping unit and classified as a paleudalf soil (Embrapa, 1999). The climate of the region is Cfa (subtropical humid), according to the Köppen classification.
The experimental area used corresponded to 16.3 ha, of which 11.7 ha were divided into 12 paddocks with variable area where the animal tests were managed, and 4.6 ha where the animal regulators remained. The implementation of the pasture occurred on April 1st, 2009 with the broadcast sowing of 31.2 kg/ha of ryegrass seeds (Lolium multiflorum Lam.) based on 100% of cultural value, and on April 2nd, 3rd and 4th, 2009 with sowing of 77.4 kg/ha of oat seeds (Avena strigosa Schreb.)based on 100% of cultural value in a row. In the based fertilizer, 141 kg/ha of NPK fertilizer formula 5-20-20 were used. The cover fertilization was accomplished in four stages: June 4th, 2009 -47 kg/ha of urea; June 16th, 2009 -124 kg/ha of NPK fertilizer formula 5-20-20;July 11th, 2009 -77 kg/ha of urea;and August 22nd, 2009 -38.5 kg/ha of urea.
The pasture establishment was of 70 days, when the heifers joined the pasture; the last 15 days before the trial period were used for adaptation to the supplement and to the management. The experimental period lasted 112 days, from July 5th, 2009 to October 24th, 2009, divided into four periods of 28 days.
Twenty-eight Charolais × Nellore heifers, with initial age of 18 months and average live weight of 274.9±4.97 kg were distributed in the following treatments: no supplementation (NS): heifers maintained exclusively on oat + ryegrass; Megalac ® (MEG): heifers maintained on oat + ryegrass pasture receiving 3% of protected fat under the estimate of the total intake of dry matter fixed at 30 g/kg live weight; brown rice meal (BRM): heifers maintained on oat + ryegrass pasture receiving brown rice meal at a level of 8 g/kg of live weight; and brown rice meal + Megalac ® (BRM+MEG): heifers maintained on oat + ryegrass pasture receiving brown rice meal at a level of 8 g/kg of live weight and 3% of protected fat under the estimation of total dry matter, fixed at 30 g/kg live weight. Each treatment consisted of three area replications, with a variable number of animals within them; two paddocks with two heifers each, and a paddock with three heifers.
The weight of the animals was measured before the beginning and end of each period of the experiment, previously fasted for 12 hours of solids and liquids. During the weighing, the body condition of heifers was assessed, assigning scores from 1 to 5, in which 1 = very thin and 5 = very fat, following the method described by Lowman et al. (1973). The pelvic area was measured rectally, with the aid of the Rice pelvimeter before the start and the end of the trial period, determining the width or horizontal measure, which corresponds to the distance between the right and left ileus, at the height of the psoas tuber, and the height or vertical measure, which corresponded to the distance between the symphysis of the pubis and the base of the sacral vertebrae body. The horizontal and vertical measurements were calibrated in centimetres and then multiplied to obtain the estimate of the pelvic area in cm 2 . The reproductive tract score, assessed at the beginning and end of the experimental period, was determined according to the methodology described by Anderson et al. (1991), with heifers sorted according to the scores: infant (1 or 2), pre-pubertal (3) and pubertal (4 or 5).
The hip height and perimeter of the chest were evaluated at the beginning and at the end of the experimental period with the aid of a measuring tape and graduated ruler with the immobilization of the animal in torso restraint, staying with the dorsal line straight.
Before the start of the breeding season, comprised between November 15th, 2009 and February 28th, 2010, the bulls for breeding were selected through andrologic examination, utilizing a bull:cow ratio of 1:20. The determination of gestational age was performed after 45 days of the end of the breeding period, through ultrasound, then it was determined by the difference between the day of the examination and the beginning of the breeding season, the probable day of conception.
The grazing method adopted was continuous with variable stocking rate, using the "Put and take" technique (Mott & Lucas, 1952), from the use of a pre-determined forage mass of 1200 kg DM/ha, utilizing regulator animals to maintain the forage availability desired. Forage mass was determined by the double sampling technique (Wilm et al., 1944). Of each cut made in the repetition, a sample was taken for the composition of the composite sample for determination of dry matter (DM). For the calculation of the estimate of dry matter (DM) intake per treatment and period, the final forage mass of the period and the forage losses occurring during the period were subtracted from the total dry matter/ha production in the period. Dividing the estimated intake of DM/ha by the ability of animal support in the period, the DM in g/kg of body weight for each period was obtained. The rate of substitution and addition were calculated from the following formulas: replacement = [(forage DM intake of non-supplemented heifers -forage DM intake of the supplemented animals)/DM intake of the supplement]; addition = [(Total DM intake of supplemented animals -forage DM intake of non-supplemented animals)/ DM intake of the supplement] * 100.
For sampling of forage consumed by heifers, grazing simulations were performed in each experimental period according to the method of Euclides et al. (1992). The samples were pre-dried in an oven with forced air circulation at 55 °C for 72 hours until they reached constant weight, then they were ground in a Willey-type mill of 1 mm mesh sieve.
The dry matter content was determined by drying in oven at 105 °C until constant weight. The total nitrogen content was determined by the Kjeldahl method (AOAC, 1995). The ether extract content was determined after treating the samples with ether in a reflux system at 180 °C for 2 hours. The content of neutral detergent fibre is in accordance with Van Soest et al. (1991) (AOAC, 1995). Total digestible nutrients were calculated from the chemical composition of the food using the equation of Weiss et al. (1992).
The experimental design was completely randomized, with three replicates per area, in a 4 × 4 factorial arrangement (four treatments × four periods). The variables were tested concerning normality by the Shapiro-Wilk test, in which all presented normal distribution. The total weight gain, heart girth, hip height, pelvic area and weight/height ratio, which consisted of only two measures, at the beginning and end of the experiment, were subjected to analysis of variance and F test at 5% of significance using PROC GLM. The pelvic area was initially used as a covariate. For pregnancy rate, the chi-square test was utilized. Other data were subjected to analysis of variance and F test at 5% significance level using the PROC MIXED, and the information criterion for choosing the best covariance structure was the AIC, and when differences between the means were detected, they were compared by Student's t test.
The mathematical model adopted in the analysis of variance was: Y ijk = µ + T i + R k (T i ) + P j + (TP) ij + e ijk in which: Y ijk represents the dependent variables; µ, the mean of all the observations; T i , the effect of the i-th dietary treatment; R k (T i ), the effect of the k-th repetition within the ith treatment (error a); P j , the effect of the j-th period; (TP) ij , the interaction between the i-th treatment and the j-th period; and e ijk , the total experimental error (error b).
Regression test, lack of fit test (Lack-of-fit), and correlation test at 5% significance level were performed. Data analyzes were performed using the statistical package SAS (Statistical Analysis System,version 8.01).
Results and Discussion
The ingestion of supplements for treatments Megalac ® , brown rice meal and brown rice meal + Megalac ® , showed means of 0.115; 2.430 e 2.594 kg/day, respectively. The Megalac ® supplement intake for the treatment which received only the product was inferior to the treatment which was mixed with the brown rice meal, a fact that was associated with their low palatability when provided as the sole source of supplementation, which stops occurring when mixed with another product accepted by cattle.
The contribution of crude protein (CP) of the pasture (Table 1) was above the requirements for maintenance and gain for beef heifers over the year, 126 g/kg of dry matter (DM), according to the NRC (1996), which would enable the use of only energy supplementation up to 11.2 g/kg of live weight (LW) with no protein deficit. Regarding the content of total digestible nutrients (TDN) (Table 1), the oat + ryegrass and brown rice meal showed similar levels of energy, 722.0 and 720.3 g/kg DM, respectively; the highest participation of TDN per unit of product was with the Megalac ® supplement with 1635 g/kg DM.
There was significant interaction between treatments and periods for the weight of the animals (Figure 1). The evolution of the weight was constant for all treatments; however, it can be seen that the heifers that received MEG and BRM+MEG gained more weight at the end of the experiment, when the forage quality began to decrease due to the advancement of the reproductive stage and consequently in function of increased lignification of forage intake. The extra input of energy via Megalac ® probably benefited this increased weight gain at the end of the experimental period, because when using BRM or just grazing, the performance observed was lower. The average weight of heifers at the beginning of the experiment of 274.90±4.97kg accounted for 61.09% of the mature weight of Charolais × Nellore cows, 450 kg. According to Barcellos et al. (2003), the heifer is ready for mating when achieving at least 60-65% of the body weight of an adult cow; however, for British × Zebu crosses, this percentage may be slightly higher. Thus, with the autumn-winter period, when the stagnation of the growth of native grass has already initiated by May, and the quality deteriorates considerably by June, the maintenance of heifers exclusively on pasture seeking to mate would lead to loss of weight and body condition, which would compromise the reproductive indexes.
In this study, the average final weight reached by the heifers was of 403.4±5.83kg and corresponded to 89.5% of the adult weight, in which the rearing of females on oat + ryegrass pasture with or without supplementation enabled the heifer to achieve the adequate weight for mating. For Barcellos et al. (2006), the weight of the heifer is the variable of greatest impact on reproductive performance during its first mating.
The body condition of heifers increased linearly (Figure 2) with each additional day of 0.012 points, correlating positively with the final weight (r = 0.79, P<0.0001).
Body condition becomes an important tool to assist the producer in the search for better reproductive rates especially for heifers that will enter the first mating season, because according to Rice (1991), it is one of the nutritional status indicators that is most directly associated with the pregnancy rate. Cows with higher body condition score at the start of the breeding season are those that phenotypically present higher body weight, lower value for days to calving, and higher success rate at calving than cows with lower body condition score (Mercadante et al., 2006). In this study, both the quality of pasture and supplements offered were decisive for the heifers to present high weight and high body condition score at the end of the experimental period (3.91 points).
There was interaction (P<0.05) between treatment and period for average daily weight gain (Table 2). The occurrence of interaction between treatment and period in temperate grassland is reported by other authors (Roso & Restle, 2000;Rocha et al., 2004;Pilau et al., 2004).
The highest daily weight gain, 1.395 kg, occurred in the first period when the animals were supplemented with BRM+MEG. One could infer that the daily weight gain during this period was the result of a likely compensatory gain; however, daily weight gain was very close to that observed in the third period, that is, the diet supplied allowed the heifers to express high daily weight gain during different periods. In the last period, the animals with no supplementation presented lower daily weight gain, 0.888 kg. This lower result is probably related to the body condition presented by heifers, averaging 3.79 points, i.e., during this period, heifers had a higher fat deposition, and as they did not have extra energy intake via supplements, as in the other treatments, the daily weight gain was lower due to lower energy density of the dietary intake. Santos et al. (2005), working with beef heifers in temperate grasslands also observed greater daily weight gain: 1.425 kg in the initial period (July to August) when the animals were supplemented (9 g/kg BW) with soybean hulls, and daily weight gain of 0.580 kg for calves grazing only on pasture in the last period (September-October). The joint analysis of the data shows that the daily weight gain does not remain homogeneous during the use of the pasture, with occurrence of periods of higher or lower daily weight gain.
During the period of pasture use, weight gain in kg/ha showed quadratic behaviour: Y = 165.16996-3.11581*Day+0.02622*Day 2 (R 2 = 0.55, P<0.0001) (Table 3). The inflection observed in the trend line for the weight gain in kg/ha occurred between August 2nd and August 30th, 2009, concurrently with the least ability of animal support and daily weight gain observed by the grazing in that period (Table 2). At the end of pasture use, with the highest accumulation rate observed in this period, it was possible to increase the ability of animal support, and the maintenance of daily weight gain close to or higher than the other periods enabled possible treatments to have higher weight gain (kg/ha/period) in the final grazing cycle. The data from Rocha et al. (2003) corroborate this result, in which the greatest weight gain in kg/ha occurred in the same period of this experiment, at the end of September to mid-October, but with higher values when supplement was used (231.5 kg/ha), but close when the animals were raised on pasture (163.5 kg/ha). According to Rocha et al. (2003), the data on live weight gain/ha quantify the potential for animal production of different alternatives of pasture utilization, and also enable the verification of the economic return of each strategy utilized.
In relation to the total weight gain (kg/ha; Table 3), there was no difference (P>0.05) between treatments, with gains ranging from 345.27 kg (MEG) to 473.25 kg (BRM+MEG). These values are in accordance with data found in the literature with variations of 298.3 kg/ha without supplementation on oat + ryegrass pasture (Roso, 2007) to 1039.3 kg/ha with increasing supplementation over the period of temperate grasslands (Freitas et al., 2005). Another factor that tends to make the overall gain in kg/ha vary is the number of days of pasture use, which was 112 days in the present study. In favourable weather, Roso & Restle (2000) managed beef heifers for 182 days, and even with a reduction in daily weight gain in the last period, the total weight gain/ha reached 726.3 kilograms, that is, even with the reduction in daily weight gain in the last days of grazing, the increase of grazing periods tends to increase the total weight gain (kg/ha).
The dry matter intake of pasture (Table 4) was not affected (P>0.05) by the supplement provision, so that the total dry matter intake also did not change. Even with a reduction in DM intake of pasture for treatments BRM and BRM+MEG, the provision of the supplement complemented the dry matter intake by heifers. These variables could explain the lack of variation in average daily weight gain, mainly because the TDN content of oat + ryegrass pasture and of the brown rice meal are similar (Table 1). Thus the replacement of the forage by the concentrate had no effect on the increase in average daily weight gain, with additive effect of the concentrates (Table 4).
The dry matter intake of pasture for all treatments of this study is above the level recommended by the NRC (1996), 26.3 g/kg BW. However, the literature has similar, as well as higher data on pasture intake values than those obtained in this experiment. Bremm et al. (2005) found mean value of 38 g/kg BW of pasture dry matter intake for animals without supplementation, and when they received wheat bran corresponding to 10 g/kg BW, the forage intake decreased to 28 g/kg BW, but during the experimental period, these authors observed extreme intake between August 18th and September 14th, 2002, which corresponded to 59 and 53 g/kg for animals not supplemented and supplemented with 5 g/kg BW, respectively. Pötter (2008), grouping several studies in one analysis, found forage intake of 40 g/kg BW for animals exclusively on pasture and 34 g/kg BW for animals supplemented with 9 g/kg BW. The dry matter intake, by the technique used, may be overestimated by errors in estimating initial and final forage production, trampling, insect activity and intake by non-experimental animals (Minson, 1990, apud Pötter, 2008).
The diets did not influence (P>0.05) the circumference of the chest and hip height of heifers at the end of the experiment (Table 5). The final chest circumference showed a high correlation (r = 0.73; P<0.05) with the final weight of heifers, so that, under grazing conditions that do not limit intake, the increase in thoracic perimeter tends to promote greater capacity for DM intake and consequently a better performance from animals, best explained by the correlation between final chest circumference and average daily weight gain, r = 0.42 (P<0.05).
In this experiment, the diets offered had 407.3 g/kg DM of NDF in the maximum, therefore not limiting intake, about which, according to Van Soest (1994) values of 550.0-600.0g/kg DM of NDF in the diet offered correlate negatively with feed intake. When the hip height was correlated with the final weight, the observed value was of 0.45 (P<0.05), the same found by Thompson et al. (1983) for the same characteristics. The results also agree with Montanholi et al. (2008), in which the assessment of correlations between chest circumference and hip height shows that the perimeter of the chest is a linear measure more appropriate to estimate the live weight than the hip height, because, according to Barker et al. (1988), the size of the skeleton is less susceptible to variations from the medium than the weight, in addition to their earlier maturation.
In the present study, there was correlation (r = 0.65; P = 0.0002) for weight/height and chest circumference and correlation (r = 0.90; P<0.0001) for the weight/height ratio and final weight of the heifers, values below those from Thompson et al. (1983), who found r = 0.85 and r = 0.96, respectively. The weight/height ratio did not differ (P>0.05) between treatments, presenting very close values, showing that body development was similar between heifers. The weight/height ratio can be considered a better estimator of body structure of the animal than its live weight, since it gathers data on body weight, which depends on body composition, with the height, expressing a qualification of the animal size (Barcellos et al., 2003). Thus the average daily weight gain of 1.147 kg for 112 days had an increase of 18 cm in chest circumference and 6 cm in height, so that the weight/height ratio increased from 2.14 to 2.99, demonstrating the structural growth that heifers had prior to mating.
Because the heifers of the present experiment were crossbred Charolais × Nellore, that is, their racial composition leads to animals that tend to have later puberty, especially when compared with animals of European blood, it becomes essential that by the moment of mating, animals have already reached puberty and be on a regular cycle. Hall et al. (1995), testing two types of animals with different growth rates, fast and medium, observed difference for the weights at puberty: 390.1 versus 361.7 kg for animals of medium and rapid growth, respectively; however, the age at puberty was similar for both (401.9 vs. 398.5),and the weight/height ratio was the same, 3.1 kg/cm. Fox et al. (1988), in turn, consider the weight/height ratio of 2.77 kg/cm as one of the suitable factors for the onset of puberty in heifers of the intermediate frame. The manifestation of puberty in ½ Charolais ½ Nellore and ½ Nellore ½ Charolais heifers managed on pasture cultivated in two subsequent winter periods occurred when the weight of 346 kg was reached at 20 months (Restle et al., 1999). In the present study, the moment of puberty onset was not verified; however, the animals showed the same weight as reported by Restle et al. (1999), approximately at 60 days of grazing at the same age mentioned by the respective authors.
At the end of the trial, the examination of the reproductive tract pointed to all treatments of score above 3 (Table 5), which, according to Anderson et al. (1991), determines pubertal heifers that can already conceive. The non-occurrence of variation between treatments in reproductive tract score can be explained by the high daily weight gain of more than 1.080 kg, i.e., in favourable environmental conditions for animal performance, the inclusion of concentrate in the diet did not alter the evolution of the reproductive tract. The response obtained in this experiment is consistent with the results of Montanholi et al. (2004) with heifers mating at 18 months, who found that when the average daily gain was higher than 0.700 kg, the reproductive tract score did not change. Even reaching the weight considered suitable for breeding at 18/20 months, 65% of the adult weight, Polled Hereford heifers with average daily gain of 0.137 kg showed less development of the reproductive tract, reproductive tract score of 1.9 in relation to heifers with average daily gain of 0.616 kg, which showed reproductive tract score of 3.5, which resulted in a significant percentage of pregnant heifers, 25 vs. 75%, respectively (Souza, 2009).
The development of the reproductive tract presented by heifers in the period before mating was crucial for the high pregnancy rate achieved (Table 5), where all treatments had 100% of pregnancy, with the exception of heifers supplemented only with MEG, which presented 85.71% of pregnancy, without statistical difference between treatments. It is pertinent to note that when an ultrasound was performed to verify pregnancy, it was found that a calf from the MEG treatment had not yet manifested puberty, which contributed to lower pregnancy rate, although it is an exception that is not related to daily supplementation but to specific factors of the animal, since in the treatment BRM + MEG, no variation was observed.
On average, the pregnancy rate of 96.30% is consistent with that reported by Silva et al. (2005), 86.7% in Hereford heifers at 24 months of age and with weight of 350.6 kg at the beginning of the mating. There was the possibility of improvement in the pregnancy rate of heifers in function of the provided supplements, which had high levels of lipids that could have positive effects on the reproduction of heifers, regardless of the input energy, since it has been demonstrated that the use of lipid supplements positively affects important reproductive functions in several tissues, including the hypothalamus, anterior pituitary, ovary and uterus, according to Funston (2004). However, this effect became null due to the non-occurrence of variations in the body, structural and reproductive tract growth of heifers. The pregnancy rate observed in this experiment is extremely satisfactory and should be considered as the objective of production systems aiming at the productive efficiency of the dams. The high rate of pregnancy, especially for heifers that will conceive for the first time, should not be the sole purpose during the breeding period; one should seek the conception of the heifers in the initial stage of mating aiming at calving as soon as possible so that the primiparous can recover the body state in time to conceive again in the next mating season.
In this regard, the daily weight gain shown by heifers associated with the structural development and evolution of the reproductive tract was crucial for 88% of heifers conceiving within the first 30 days. This variable is important for the next breeding season, since healthy cows with appropriate nutritional management tend to show estrus from 40 to 50 days postpartum, that is, they will tend to have their cycle regularly in the next breeding season (Rovira, 1996). Another important factor refers to the conception in the initial period of the first mating of heifers, taking effect on the production system, because they will give birth earlier in the birth season of the following year, weaning heavier calves and in greater numbers throughout their productive life (Lesmeister et al., 1973).
Conclusions
Supplementation with brown rice meal and/or protected fat does not interfere in the intake of pasture for heifers or increase the total intake of dry matter, so the average final daily weight gain is not changed. The use of temperate pasture with or without supplementation promotes proper development of the structure and reproductive tract of heifers, improving the animal performance indexes in the first mating at 25/27 months of age.
Figure 1 -
Figure 1 -Evolution of the live weight of heifers supplemented with different concentrates, over the period of use of oat + ryegrass pasture.
Figure 2 -
Figure 2 -Estimate of body condition of heifers during the grazing periods.
Table 1 -
Average contents of chemical analysis of forage from the grazing simulation, brown rice meal and Megalac ®
Table 2 -
Means and regression equations for daily weight gain of heifers receiving different supplements on oat + ryegrass pasture
Table 4 -
Estimates of pasture dry matter intake and total dry matter (pasture + supplement) intake, replacement rate and addition rate
Table 5 -
Means and standard error of initial and final chest circumference (CC), hip height (HH), gains of CC (GCC) and HH (GHH), weight/height ratio (W/H ratio), pelvic area (PA), reproductive tract score (RTS) and pregnancy Means followed by different letters in the row differ (P<0.05) by the t test.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-03-30T13:04:40.902Z", "created": "2012-10-01T00:00:00.000", "id": "85646407", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/rbz/a/yYNj5GXHd6f8m5RQGrBZQjc/?format=pdf&lang=en", "pdf_hash": "0681e28a0d463132a5fb20277ed95fd6aea93f08", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:34537", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "0681e28a0d463132a5fb20277ed95fd6aea93f08", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Development of a Multipurpose Core Collection of New Promising Iranian Pomegranate ( Punica granatum L.) Genotypes Based on Morphological and Pomological Traits
: Establishment of a core collection, of limited size and better representation of the whole germplasm phenotypic diversity, is fundamental for fruit tree breeding programs from an economic and management points of view. To achieve this goal with pomegranate fruit trees, 221 genotypes were evaluated for 25 morphological and pomological traits during two successive years. Using the maximization strategy in Power Core software, 12 out of 221 pomegranate genotypes were selected for the new core collection, reducing the population size to 5.42% of the entire collection. Variance difference (VD%) and mean difference (MD%) were calculated as 42.68% and 7.03% in core collection, respectively. This indicates an excellent diversity amongst genotypes within the core collection. The Shannon’s diversity index ( H (cid:48) ) in the formed collection suggested that 19 out of 25 phenotypic variables were of high diversity. Results showed that core collection genotypes are equally presented in all three population groups formed by cluster analysis through the original collection. The current research is the first in using phenotypic data to establish a core collection of Iranian pomegranate germplasm. The formation of this core collection will be an effective step towards examining the diversity of the original population and breeding prospects.
Introduction
Pomegranate (Punica granatum L.) is native to central Asia, particularly some areas of Iran and that was broadcasted to other parts of the world [1][2][3][4]. Extensive cultivation of pomegranate in tropical and subtropical regions in a new changing climatic context indicates the high adaptability and flexibility of this species [5]. Pomegranate fruits are characterized by their high antioxidant properties and phenolic contents, which can amend metabolic syndrome [6,7]. With an annual production of more than 1 million tons and a cultivated surface of 89,400 ha, Iran is one of the largest pomegranate producers in the world [8].
Phenotypic variability in this tree species is the result of the creation of none "trueto-type" seedlings of pomegranate due to about 13% outcrossing [9] and morphological changes created through the domestication process. The genetic and phenotypic base of these variations in Iranian pomegranate germplasm has been studied using morphological characteristics [10,11] and molecular features [12]. Accordingly, various studies have been conducted worldwide on the biological diversity of pomegranate to collect, preserve, and evaluate its germplasm [13][14][15].
Despite the large number of pomegranate accessions with different phenotypic characteristics throughout the world, only a few are commercially acceptable and widely cultivated. Therefore, due to the selection of superior accessions for breeding purposes as well as orchard establishment, most of the existing genetic diversity was not included in the commercial cultivation of this fruit species [16]. The pomegranate breeding infrastructure includes highly variable pomegranate collection [17], segregating populations for important morphological and pomological features, assembled transcriptome, SNP markers, and a genetic map [18]. Fruit related characteristics have high potential for discrimination among different pomegranate accessions [15]. Traits with commercial significance in pomegranate which have the most goals in its breeding program are included tree yield, fruit skin color, aril color, fruit size and shape, seed softness, active pharmaceutical compounds, and resistance to fruit cracking and sunburn [16,19].
Classifying genetic resources for the identification of superior genotypes is of utmost importance for the successful design of breeding programs, and has a significant role in facilitating the management of genetic collections' conservation [15]. Gene banks can provide researchers with genetic resources for long-term protection of plants, mainly perennial fruit trees, and rich germplasm sources. Still, the formation of a core collection, more limited in size and more representative of the genetic diversity of the whole gene bank, is an effective approach in terms of spending less time and money in the production and breeding context [20,21].
Formation of most core collections based on the so-called maximizing method using the MSTRAT software is done by optimizing the number of traits for germplasm protection, while the number of accessions and diversity in the collection determines the size of the core [22]. Genetic diversity and variability determine sample size, which cannot be a uniform standard because of different crop species' number and the creation of specific characteristics through evolution and human intervention [20]. The core collection size varies from 5 to 30% of the entire population [23,24]. According to Brown's [25] suggestion, a sample size of 5 to 10% of the entire collection with a maximum of 3000 samples per species can preserve about 70% of the alleles of the entire collection. Over the past decade, a core collection has been created in several perennial fruit tree species, such as apricot [20,26], olive [27], apple [28], and walnut [29].
To the best of the researcher's knowledge, no study or guideline exists on establishing a core collection for pomegranate fruit trees based on neither molecular nor phenotypic traits. Hence, the present study aimed to create a multipurpose core collection based on the evaluation of different morphological and pomological attributes. It should contain the diversity present in the Iranian National Pomegranate Collection (INPC) so that while keeping the most important phenotypic traits, superior genotypes are identified.
Plant Material
The INPC held in Yazd Province was used as a plant material supplier for this study. Since 1987, more than 760 new promising pomegranate genotypes from the whole country have been collected in Yazd pomegranate collection in Agriculture and Natural Resources Research Centre of Yazd (31 • 55 N, 54 • 16 E, and 1216 m alt.). Trees in this collection were collected from different regions of Iran and asexually propagated. To represent as much of the diversity of the germplasm, 221 out of more than 760 genotypes of this collection were selected based on preliminary phenotype characterizations due to its superior behavior. Supplementary Table S1 indicates that most samples are originated from the central provinces of Iran and provides the genotype accession codes and names. The accession codes are based on the province code and sample number in the Yazd collection.
Agro-Morphological Evaluation
A total of 25 traits related to tree appearance, leaf, flower, and fruit were evaluated during two successive years (2012-2013) (The average monthly air temperature for the growing season = 28.2 • C, average annual values = 20.8 • C and average amount of precipitation = 27 mm for those two years) according to international descriptor of International Union for the Protection of New Varieties of Plants [30] by obtaining 30 samples from each accession. Evaluated traits as designated with their respective codes and categories are presented in Supplementary Table S2.
Establishment of the Multipurpose Core Collection
Development of the core collection was performed based on analysis of the morphological attributes, using the maximization strategy in the PowerCore V.1.0 software as proposed by Kim et al. [31]. In order to evaluate superior genotypes that will constitute the core germplasm, four parameters were evaluated including variance difference percentage (VD%), mean difference percentage (MD%), changeable rate of the coefficient of variation (VR%), and coincidence rate of range (CR%).
Statistical Analysis
Descriptive statistics were carried out for both the entire and core collections. Morphological diversity for each morphological and pomological trait was estimated using the H index according to Shannon [32]: where H is the diversity index, Pi is the proportion of each phenotypic trait in the sample, and ln(Pi) is the natural logarithm of this proportion. In factor analysis and according to construction factorial coefficients matrix, the factors with eigenvalues greater than 1.0 were evaluated [33]. The percentage of explained variability for each factor and the communalities for each trait were determined, and the scatterplots were established based on two first factors. Besides, the non-parametric Spearman correlation approach was used to assess the correlation between traits with a breeding purpose. The genetic dissimilarity component was analyzed using Euclidean distance, and the agglomerative hierarchical clustering algorithm was performed by Ward's method. Statistical analysis and drawing the plots were performed by R software 4.0.4 and XLSTAT software (Version 2016.1).
Description of Morphological and Pomological Traits
The observations on 25 traits were recorded for all 221 genotypes. Data analysis is presented in Supplementary Table S2 and Figure 1. Based on frequency distribution, a high variability was recorded for the number of fruits per tree, fruit mean weight and thorn in mature wood branch, fruit size, fruit skin thickness, fruit symmetry, and aril color. In contrast, tree crown shape, length-to-width leaf ratio, fruit skin sensitivity to sunburn, fruit crown shape, aril size, and seed hardness showed less variability. The descriptive statistics of minima, maxima, mode, median, and H are presented in Table 1. Modified from Eticha et al. [34] and Jamago and Cortes [35], an ideal classification of high (H ≥ 0. Most pomegranate genotypes showed a wide tree crown shape, a strong tree growth power, and a medium length-to-width leaf ratio. In addition, most of the genotypes formed flowers on the yearly branches with lateral positions and medium cup diameters. The high fruit skin sensitivity to burst, medium fruit skin sensitivity to sunburn, and medium fruit skin thickness were the most common. Also, most of the genotypes had medium fruit size, sour-sweet flavor, medium aril size, and light red aril color with hard seeds (Tables 1 and S2). Most pomegranate genotypes showed a wide tree crown shape, a strong tree growth power, and a medium length-to-width leaf ratio. In addition, most of the genotypes formed flowers on the yearly branches with lateral positions and medium cup diameters. The high fruit skin sensitivity to burst, medium fruit skin sensitivity to sunburn, and medium fruit skin thickness were the most common. Also, most of the genotypes had medium fruit size, sour-sweet flavor, medium aril size, and light red aril color with hard seeds (Tables 1 and S2).
Correlations between Morphological and Pomological Traits
Flower position showed a positive correlation with fruitful flower percentage, fruit symmetry, and fruit crown shape and a negative correlation with dominant flowering habit and fruit shape. A positive relationship was observed between flower cup diameter and productivity traits such as mean tree yield, mean fruit weight, and number of fruits per tree. There was a negative relationship between length-to-width leaf ratio with fruit skin sensitivity to sunburn and burst. In addition, it was found that the fruit skin sensitivity to burst had a positive relationship with the fruit skin sensitivity to sunburn and a negative correlation with fruit skin thickness ( Figure 2).
Multivariate Analysis
The results of the factor analysis using the Varimax Rotation method [36] allowed the classification of the 25 morphological traits in seven main groups, which justified and covered 76.53% of the total variance. The first and second factors each justified 25.41 and 11.93% respectively, and 37.34% of the cumulative variance. Examining the relationship between traits and factors showed that fruit skin color, shape, and crown shape are most related to the first factor ( Figure 3A), and flower position and fruitful flower percentage were dominant in the second factor ( Figure 3B). The third factor (11.24%) consisted of traits including fruit skin sensitivity to burst and sunburn, fruit skin thickness, and fruit ripening time. The fourth to seventh factors explained 9.26, 7.79, 5.88, and 4.99% of the total variance, respectively. The classification of the 25 evaluated traits divided the traits into four separate groups. In the first group, tree-related attributes LWLR, TCS, TGP, and DFH were included. AS, SH, FSH, and FSC were in the second group. FP, FFP, and five traits related to fruit established the third group. In the fourth group, there were ten traits that were mainly related to fruit yield and appearance ( Figure 2).
Multivariate Analysis
The results of the factor analysis using the Varimax Rotation method [36] allowed the classification of the 25 morphological traits in seven main groups, which justified and covered 76.53% of the total variance. The first and second factors each justified 25.41 and 11.93% respectively, and 37.34% of the cumulative variance. Examining the relationship between traits and factors showed that fruit skin color, shape, and crown shape are most related to the first factor ( Figure 3A), and flower position and fruitful flower percentage were dominant in the second factor ( Figure 3B). The third factor (11.24%) consisted of traits including fruit skin sensitivity to burst and sunburn, fruit skin thickness, and fruit ripening time. The fourth to seventh factors explained 9.26, 7.79, 5.88, and 4.99% of the total variance, respectively. A two-dimensional scatter plot was drawn based on the first and second factors, and genotypes were grouped from 1 to 10 groups (Figure 4). A two-dimensional scatter plot was drawn based on the first and second factors, and genotypes were grouped from 1 to 10 groups (Figure 4). A two-dimensional scatter plot was drawn based on the first and second factors, and genotypes were grouped from 1 to 10 groups (Figure 4). According to k-mean partitions comparison analysis based on Calinski criterion, K = 3 was selected as the best k ( Figure 5A), and 221 genotypes were divided into three main groups ( Figure 5B). According to k-mean partitions comparison analysis based on Calinski criterion, K = 3 was selected as the best k ( Figure 5A), and 221 genotypes were divided into three main groups ( Figure 5B). On the other hand, consistent with the results of K analysis, the cluster analysis of the assayed genotypes, using the ward's method and Euclidean distance, revealed three main groups. The first (C1) and the second (C2) groups consisted of 26 and 60 genotypes, respectively. The remaining 135 genotypes were classified in the third group (C3) with two sub-clusters i.e., C3a and C3b, consisting of 10 and 125 individuals, respectively (Figure 6). The relationship between traits and each group of genotypes was determined. Based on these results, fruit shape, fruit skin color, and dominant flowering habit had the most, and fruit crown shape had the least with the first group. Fruit ripening time, fruit skin thickness, fruit size, flower cup diameter, fruit mean weight, fruit flavor, tree mean yield, and number of fruits per tree showed the most association with the second group. The traits most related to the third group were fruit skin sensitivity to sunburn and burst, flower position, fruitful flower percentage, and fruit crown shape (Figure 7). On the other hand, consistent with the results of K analysis, the cluster analysis of the assayed genotypes, using the ward's method and Euclidean distance, revealed three main groups. The first (C1) and the second (C2) groups consisted of 26 and 60 genotypes, respectively. The remaining 135 genotypes were classified in the third group (C3) with two sub-clusters i.e., C3a and C3b, consisting of 10 and 125 individuals, respectively ( Figure 6). According to k-mean partitions comparison analysis based on Calinski criterion, K = 3 was selected as the best k ( Figure 5A), and 221 genotypes were divided into three main groups ( Figure 5B). On the other hand, consistent with the results of K analysis, the cluster analysis of the assayed genotypes, using the ward's method and Euclidean distance, revealed three main groups. The first (C1) and the second (C2) groups consisted of 26 and 60 genotypes, respectively. The remaining 135 genotypes were classified in the third group (C3) with two sub-clusters i.e., C3a and C3b, consisting of 10 and 125 individuals, respectively (Figure 6). The relationship between traits and each group of genotypes was determined. Based on these results, fruit shape, fruit skin color, and dominant flowering habit had the most, and fruit crown shape had the least with the first group. Fruit ripening time, fruit skin thickness, fruit size, flower cup diameter, fruit mean weight, fruit flavor, tree mean yield, and number of fruits per tree showed the most association with the second group. The traits most related to the third group were fruit skin sensitivity to sunburn and burst, flower position, fruitful flower percentage, and fruit crown shape (Figure 7). The relationship between traits and each group of genotypes was determined. Based on these results, fruit shape, fruit skin color, and dominant flowering habit had the most, and fruit crown shape had the least with the first group. Fruit ripening time, fruit skin thickness, fruit size, flower cup diameter, fruit mean weight, fruit flavor, tree mean yield, and number of fruits per tree showed the most association with the second group. The traits most related to the third group were fruit skin sensitivity to sunburn and burst, flower position, fruitful flower percentage, and fruit crown shape (Figure 7).
The number of accessions for each class, the core count, and the entire count for all 25 studied morphological and pomological traits are presented in Figure 8. Statistical indices and Shannon's index evaluated the efficiency and variation of candidate core collection (Table 1). In order to evaluate the efficiency of candidate core collections, four traits including MD%, VD%, CR%, and VR% were calculated as 7.03, 42.68, 100, and 150.2%, respectively.
The number of accessions for each class, the core count, and the entire count for all 25 studied morphological and pomological traits are presented in Figure 8. Statistical indices and Shannon's index evaluated the efficiency and variation of candidate core collection (Table 1). In order to evaluate the efficiency of candidate core collections, four traits including MD%, VD%, CR%, and VR% were calculated as 7.03, 42.68, 100, and 150.2%, respectively.
Discussion
A core collection can be formed based on different molecular, morphological, and agronomic markers. The morphological data are usually applied extensively as they are
Discussion
A core collection can be formed based on different molecular, morphological, and agronomic markers. The morphological data are usually applied extensively as they are recorded comprehensively [37,38]. The PowerCore software selects genotypes with higher diversity, representing trait states and total coverage of marker alleles of the entire collection through a modified heuristic algorithm [39]. As the formed core collection must be validated, one of the ways to measure and evaluate the formed core collection is to compare its diversity with the original collection's diversity. Therefore, study of the entire collection in terms of morphological and agronomical diversity of existing genotypes and their relationships and grouping is inevitable.
There existed a positive relationship between length-to-width leaf ratio and flower cup diameter, dominant flowering habit, and flower formation site. This means that increased leaf area makes flowering more regular and on perennial branches with larger flowers. Accordingly, the superior genotype can be selected for breeding at the seedling stage through the calculation of leaf area before reaching the maturity and fruiting stage. Regarding the negative relationship of fruit skin sensitivity to sunburn and burst with length-to-width leaf ratio, it should be noted that leaves provide shade and prevent direct sunlight falling on fruits. Hence, growing cultivars with more leaf areas in places where the maximum mean temperature is high can reduce incidences of sunburn [16]. As the lengthto-width leaf ratio increases, the leaf area and consequently, the transpiration rate increase. Leaf transpiration rate is known to affect calcium and boron absorption and translocation through fruit peel, which is related to diversity of fruit cracking among cultivars [16,40]. In this regard, Rodriguez et al. [41] reported that anti-transpiration treatments decreased stomatal conductance and resulted in more sunburned fruit in treated trees.
A positive relationship between fruit skin sensitivity to burst and sunburn should be due to the fact that free radicals formed by sunburn destroy the integrity of cell fats, proteins and membrane, which results in cracking [42]. A similar result in pomegranate has been reported by Shakeri and Sadat Akhavi [43]. The relationship between flower position and other traits showed that terminal flowering genotypes have a more regular flowering habit with a higher fruitful flower percentage, and the formed fruit will be more symmetrical and spherical with shorter crowns. In pomegranate, lateral flowers have been reported to have fewer and no developed ovules that are less fertile [44]. Similar results have been achieved in this research, as flowers with large cup diameter were accompanied with high tree productivity.
Examining the relationship between different morphological and pomological evaluated traits through factor analysis, the redundancy among variables was reduced using a smaller number of factors. The first factor, namely "fruit appearance", consisted of traits related to fruit shape such as FSH, FSC, and FCS. In the same vein, FP and FFP were dominant traits in the second factor, namely "flower attributes". The "physiological disorder" was the third factor in which FSSS, FSSB, and FST were representatives. These attributes were more effective in distinguishing and identifying the studied population. Besides, they could be useful for selecting superior genotypes or new cultivars from an economic point of view. In accordance with these results, it has been reported that fruit-related traits are more important in analyzing and differentiating breeding materials, and dealing with the phenotypic characterization of cultivated pomegranates [45,46].
Based on the scatter plot obtained using the first and second factors, genotypes were divided into three groups without any special geographical relationships. The uniform distribution in an area, independent from the geographical origin, has been reported in the grouping of pomegranate genotypes of China [47], Iran [12], Tunisia [48], and India [49] based on genetic and phenotypic data. Accordingly, there was no specific relationship between samples' placement into the groups and provincial divisions in the cluster analysis approach. One of the reasons for this dispersion should be the lack of knowledge on the exact origin of the genotypes, and each may have been given different names in different geographical areas by indigenous people as in the cases of Lebanese [14] and Turkish pomegranate genotypes [13]. Mislabeling was also reported for pomegranate accessions growing in Tunisia [50]. They demonstrated that the genetic distance between genotypes is rarely correlated with their geographical origins, and the same planting material, depending on the cultivation area, may have different denominations. Two important seedless genotypes, namely 11-223 and 5-220, and several wild genotypes were classified together in the C3a cluster. Most C1 members were commercial genotypes with higher fruit quality attributes and more tolerance to cold stress [51].
Fruit shape, fruit skin color, and dominant flowering habit were most associated with this group. The C2 genotypes were often cold-sensitive, and most of their fruits are used for processing purposes. Productivity properties, such as fruit size and mean weight, number of fruits in tree, tree mean yield and also flower cup diameter, fruit ripening time, and fruit skin thickness were the traits associated with this group. Fruitful flower percentage, flower position, fruit crown shape, and skin sensitivity to sunburn and burst were mostly associated with C3.
After the establishment of a core collection, an important issue is the extent to which it meets its original objectives in terms of the lack of repetition and diversity representation. In this concept, comparison of the core collection with its original one is usually used for validation purposes. Examination of the formed core collection revealed that the CR value is 100%. This index indicates whether the distribution ranges of each variable in the core set are well represented in comparison to the entire collection, and it should be no less than 80% order to represent the whole accessions by core collections [31,51]. MD%, VD%, and VR% are used to measure the statistical consistency between the core and entire collections [31]. If MD% between the initial collection and the core collection is less than 20% of the traits, the VD% within a core collection will be high, indicating the formation of a good core collection [52].
In addition, an ideal core collection encompasses the maximum diversity of the entire germplasm with minimum repetitiveness. Based on these explanations, MD% and VD% of this study were 7.03 and 42.68% respectively, which means a good diversity amongst genotypes within the core collection. A comparison of different values existing in the core collections and the entire collection is calculated by VR%, which determines how well it is being represented in the core sets [31]. The average of this value was 150.2% in the current study, which means that the core collection selected by Power Core is similar to the entire collection.
Calculation of the H index in the formed core collection showed that of 25 phenotypic variables, 19 traits had a high diversity with an H value of more than 0.67. Four traits, including tree crown shape, flower formation site, fruit ripening time, and aril size had a medium diversity, and fruit skin sensitivity to sunburn and seed hardness had a low diversity. Hence, it should be said that the formed core collection has a good quality with a variety of high attributes.
In this study, the core collection formed with 12 genotypes included 5.4% of the total genotypes of the original collection. In coordination with current results, Yanfang et al. [53] evaluated a collection of 560 mulberry accessions with 40 morphological features in China, declaring that 5% is the best ratio for an ideal core collection in this tree species. However, in the formation of a core collection of olive [54], apricot [20], and apple [55] using molecular markers, collections with a rate of 10-19, 8, and 12.4% were formed respectively, which are higher than that of pomegranate. One reason for these differences is the use of molecular markers that are better able to demonstrate differences among the promising genotypes, resulting in a larger core collection. Also, in a study on 104 Iranian walnut genotypes, a core collection of 27 genotypes, i.e., 26% of the original collection, was formed based on phenotypic diversity [29]. The reason for the higher ratio of walnut core collection to primary collection compared to the results of this study on pomegranate may be due to their different propagation methods. The vegetative propagation of pomegranate trees resulted in less diversity in comparison to walnut trees that are often propagated by seed [56].
Production or improvement of cultivars with seedless fruits is one of the main objectives of pomegranate breeding programs. Taking into account that there were only two seedless individuals in the original collection, one of them, i.e., "Bihasteh-Sangan", was included in the constructed core collection. This, in turn, confirm the good quality of the established core collection for breeding objects.
Conclusions
In this study, as a part of an ongoing project for pomegranate breeding program in Iran, 221 genotypes were clustered into three main groups based on 25 morphological and pomological attributes. Among all of the traits analyzed, those related to flower and fruit had the highest power of discrimination. They are, therefore, the most useful attributes for genetic characterization studies in pomegranate germplasm. The applicable core collection, with 12 representative superior genotypes derived from a whole primary collection, could well represent the Iranian pomegranate variation. These superior genotypes should be used for both production and breeding purposes. To the best of our knowledge, the present study was the first to use phenotypic data for developing a core collection in the pomegranate population. The established core collection will be an efficient step toward exploring, characterizing, and capturing the genetic diversity of large original populations. Also, the conservation of this core collection with only 12 superior genotypes should be more economical and manageable for breeding and production objects. This core collection, in spite of being reduced in size, will provide access to variation quite at the same variation level of initial collection and can be effectively used to improve pomegranate production in Iran and future breeding programs. In order to better evaluate the selected core collection, a complementary characterization of this core collection using molecular markers is also underway.
Supplementary Materials: The following are available online at [URL]/10.3 390/horticulturae7100350/s1, Table S1: Description of the evaluated Iranian pomegranate genotypes consisting of their name and identification code, Table S2: Description of the various phenotypic traits and their frequency in the entire and core collection of the evaluated 221 pomegranate genotypes.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-10-18T17:36:57.073Z", "created": "2021-10-01T00:00:00.000", "id": "244213462", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/2311-7524/7/10/350/pdf?version=1633690247", "pdf_hash": "f33a58295bc505b8ca3ea53201e0f8d6a161c424", "pdf_src": "Adhoc", "provenance": "peS2o-0075.json.gz:37257", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "33e67fa9eef3e67202d0ee76eb1a856ad5a554f7", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Application of sumac and dried whey in female broiler feed
The present study aimed at investigating the effects of sumac and dried whey powder on growth performance, carcass traits, intestinal morphology, microbial population, and some biochemical parameters and antibody titer against Newcastle disease of female broiler chicks. A total of 360 1-day-old female broiler chicks with an average weight of 38± 0.42 g were randomly divided into three treatments. Each treatment was further divided into six replicates. Three treatments were used: chicks were fed by basal diet as control group, basal diet + 0.02 % sumac powder (S), and basal diet + 0.02 % dried whey powder (DW) for 42 days. Results showed that feed intake of chicks increased significantly in S and DW in comparison with the control group (P < 0.05). Body weight gain was also significantly higher in the treated groups. The serum concentration triglyceride and cholesterol of chicks decreased significantly by S and DW feeding. While low-density lipoprotein (LDL) level decreased significantly, high-density lipoprotein (HDL) levels increased in the S group. Antibody level increased titer against Newcastle disease significantly by feeding treated diet compared to the control group. Data from microbial population investigation showed that E. coli population decreased while Lactobacillus increased in S and DW groups. The data revealed an improvement in the body weight gain, feed conversion ratio because of increased intestinal morphology, antibody level, and some useful microbial population in female broiler chicks receiving the sumac and dried whey powder.
Introduction
Sumac (Rhus coriaria L) is a plant species in the anacardiaceous family that is used as a spice and herbal medicine. Sumac is found in hot, temperate, and tropical regions worldwide (Kurucu et al., 1993). It has a long history of use by indigenous people for medicinal and other applications (Rayne and Mazza, 2007). The fruits of sumac contain flavonols, phenolic acids, hydrolysable tannins, anthocyanins, and organic acids such as malic, citric and tartaric acids (Ozcan and Haciseferogullari, 2004;Greathead, 2003;Jung, 1998). Sumac is used as an herbal remedy in traditional medicine because of its assumed analgesic, antidiarrheal, antiseptic, anorectic, and antihyperglycemic properties (Rayne and Mazza, 2007). Probiotics have been defined as a live microbial feed supplement that can beneficially affect the intestinal microbial balance, resulting in improved body weight gain and reduce mortality in broiler chickens. Prebiotics are defined as nondigestible, but fermentable food ingredients that beneficially affect the host by selectively stimulating the growth and activity of one or a limited number of species of probiotics bacteria in the colon (Panda et al., 2006). Symbiosis is defined as a combination of a probiotics and a prebiotics, aimed to increase the survival activity of probiotics in vivo and stimulating indigenous bifido bacteria and Lactobacillus (Dughera et al., 2007). Whey, a liquid remaining from cheese or casein production, is one of the most valuable protein sources in the human food chain. In spite of its balanced nutrients, liquid whey is disposed as a waste product and dried whey that is produced from its liquid form can be used in bird diets (Aghaei et al., 2010). It is shown that dietary supplementation of whey powder linearly increases body weight gain and nitrogen retention in turkey and broiler Published by Copernicus Publications on behalf of the Leibniz Institute for Farm Animal Biology.
chickens (Shariatmadari and Forbes, 2005). The purpose of this study was to investigate the response of broiler chicken consumption of dried whey powder as symbiotic and sumac powder as prebiotic on performance, intestinal morphology, and microbial population of Escherichia coli and Lactobacillus in the gastrointestinal tract.
Birds and diets
All procedures used in this experiment were approved by the Department of Animal Science, Faculty of Agriculture, Shahrekord Branch, Islamic Azad University, Shahrekord, Iran. A total of 360 1-day-old Ross 308 commercial female broiler chickens with an average weight of 38 ± 0.42 g were randomly distributed into three dietary treatments with six replicates each. The birds were housed in groups of 3 in 18 pens under standard conditions of temperature, humidity, and ventilation. The study was carried out in the poultry farm of Animal Science College, Shahrekord Branch, Islamic Azad University, Shahrekord, Iran, for a period of 6 weeks. The sumac was purchased from the local market, and the whey was purchased from PAK dairy company. It was ground to a fine powder and then mixed with the basal diet. The basal diet (control group) was formulated according to requirements recommended by the Ross manual (Ross, 2007) on the basis of corn, wheat, soybean meal, and soybean oil (Table 1). The three experimental treatments were the following: control group (basal diet), sumac group (basal diet + 0.02 % of sumac powder (S)), and dry whey (basal diet + 0.02 % of dried whey powder (DW)). In the basal diet, sand was used as neutral substance instead of sumac and dried whey powder. The chemical compositions of sumac and dried whey are shown in Table 2. Feed and fresh water were provided ad libitum during this experiment. Body weight gain, feed intake, and feed conversion rate (FCR) were calculated for the experimental period. A lighting programme of 23L : 1D was used for the trial period.
Data collection
On day 42, all birds were processed after a 4 h feed withdrawal period; two birds per pen were randomly selected and killed in a commercial slaughter house. Weight of carcass, breast, and thigh were recorded on 12 birds per treatment (n = 36). Carcass yield was calculated as eviscerated carcass with neck, feet, and abdominal fat pad removed, as percentage of live body weight at the time of feed withdrawal.
Evaluation of blood parameters
The blood samples were taken from the brachial vein from two birds per replicate and stored at refrigerator at + 4 • C. Also the blood serum samples were subjected to biochemical analysis for cholesterol, triglycerides, high-density lipoprotein (HDL) and low-density lipoprotein (LDL) by Pars Azmoon commercial kits (Ellefson and Garaway, 1967).
Detection of antibody titer to ND
Using hemagglutination inhibition test (HI) according to Beard (1989) with chicken red blood cells and four units of NDV (Newcastle disease vaccine) antigen, then geometric mean titers were calculated.
Microbial count
The internal organs were removed after slaughter. About 7 cm from the length of the ileum was sampled to determine the microbial population. Also 1 g of ileum content was used to make 10-fold dilution using buffered peptone water and then 0.1 mL of the appropriate ileum dilution was spread on Lactobacillus MRS1 Agar-Hi Media Laboratories to detect lactic acid bacteria and VRBA2 (Violet Red Bile Agar) to detect E. coli form. The cultures of Lactobacillus and E. coli bacteria were made an aerobically form. The plates were incubated at 37.5 • C for 48 h. After counting the number of colonies in each plate, the number so obtained was multiplied by inverse of the dilution and the result was stated as the number of colony forming unit (cfu) in 1 g of the sample (Downes and Ito, 2001).
Histomorphometric examination of small intestine
The histomorphometric examination was performed by light microscopy, and the measurement was done using public domain image analysis software (Image J, National Institute of Mental Health, Bethesda, MD, USA) (Rezaian, 2006).
Statistical analysis
Data analysis was performed by using the general linear model procedure and the comparison of means was made through Duncan's multiple range test by using SAS 9.1 software (SAS, 2004).
Results
Data from this study showed that feeding of S and DW diets significantly increased body weight gain and reduced feed intake of chicks (P < 0.05) compared to the control diet. The feed conversion ratio of female broiler chicks significantly (P < 0.05) improved by feeding sumac and whey powder (Table 3). The triglyceride, cholesterol, and LDL level decreased significantly in the treated groups compared to control group while S group had the lowest level of triglyceride, cholesterol, and LDL blood plasma. The antibody titer against Newcastle vaccine disease in control and treated groups is given in Table 4. Antibody titer was significantly higher (P < 0.05) when female broiler chicks were fed with S and DW diets. It was determined that sumac and dried whey had significant (P < 0.05) effects on antibody titer against Newcastle disease vaccine compared to control group (Table 4).
It was showed that sumac and dried whey had significant (P < 0.05) effects on intestinal microbial population compared to control group (Table 5). Data showed that E. coli population decreased while Lactobacillus population increased significantly in DW and S compared with the control group, respectively (P < 0.05). The effects of sumac and whey powder on some carcass traits of female broiler chicks are given in Table 6. The results showed that using sumac and whey powder already reduced abdominal fat and bursa of Fabricius percentage. These treatments also significantly increased liver and spleen percentage (P < 0.05).
The effect of sumac and dried why powder supplementation on the histology of intestines is given in Table 7.
As results reveal, we demonstrated that mucosa, muscularis, serosa, and total length increased significantly in S and DW compared to control group. Also female broiler chicks fed with dried whey powder had thicker mucosa, muscularis, serosa, and total than other groups (P < 0.05).
Discussion
Recent research and development of symbiotic products have been increasingly focused on functional benefits including resistance to gastrointestinal bacterial infection and improved immune status in broiler chicks. The consumption of a probiotic in combination with a suitable prebiotic (symbiotic) can result in synergistic effects (Zanoni et al., 2008). Hosseini Mansoub (2012) showed that using different levels of sumac had significant effects (P < 0.05) on feed intake, weight gain, body weight, and feed conversion ratio of broiler chickens. According to Ghasemi et al. (2014) the improvements in body weight gain and feed conversion ratio are due to the active materials (cinnamaldehyde and eugenol) found in sumac, causing greater efficiency in the utilization of feed, resulting in enhanced growth. Hosseini Mansoub (2012) found that antimicrobial substances present in sumac and cinnamon can reduce the harmful bacteria pop-Table 4. The effect of sumac and whey powder diets on some blood biochemical parameters and antibody titer against Newcastle vaccine disease of female broiler chick.
Treatments Triglyceride (mg dl −1 ) Cholesterol (mg dl −1 ) HDL (mg dl −1 ) LDL (mg dl ulations in the gastrointestinal tract and improve the levels of absorbed amino acids. Aghaei et al. (2010) showed that using dried whey in laying hen diet had significant effects (P < 0.05) on feed conversion ratio, and 5 % dried whey diet had least feed conversion ratio while adding probiotic did not affect feed conversion ratio in laying hens. Sumac, which has hypocholesterolemic effects that lead to d-limonene (l-methyl-4-,1-methyletheny l-cyclohexane), is a monocyclic monoterpene component. The hypocholesterolemic action of sumac is possibly related to its polyphenolic components. Poly phenols have been shown to depress the reverse-cholesterol transport, reduce the intestinal cholesterol absorption and even increase bile acid excretion (Jung, 1998). Also free gallbladder acids attach to bacteria and fibers and this can increase the excretion of them. This is consistent with the well-observed effect of herbs on lowering blood cholesterol level. Sumac extracts have been found to have antimicrobial, hypoglycemic, and antioxidant activities (Hosseini Mansoub, 2012). These results on biochemical blood parameters were not in accordance with Golzadeh et al. (2012), who declared that there was no significant difference between the treatments for plasma triglyceride, HDL, and LDL concentrations. However, they were in similar to the results of Hosseini Mansoub (2012), who showed that the serum total cholesterol, triglycerides, and LDL concentration were significantly reduced and the concentration of serum HDL was significantly induced in treatment groups compared to the control group (P < 0.05). Concerning humeral immune response, high doses of symbiotic improve the HI antibody titers for NDV compared to that of the control group. These results are in agreement with Ashgan and Samah (2011), who showed a high concentration of symbiotic improves the antibody response to NDV and infectious bronchitis virus vaccines. Researchers suggested that sumac and probiotic can control microbial growth by acting on the microflora's biochemical processes such as protein synthesis or inhibiting the elongation of Methanobacterium and Escherichia coli, or by reducing lactic acid producing bacteria in the gastrointestinal tract (Guvenc and Koyuncu, 1994). Jung (1998) indicated that sumac is effective against both gram-positive and gram-negative bacteria, while it is more effective on gram-positive than gram-negative microorganisms. Fairchild et al. (2001) and Spring et al. (2000) reported that the use of probiotics in the ration reduces the total population of E. coli forms in the intestinal lumen. Rada et al. (1999) found that the use of Lactobacillus salivary in the chicken's diet can reduce the coliform bacteria population significantly as compared with the control treatment by reducing the intestinal pH level. Chung and Day (2004) showed that lactose in dried whey cannot be digested or absorbed so fermented by lactic acid bacteria, production of lactic acid and reduction in the pH and multiplication of pathogenic bacteria. Despite of this, Yang et al. (2007) observed no change in the gastrointestinal microbial flora by adding probiotic similar to dried whey diet. The probiotic microorganisms prevent the growth of pathogenic bacteria in the intestinal environment by production of acids, such as acetic acid and lactic acid and other components (Fuller, 1989). Zavaragh (2011) found that there is a possibility that gathering these antimicrobial herbs caused a remarkable decrease in the amount of intestine microbial colony. In this study sumac and whey powder reduced abdominal fat and bursa of Fabricius percentage similar to Hosseini Mansoub (2012), who showed that use of sumac had decreased abdominal fat and increased gizzard and liver weight. Langhout (2000) showed that herbal plants could stimulate the digestion system in poultry, improve the function of the liver, and increase the pancreatic digestive enzymes. Enhancement of the metabolism of herbal plants, carbohydrates and proteins in the major organs would increase growth rate of these organs. Majewska et al. (2009) showed that use of whey increased liver weight percentage and decreased heart, abdominal fat, and gizzard weight percentage in broiler chickens. The digestive tract is influenced by some herbal plant, which has an exact effect on intestinal activity and structural of intestinal villi in broiler chicks (Rafiee et al., 2014). Samanya and Yamauchi (2002) demonstrated that by increasing villus height and absorptive surface area absorptive function, digestive performance will increase. In fact some growth promoters in herbs can modify brush border enzymes and the nutrient transport systems. Miles et al. (2006) showed that the reduction of the inflammatory reactions at the intestinal mucosa can increase the villus height and secretion of enzymes, digestion performance, and absorption of nutrients by the them. Ghasemi et al. (2014) demonstrated that the use of sumac extract affects intestinal characteristics such as villus height, crypt depth, villus width, epithelium layer, and goblet cells. Szczurek et al. (2013) showed that the overall effect of feeding duration with the whey containing diets was seen only for breast meat yield and relative spleen weight. Their value increased (P < 0.05) as feeding duration increased from 1-7 to 1-42 days. The results of current study are in agreement with Ghasemi et al. (2014), Rafiee et al. (2014), and Samanya and Yamauchi (2002). It can be hypothesized that, in the current experiment, increased integrity of the intestinal tract associated with a greater villi height and surface area after whey supplementation resulted in improved live performance and carcass traits of broiler chickens.
Conclusions
It is concluded that dietary supplementation of sumac (0.02 %) and dried whey powder (0.02 %) can reduce triglyceride, cholesterol, and LDL in plasma of female broiler chicks at 42 days of age. According to our results, the use of dried whey and sumac can reduce the intestinal pH and provide a good environment for Lactobacillus growth and limited multiplication of coliform bacteria. It can be suggested that increased integrity of the intestinal tract associated with a greater mucosa, muscle layer, serosa, and total length of intestine after whey and sumac supplementation resulted in improved performance and carcass traits of female broiler chicks.
Table 1 .
Composition (measured in %) of the experimental diets for broiler chicks.
Table 2 .
Chemical composition of the sumac and dried whey (% dry matter).
Table 3 .
The effect of sumac and whey powder on growth performance of female broiler chicks.
1 Standard error of the means. Column values with same superscript or no superscript are not significantly different (P > 0.05). Control group = basal diet, S = basal diet + 0.02 % of sumac powder and DW = basal diet + 0.02 % of dried whey powder.
Table 5 .
The effect of sumac and whey powder diets on intestinal microbial population of female broiler chicks.
Table 6 .
The effect of sumac and whey powder diets on carcass traits (%) of female broiler chicks.
Table 7 .
The effect of sumac and whey powder diets on intestinal characteristics of female broiler chicks.
* Total = mucosa + muscularis + serosa; column values with same superscript or no superscript are not significantly different (P > 0.05).
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-06T23:29:45.799Z", "created": "2015-05-21T00:00:00.000", "id": "55484197", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://aab.copernicus.org/articles/58/205/2015/aab-58-205-2015.pdf", "pdf_hash": "7ce7ea11fd93ecf15715b447a36af5a00df8893d", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:34723", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "7ce7ea11fd93ecf15715b447a36af5a00df8893d", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Paclobutrazol as growth regulator in Bahiagrass
Bahiagrass (Paspalum notatum) is a native species widely found on highways margins, however, cutting practices are necessary, which increase the maintenance cost, being an alternative, the use of growth regulators, such as Paclobutrazol. Thus, the objective of the present study was to evaluate the use of Paclobutrazol doses as growth regulator in Bahiagrass. The turf was implanted in black plastic containers (8.46 L), previously prepared with the mixture of soil + sand + organic compost (1:1:1). The experimental design was completely randomized, composed of 3 doses of Paclobutrazol [1.1 and 2.2 kg a.i. (active ingredient) ha-1 + control (0 kg a.i. ha-1)], in four applications of 2 L of syrup per container, thus forming a factorial 3x4 (doses x response time) with 3 repetitions. The fresh and dry mass, relative chlorophyll index, leaf N and the visual aspect of the green cover rate were evaluated on four different dates. It was observed that the product controlled the growth of the lawn, with less production of fresh and dry mass, and increase of the relative index of chlorophyll and concentration of leaf nitrogen; however, it decreased turfgrass density of green mass, which influenced the visual lawn aspect. Thus, Paclobutrazol controlled Bahiagrass growth at a dose of 2.2 Kg a.i. ha-1 in regular applications of 30 to 45 days, however it affected the density and consequently the aesthetics of the turfgrass.
Introduction
Bahiagrass (Paspalum notatum) is a grass species native of South America, including Brazil, which highlights for its aesthetics, functionality and rusticity (Amaral and Castilho, 2012;Steiner et al., 2017;Souza et al., 2020a). With leaves concentrated in the basal part of plant, it easily covers the ground, being used in green areas and embankments along road margins, which protects soil against erosion, as its surface rhizomes and roots intertwine, managing to retain the soil (Marchi et al., 2016;Antoniolli, 2019;Souza et al., 2020a). It is resistant to water deficit and well adapted to soils with low fertility. However it requires frequent maintenance in relation to pruning in periods of rain and intense heat, and therefore, there may be an increase in cost of constant management (Amaral and Castilho, 2012;Marchi et al., 2016;Santos and Castilho, 2018).
However, most lawns formed by this grass are the result of planting by "pieces" collected manually in degraded pasture areas because the commercial availability of this species is scarce (Souza et al., 2020a). Thus, there is a need to search for information about it for use on highways, urban avenues, or airport runways (Souza et al., 2020b). In recent years, research on Bahiagrass has gained prominence to be used for this purpose (Souza et al., 2016;Silva et al., 2018;Maximino et al., 2017;Souza et al., 2020b) and recently being launched commercial cultivars of Paspalum notatum registered by EMBRAPA (Brazilian Agricultural Research Corporation) (Souza et al., 2020a).
According to Souza et al. (2020b) it is essential that Paspalum species, when used as vegetation cover, have slower vertical growth, to reduce pruning need. Marchi et al. (2016) states that 5 to 7 mowing operations are necessary for Bahiagrass during its main growing season (summer), and Chaer (2015) mentions that in vegetation areas such as highways, it is estimated that the concessionaires performed up to 15 annual pruning. Thus, an alternative to reduce cutting operations would be the use of growth regulators, which has been shown to be an option for mechanical handling (March et al., 2013).
Plant regulators are gaining importance in lawn management programs, as they suppress vegetative growth without impairing their visual quality and without promoting phytointoxication, discoloration or leaves thinning (McElroy, 2012;March et al., 2013;Gazola et al., 2019). However, there are no registered products for lawns in Brazil for this purpose (Dias et al., 2019;Gazola et al., 2019). An alternative would be the use of Paclobutrazol, which is a regulator already studied and registered for other crops such as mango, tomato (Syngenta, 2019), and registered for turfgrass in the European Union (Semillas Fitó, 2010) and in the USA (McElroy, 2012).
Paclobutrazol acts on the synthesis route of gibberellins, and interferes in phase 2, that is, it blocks the action of the enzyme Kaurene Oxidase, which inhibits the conversion of kurene into kaurenol and, consequently, prevents the formation of any type of gibberellin as well as cellular elongation, creating compact leaf (McElroy, 2012;March et al., 2013;Glab et al., 2020). When applied directly to the soil or substrate, it has greater activity and efficiency compared to foliar application, as it is translocated almost exclusively by xylem and less by phloem, where it is absorbed by the roots and moved directly to the growth points (McCarty, 2008;Blank et al., 2009).
Studies on Brazilian conditions for growth control in Bahiagrass have been published in recent years (Marchi et al., 2016;Marchi et al., 2017;Barbosa et al., 2017;Dias et al., 2019), however none of them use Paclobutrazol as a growth regulator. In this context, the objective of the present study was to evaluate the use of Paclobutrazol doses as growth regulators in Bahiagrass.
Material and Methods
The experiment was conducted in the Northwest region of São Paulo state, in full sun, from September 2019 to February 2020 (average temperature of 28 ºC, average relative humidity of 72.8%). For planting the lawn, the Bahiagrass came from the experimental area of the University where the experiment was carried out, and it was harvested in the form of «plates». And was implanted in black plastic containers (47.5 x 17.5 cm top dimensions, 41.5 x 11.3 cm bottom dimensions, height of 15.5 cm, volume 8.46 L), previously prepared with a mixture of soil + sand + organic compost (1:1:1); height standardization cut was also performed, and the green mass obtained was discarded. After implantation, fertilization was performed with formulated 13-5-13, using 20 g diluted in 2 L of water per container. Weed control was performed whenever necessary manually, and irrigation was daily with 2 L of water per container, done in late afternoon, using a watering can.
Twenty-two days after planting, the installation of the experiment began, being applied 3 doses of Paclobutrazol (in 4 regular applications ranging from 22 to 28 days), and evaluated the product response time on 4 different dates, thus forming a factorial 3x4 (doses x response time) with 3 repetitions. For doses, Paclobutrazol was diluted in two concentrations being 1.1 and 2.2 kg a.i. (active ingredient) ha -1 + control (0 kg a.i. ha -1 ), applied 2 L of syrup per container, directly on the substrate, as product is absorbed by roots (McCarty, 2008;Blank et al., 2009). The Paclobutrazol application dates were 0, 28, 55 and 78 days after the installation of the experiment (DAIE). All applications being performed in the morning presented data of relative humidity and temperature of 23.5 ºC, 61.7%; 31.0 ºC, 55.5%; 28.1 ºC, 70.7%; 28.0 ºC, 78.5%, respectively on each application date.
For the response time of Paclobutrazol on the lawn, evaluation was carried out 28, 55, 78 and 132 DAIE, being they: fresh and dry leaf mass; relative chlorophyll index (RCI); green coverage rate and leaf nitrogen content. For fresh and dry leaf mass, the material was collected manually, using scissors to remove all leaves from treatments. Thus, after mowing, turfgrass clippings were stored in a Kraft paper bag and weighed to obtain fresh mass. After that, they were allocated in a forced circulation oven, 60°C for 72 hours and again weighed to obtain the dry mass. The RCI was realized with the manual chlorophyll meter (atLEAF), being collected in 3 points per experimental plot, measured Bruno HorscHut de Lima et al. 415 in the middle third of leaves. For green coverage rate, the treatments were photographed, with a 12 Mp camera, at a distance of 1 m; these images were analysed using the Canapeo TM software to obtain green coverage rate of the turfgrass. It was also evaluated the leaf nitrogen content and was done by using methodology described by Malavolta et al. (1997).
The collected data were subjected to analysis of variance and, subsequently, the average were compared by the LSD test at 5% of significance using SISVAR program (Ferreira, 2019).
Results and discussion
Results demonstrate for fresh mass that there was statistically significant difference between Paclobutrazol applied doses, and it was observed that the highest concentration (2.2 kg a.i. ha -1 ) managed to reduce growth at 28 and 55 DAIE, differing from control. At 78 and 132 DAIE, doses of the product differed from the treatment without application, inferring that there is greater effect of Paclobutrazol, when applied constantly at regular intervals. The fact of this reduction has occurred, is probably by execute mechanism of Paclobutrazol action, which acts in phase 2 of gibberellin synthesis root. Active ingredient blocks action performs of Kaurene Oxidase enzyme preventing formation of any gibberellin type (March et al., 2013;Glab et al., 2020), and thus, formation of the most compact leaf and, consequently, the lowest lawn growth (McElroy, 2012).
In absolute values, it is noted that for all evaluated dates (Figure 1), the greatest reduction was at treatment 2.2 kg a.i. ha -1 , where the decrease in growth was 25.35%; 33.90%; 53.67% and 62.79% at 28, 55, 78 and 132 DAIE, respectively. The drop in mass production increases over time, when the application of product is constant. Mass reduction was also reported up to 91% in 'Patriot' Bermuda grass (Cynodon dactylon x C. transvaalensis) after 4 weeks after Pacloburazol application (Volterrani et al., 2015), and 86% in 'TifEagle' (C. dactylon x C. transvaalensis) (McCullough et al., 2004) after two applications, values higher than those of the present study.
In the same way, as there was fresh mass reduction after Paclobutrazol application (Figure 1), there was also decreased in dry mass (Figure 2), where 2.2 kg a.i. ha -1 showed less growth, following the results of fresh mass. According to Santos and Castilho (2018), lower dry mass production decreases the need for mowing, which is recommended for species such as Bahiagrass. These results reported, therefore, the action of Paclobutrazol in the species as the product is indicated in the USA and the European Union for some turfgrass of warm climate such as Bermuda grass (Cynodon spp), Emerald grass (Zoysia japonica) and Saint Augustine grass (Stenotaphrum secundatum) (McCarty, 2008;Semillas Fitó, 2010). On the other hand, other growth regulators such as Trinexapac-ethyl in different doses had no effect on reducing dry mass for studied species (Marchi et al., 2017), as well as Prohexadione-calcium in split applications (Marchi et al ., 2016). When a product is used as a growth regulator, it is essential that it does not decrease leaf chlorophyll concentration (Gazola et al., 2016), as this will affect the turfgrass physiological processes, as well as green colour characteristic (Santos et al., 2019), and the relative chlorophyll index (RCI) results demonstrate this fact (Figure 3). It was observed that at 28 and 132 DAIE were the only dates where the RCI showed statistical difference between treatments, while at 55 and 78 DAIE doses did not differ between them. Thus, these results show that Paclobutrazol did not affect RCI and even in some cases, such as at 28 and 132 DAIE it provided chlorophyll increase.
According to Brito et al. (2016), Paclobutrazol causes chlorophylls concentration in a smaller volume in cells, due to product preventing cell elongation, and consequent leaf development, which was observed visually, and is also inferred at fresh mass (Figure 1). In addition, it stimulates endogenous cytokines biosynthesis, maximizing chloroplast differentiation, chlorophyll biosynthesis and delaying its degradation (D'Arêde et al., 2017). The results of the present study were close to those found by Amaral and Castilho (2012) in Bahiagrass from 39.67 to 49.60 atLEAF, when converting the values to the same unit.
The results of the present study still demonstrate that leaf N increased with the Paclobutrazol application ( Figure 4), thus indicating a small relationship between RCI and leaf N, as also reported by Santos and Castilho (2015) and Oliveira et al. (2018) in Emerald grass and by Santos et al. (2019) in Bermuda grass. This is because chlorophylls are magnesian porphyrins composed of a central atom of Mg linked to another four atoms of N (Taiz et al., 2017). In addition, the leaf N values for the dose 2.2 kg a.i. ha -1 differed to all evaluated dates from control, indicating that the product did not interfere with plant nutritional status and increased N concentration. For both applied Paclobutrazol doses, the values were higher than those found by Amaral and Castilho (2012) in an experiment with fertilization of Bahiagrass (9.7 to 14.12 g kg -1 ). For Bahiagrass green coverage rate after paclobutrazol application, it was observed that at 2.2 kg a.i. ha -1 showed lower density ( Figure 5). This fact is because Bahiagrass has only rhizomatous growth, presenting superficial rhizomes (Souza et al., 2020a), and thus does not form an extremely dense lawn, where Paclobutrazol application exposed the rhizomes decreasing grass leaves density.
At 78 and 132 DAIE it was observed that the green coverage rates were low, being less than 50%, showing that after each cut, the surface rhizomes are increasingly exposed, and when using Paclobutrazol, this exposure is more visible, and the visual aspect highlights this fact ( Figure 6). The visual quality of species is essential for product acceptance as growth regulator , especially on highways margins, where the importance of green lawns has been emphasized, as they are pleasant to drivers' eyes (Affonso and Freitas, 2003). However, in the present study, Paclobutrazol application interfered in coverage rate, regardless of dose used, being higher in the dose after 78 DAIE ( Figure 5). As previously described, Bahiagrass only has rhizomatous growth, presenting superficial rhizomes, (Souza et al., 2020a), and thus does not form an extremely dense lawn such as Emerald grass and Bermuda grass, which has rhizomatous and stoloniferous growth (Santos et al., 2019;Gazola et al., 2019). Thus, by creating a more compact leaf, the application of Paclobutrazol ends up causing the turfgrass to leave its rhizomes exposed, which reduces the green density of the lawn. Dias et al. (2019) also observed reduction in green visual quality of Bahiagrass after application of glyphosate as growth regulator from the dose of 45 g a.e. (acid equivalent) ha -1 .
According to Gazola et al. (2016), when growth regulator is used on lawns, it is essential that it does not impair aesthetics, such as green colour characteristic and coverage rate (closed, flawless lawn). However, in the present study, from 78 DAIE, the great lack of green mass in all treatments is clear (Figure 6), as all leaves were cut for analysis and the lawn did not have time for full recovery, inferring that turfgrass mowing height might be high to not damage the aesthetics. Sampaio (2012) recommends not cutting more than 1/3 of the lawn during maintenance, and for Bahiagrass author suggests maintaining the leaf height of 3-6 cm, precisely to avoid this exposure of surface rhizomes.
Thus, Paclobutrazol can be a great option to be used as a regulator and growth in Bahiagrass, to reduce the need for mechanical mowing and consequently the cost of the operation. However, the use of the product must be associated with the correct management of the lawn, observing important issues, such as regular applications of the regulator and mowing at the ideal height so as not to interfere with the aesthetics of the turfgrass.
Conclusion
Paclobutrazol controlled Bahiagrass (Paspalum notatum) growth with dose of 2.2 kg a.i. ha -1 in regular applications of 30 to 45 days, however it affected the density and consequently the aesthetics of the turfgrass.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-10-28T18:34:21.938Z", "created": "2020-09-01T00:00:00.000", "id": "225189485", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.scielo.br/j/oh/a/mVfkMMsRwnThVmhmVMgZStP/?format=pdf&lang=en", "pdf_hash": "31851c4bf90ba7c498e5d68d24780af45a7701fc", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:34718", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "401bf8d5869c967b3e2af5f98828d1944e1f6f81", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effect of linseed oil sediment in the diet of pigs on the growth performance and fatty acid profile of meat
The objective of this investigation was to examine the influence of dietary linseed oil sediment on the growth performance and fatty acid composition in the muscle tissue of pigs. Sixty-eight crossbred Swedish Yorkshire × Norwegian Landrace pigs were allocated to two trials with two different levels of linseed oil sediment. Twenty-four pigs in Trial 1 were allotted into control 1 and experimental 1, of 12 animals each, and forty-four pigs in Trial 2 were allotted into control 2 and experimental 2, of 22 animals each. In both treatments, control and experimental groups were formed by animals analogous by origin, gender, weight, and condition score. Control pigs were fed identical diets ad libitum in both trials. The treated pigs were fed the same diet as control pigs, but vegetable oil was replaced by linseed oil sediment at a rate of 25 g kg−1 (experimental group 1) in Trial 1 and 50 g kg−1 (experimental group 2) in Trial 2. The results indicated that in both trials, vegetable oil replacement for linseed oil sediment had no significant influence on the growth rate of pigs, though a tendency was observed for a more rapid daily gain. Addition of linseed oil sediment to the diets increased the content of n-3 α-linolenic (C18:3n-3), eicosatrienoic (C20:3n-3), and eicosapentaenoic (C20:5n-3) acids and total n-3 polyunsaturated fatty acids (PUFA) and decreased the C18:2n-6/C18:3n-3 and n-6:n-3 ratios and the thrombogenic index of meat. Moreover, the addition of 50 g kg−1 linseed oil sediment resulted in higher content of docosapentaenoic (C22:5n-3) fatty acid, total PUFA, and PUFA:SFA ratio. Supplementation of pig diets with linseed oil sediment increases the content of α-linolenic, eicosatrienoic, eicosapentaenoic, and docosapentaenoic fatty acids and total content of n-3 polyunsaturated fatty acids and have a positive effect by improving the polyunsaturated fatty acids:saturated fatty acids and n-6:n-3 fatty acid ratios in meat.
Introduction
The cereal-based diet that is mostly used for pig feeding supplies a small amount of n-3 polyunsaturated fatty acids (PUFA). The aspiration to increase the n-3 PUFA content in pork requires a supply of n-3 PUFA from the diet (Nieto and Ros, 2012). Therefore, in the past few years, several studies have focused on the improvement of the nutritional value of pork. Pig feed has been oriented to a higher content of natural sources of PUFA to increase the tissue deposition of PUFA and to improve the health status of consumers (Boselli et al., 2008;Peiretti et al., 2015). Linseed is one of n-3 PUFA sources. Linseed contains from 18.9 to 27.0% protein and from 34.1 to 40.7% fat (Čolovic et al., 2016). Adding 5 or 10% of linseed in the diet of pigs increases concentrations of α-linolenic acid and decreases n-6:n-3 ratio (Matthews et al., 2000).
Linseed oil in the diet throughout the fattening period increased the growth rate of pigs during the finishing period (Więcek et al., 2010). It also increased the content of C18:3 in neutral lipids and phospholipids in the longissimus muscle and biceps brachii muscle (Lu et al., 2008). The diet enriched with 3% linseed oil produced the highest level of α-linolenic acid (Nguyen et al., 2003). However, as an ingredient in pig rations, linseed will only be used as a secondary linseed oil product -as a meal or a cake -if its use is economically justified. Corino et al. (2008) showed that inclusion of extruded linseed in the diet of pigs increased n-3 PUFA content in both longissimus dorsi muscle and backfat and decreased n-6:n-3 PUFA ratio. Inclusion of 13.4% ground linseed in the diet of pigs increased the content of linoleic and alpha-linolenic acids in the muscle tissue and backfat (Bečková and Václavková, 2010).
Storage of cold pressed oil in various containers produces sedimentation with a comparatively high content of fat, protein, and phospholipids. Obviously, linseed oil sediment has a lower content of α-linolenic fatty acid (C18:3n-3) compared with linseed or linseed oil, but at the same time also a lower content of linoleic fatty acid (C18:2n-6), harmful to health. However, there is no sufficient data on the use of this sediment in animal feeding. Therefore, the objective of this study was to investigate the effect of cold pressed linseed oil sediment in the diet of pigs on the growth performance and fatty acid composition in meat.
Material and Methods
This trial was carried out in accordance with the Directive 2010/63/EU of the European Parliament and the Council of 22 September, 2010 on the protection of animals used for scientific purposes.
Sixty-eight crossbred Swedish Yorkshire × Norwegian Landrace pigs were used in two dietary experiments. Twenty-four pigs of 16.50±0.47kg body weight in Trial 1 were allotted into control 1 and experimental 1 groups, of 12 animals each, and forty-four pigs of 14.84±0.45kg body weight in Trial 2 were likewise allotted into control 2 and experimental 2 groups, of 22 animals each. In both trials, control and experimental groups were formed by animals analogous by origin, gender, weight, and condition score. The pigs of control groups were fed an identical diet which was adapted according to the physiological needs of pigs at different ages during growing and finishing periods (Table 1). In the finishing period, the dietary composition was slightly changed by reducing the soybean meal content and increasing the amount of peas. The pigs of experimental groups were fed the same diet as control pigs, but vegetable oil was replaced by linseed oil sediment at a rate of 25 g kg −1 (experimental group 1) in Trial 1 and 50 g kg −1 (experimental group 2) in Trial 2 (Table 1).
The pigs in control and treated groups were fed ad libitum and raised in pens of 12.6 m 2 area with 11 or 12 pigs per pen. Feed intake was recorded by pen and average daily feed intake per pig per day and per kg gain was estimated by group of pigs. The finishing period ended when the pigs reached a final live weight of 110-120 kg.
The chemical composition and nutritive value of the feeds were analysed according to the standard methods (AOAC, 1990). The analysis of nutritional value of vegetable and linseed oil sediment used in the pig feed showed that linseed oil sediment contained 4,51 times more α-linolenic (C18:3n-3) and 3,61 times less linoleic (C18:2n-6), fatty acids in comparison with vegetable oil (Table 2).
Processing of the data was performed using software Statistica (Data Analysis Software System, Version 7.0; StatSoft, Inc., Tulsa, OK, USA). The individual pig served as the experimental unit for all growth and polyunsaturated fatty acid indicators of meat. The statistical evaluation of the results was performed using descriptive statistics and Student's t-test for independent samples. In the figures, all data are displayed as mean () and standard error (SE) of the mean. The probability level of P<0.05 was considered as statistically significant, whereas the differences of 0.05<P<0.10 were considered as tendency to difference.
Results
The results of our study showed that growth performance was not influenced by treatment. However, in experimental group 1, a tendency was observed for a more rapid growth of pigs as the daily gain was 73 g higher (P<0.10)than that of the pigs in control group 1 in Trial 1 (Table 3). The daily gain of the treated pigs in the growing (1.5-4 months of age) and finishing (over four months of age until slaughter) periods was higher by 60 and 87 g, respectively, than those of the control group, although no significant differences were found. In Trial 2, when a higher content (50 g kg −1 ) of linseed oil sediment was used, the daily gain of the treated pigs in experimental group 2 was 73 g higher (P<0.10)than that of control group 2 within the whole growing-finishing period. Meanwhile, in the growing period pigs of experimental group 2 gained 74 g (P<0.10)daily more than pigs of control group 2. Treated pigs consumed more feeds than control pigs (Table 4). Within the whole experimental period, feed intake per kg gain of pigs of experimental groups 1 and 2 was lower, respectively, by 80 and 190 g in comparison with pigs of control groups 1 and 2. Although in Trial 1 no differences were found between the groups for the duration of the fattening time, there was a significant effect on this indicator (10.1 day less; P<0.025) of treated pigs fed linseed oil sediment in Trial 2.
The results of our study showed a tendency for a higher content of linoleic (C18:2n-6) fatty acid (1.22%; P<0.10) in the intramuscular fat of pigs of experimental group 1, whereas in experimental group 2, there were tendencies observed for a higher content of docosadienoic (C22:2n-6) fatty acid (0.03%; P<0.10) and a lower content of docosatetraenoic (22:4n-6) fatty acid (0.03%; P<0.10) in comparison with the pigs of control groups 1 and 2. Furthermore, it was found that the muscles of pigs of experimental group 1 had higher total PUFA (2.63%; P<0.10) and total PUFA in the muscle tissue of pigs of experimental group 2 was 2.26% higher (P<0.05). The PUFA:SFA ratio was higher and n-6:n-3 fatty acid ratio was lower in the intramuscular fat of experimental groups 1 and 2 by, respectively, 1.30 (P<0.10)-1.31 times (P<0.05) and 2.11-2.14times (P<0.001).
It was found that the trombogenic index of meat of experimental groups 1 and 2 was, respectively, 17.1 (P<0.001) and 17.9% (P<0.005)lower in comparison with control groups 1 and 2. No significant decrease of the meat atherogenic index in both treated groups was observed in comparison with the control groups.
Discussion
The results of our study showed that in both trials, no significant difference in the growth rate between the pigs of the treated and control groups was found and this is in agreement with the findings of Morel et al. (2006), Nurnberg et al. (2011), andOkrouhlá et al. (2013), who used linseed or linseed oil in the diet. However, in our trials, a tendency was observed for a more rapid daily gain and lower feed intake per kg gain and significantly shorter duration of fattening time in the pigs treated by linseed oil sediment. Higher feed conversion value in barrows fed linseed was also determined in the findings of Okrouhlá et al. (2013).
The analysis of amino acids showed that palmitic (C16:0) and stearic (C18:0) fatty acids were dominant in the meat of both control and treated pigs and there was no difference for these fatty acids between the groups. A similar result for the stearic (C18:0) fatty acid was reported by Okrouhlá et al. (2013), although the same author, in contrast to our findings, reported a significantly lower content of palmitic (C16:0) fatty acid in the meat of pigs fed linseed. It was found that the treatment of experimental group 1 resulted in significantly lower content of palmitoleic (C16:1n-7) and vaccenic (C18:1n-7) MUFA. Okrouhlá et al. (2013) also reported lower content of palmitoleic (C16:1n-7) fatty acid in the study with linseed, but as MUFA can be synthesized in the pig body, their contents in meat are not so important as those of PUFA (Enser et al., 2000;Kralik et al., 2010). Bečková and Václavková (2010) and Okrouhlá et al. (2013) indicated that linseed in the diet of pigs decrease total MUFA; however, no similar results were found in our study. The current results revealed that the MUFA:PUFA ratio decreased by 1.02-0.89units in the meat of experimental groups 1 and 2, thereby improving the quality of pork. This agrees with the results reported by Okrouhlá et al. (2013), who also showed that dietary linseed supplement decreased the MUFA:PUFA ratio.
Pigs fed standard compound feed produce "unhealthy" meat due to improper n-6:n-3 ratio in it. This is because of the existing regularity that the composition of polyunsaturated fatty acids in pig feeds influences the PUFA composition in meat (Nguyen et al., 2003). A conventional feed for fattening pigs contains insufficient amount of n-3 fatty acids and has improper n-6:n-3 and PUFA:SFA ratios (Kralik et al., 2010). As linoleic (C18:2n-6) and α-linolenic (C18:3n-3) PUFA cannot be synthesized by porcine organism, the contents of these fatty acids in meat are dependent on their contents in the feed (Okrouhlá et al., 2013). As expected, the results of our trials indicated that there was a significant increase in α-linolenic (C18:3n-3), eicosatrienoic (C20:3n-3), and eicosapentaenoic (C20:5n-3) fatty acids in the meat of both treated groups. This agrees with the findings of Riley et al. (2000), Enser et al. (2000), Rey et al. (2004), andKralik et al. (2010), who also indicated that dietary linseed or linseed oil supplement resulted in a significant increase of the above-mentioned fatty acids.
Our study resulted in significantly higher content of docosapentaenoic (C22:5n-3) fatty acid and this confirms the findings of Enser et al. (2000), who supplemented the pig diet with linseed oil. However, it should be noted that no significant differences for the docosahexaenoic (C22:6n-3) fatty acid were found in our study between the treated and control groups, notwithstanding the intake of linseed oil sediment. This agrees with the findings of Riley et al. (2000), Wood et al. (2003), and Václavková and Bečková (2007), who supplemented pig feed with linseed or linseed oil, but contradicts the findings of Cunnane et al. (1990), Enser et al. (2000), and Kralik et al. (2010), who reported a significant increase of C22:6n-3 in the meat.
The meat of the treated pigs contained a significantly higher amount of α-linolenic (C18:3n-3), eicosatrienoic (C20:3n-3), and eicosapentaenoic (C20:5n-3) fatty acids. Also, the meat from experimental group 2 had a higher content of docosapentaenoic (C22:5n-3) fatty acid. The total of eicosapentaenoic (C20:5n-3) and docosahexaenoic (C22:6n-3) fatty acids, very important for human nutrition, in the meat of experimental group 2 was 0.37 versus 0.18% in the control group 2. This agrees with the finding of Enser et al. (2000), Nuernberg et al. (2005), and Okrouhlá et al. (2013) that, in the body of non-ruminants, long chain polyunsaturated C20-C22 n-3 fatty acids are synthesized from the α-linolenic (C18:3n-3) fatty acid found in the feed. In other words, our study showed that addition of linseed oil sediment in the pig diets had a positive effect on these fatty acids.
The results of our study indicated that the higher the dietary content of α-linolenic (C18:3n-3) fatty acid, the higher the content of the same fatty acid and other n-3 fatty acids in pork, which agrees with the findings of Rentfrow et al. (2003) and Okrouhlá et al. (2013). Nuernberg et al. (2005) and Václavková and Bečková (2007) reported that supplementation of pig diets with linseed or linseed oil resulted in lower total content of n-6 fatty acids and specifically of arachidonic (C20:4n-6) fatty acid; however, our findings in both experimental groups do not confirm this. Enser et al. (1996) and Wood et al. (2003) observed that PUFA:SFA ratio in the meat of pigs fed conventional compound feeds is lower than that recommended for human nutrition (over 0.4). In our study with linseed oil sediment, this ratio in treated groups increased by 0.08 unit (P<0.05 and P<0.10). Similar results were reported by Nurnberg et al. (2011) and Okrouhlá et al. (2013), who supplemented pig feeds with linseed or linseed oil.
Additionally, Wood and Enser (1997) and Enser et al. (2000) indicated that the unusual n-6:n-3 fatty acid ratio in pork is above 10. The recommended ratio for human nutrition should be 4-5 or even lower (Okrouhlá et al. 2013). In our study, this ratio in the meat of the treated pigs, if compared with that of the control pigs, was 2.11-2.14times lower and was equal to 3.16-3.27. This is in agreement with the findings of Sheard et al. (2000) and Wood et al. (2003), who indicated that dietary linseed supplement reduced the ratio to 4 or even less. Rey et al. (2004) and Kralik et al. (2010) also reported that supplementation of pig diets with Effect of linseed oil sediment in the diet of pigs on the growth performance and fatty acid profile of meat R. Bras. Zootec., 47:e20170104, 2018 linseed oil had a significant effect on the ratio in pork, making it more suitable for human nutrition.
Pork contains rather high content of linoleic (C18:2n-6) fatty acid, which has a negative influence on the C18:2: C18:3 ratio important for human nutrition. In our study, this ratio decreased from 2.3 in Trial 2 to 2.9 in Trial 1 in the treated groups as compared with the control groups and was very close to the recommended value (4.0) in a healthy food (Wood et al., 2003).
It was found that in experimental groups 1 and 2, the thrombogenic index was significantly lower. The same conclusion was reached by Okrouhlá et al. (2013), who also reported lower thrombogenic index in the trials with linseed. Furthermore, as it was shown in our study, the atherogenic index of meat was by 0.03 times lower in experimental groups 1 and 2, but the differences were insignificant. Analogous data were reported by Okrouhlá et al. (2013) in the study of linseed in pig diets.
Conclusions
Supplementation of pig diets with linseed oil sediment has no significant influence on the growth performance of pigs; however, it increases the content of α-linolenic, eicosatrienoic, eicosapentaenoic, docosapentaenoic fatty acids and total content of n-3 polyunsaturated fatty acids and have a positive effect by improving the polyunsaturated fatty acids/saturated fatty acids and n-6:n-3 fatty acid ratios in meat and reducing the thrombogenic index of pork.
Table 2 -
Composition of vegetable oil and linseed oil sediment
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-04-02T13:13:39.665Z", "created": "2018-02-01T00:00:00.000", "id": "89749997", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.scielo.br/pdf/rbz/v47/1806-9290-rbz-47-e20170104.pdf", "pdf_hash": "c1e4a464e3d2403e9173208f19b987695c53f229", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:33968", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "c1e4a464e3d2403e9173208f19b987695c53f229", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of dietary protein quality on energy metabolism and thyroid hormone Status in growing pigs
To estimate long-term effects of dietary protein quality on energy metabolism and thyroid hormone Status in growing pigs two experiments were carried out, each using 6 growing German Landrace barrows (40 to 90 kg body weight (BW)) per treatment group, which were fed semisynthetic isoenergetic diets based on either casein or soy protein isolate at 1875 kJ ME/(kg BW x d). Casein was tested with (CAS+) amino acid (AA) supplementation (methionine + cystine, threonine, tryptophane) and soy protein isolate was tested without (SPI-) AA supplementation at the recommended protein supply of 100% (normal protein level (NP)) and at a protein supply of 50% of NP. During experiments pigs were housed individually in metabolic cages at 23 ± 1°C. At both protein supply levels, SP1in comparison to CAS+ caused a lower protein energy retention (PER), which was compensated mainly by an increased fat energy retention (FER). The reduction of the protein supply to 50% caused a lower PER by 30 to 50% in both dietary qualities, which was compensated by a significantly higher FER. However, the heat production (HP) was neither affected by the protein quality nor by the quantity, and resulted in nearly similar values of 60% of ME intake. The thyroid hormone concentrations were dependent primarily on the amount of protein supply, and after decrease of supply to 50% secondly on the dietary protein quality. The increased thyroid hormone concentrations at the 50% protein level were in euthyroid ränge of pigs and obviously not associated with HP.
Schlüsselwörter: Proteinqualität, Sojaproteinisolat, Casein, Schilddrüsenhormone, Gesamtenergieverwertung, Schweine Introduction Long-term feeding of diets, which meet the energy requirements for maintenance and growth, but not the requirement for essential amino acids (AA), decreases growth rate and protein deposition in growing animals. The excess of dietary energy, which cannot be deposited as protein can either be used for fat deposition (KEAGY et al., 1987) or dissipated as heat (TULP et al., 1979;GURR et al., 1980). Although several studies were carried out to measure energy balance under these conditions, only a few studies have attempted to investigate the regulatory factors of thermogenesis. It is known that thyroid hormones may play a role in mediating the thermogenic response to low protein diets. Less research has been done in examining the effect of feeding dietary proteins with a low biological value on energy metabolism of growing animals. From previous data one cannot conclude unambiguously whether specific AA patterns are responsible for differences in efficiency of the utilization of metabolizable energy (ME). It is observed that long-term feeding of plant protein (e.g.soy protein, wheat gluten) with somewhat lower biological value, in comparison to high quality animal proteins (casein, egg protein) induced higher serum thyroid hormone levels, particularly higher total thyroxine (T4) levels, in monogastric animals (CREE and SCHALCH, 1985;FORSYTHE, 1986;BARTH et al., 1988;SCHOLZ-AHRENS et al., 1990;POTTER et al., 1996). The question arises whether a higher T 4 level in AA deficient fed animals represents a true change in biological activity and does play a role in dissipating excess dietary energy. The objective of the present study was to investigate the influence of dietary protein quality (AA pattern) on energy metabolism and thyroid hormone Status, as well as to explore whether differences in energetic efficiency are associated with changes in circulating thyroid hormone levels.
Animals and diets
Two separate experiments were carried out each with six castrated male pigs of the German Landrace, weighing between 30 and 90 kg. At the initial body weight (BW) of 30 kg animals were equipped with permanent vein catheters for stress-free blood sampling. During the experiments the pigs were housed individually in metabolic cages. To meet the thermic demands of the animals the environmental temperature at the start of the trials was kept at 25°C. With increasing body weight the temperature was decreased up to 22°C. The relative humidity was 60-70%. They were fed twice daily with semisynthetic, isoenergetic diets, which provided 2.5 times the maintenance requirement for metabolizable energy (1875 kJ ME/(kg BW 062 x d). This corresponded to a daily feed intake of approximately 110g DM/kg BW 0 62 . Water was offered to ad libitum intake. To enable a comparison with human nutrition, nutrient composition of the diets was similar to the composition of human diets in western industrial countries (protein, 8-18%; starch, 32-42%; sucrose, 20%; fat, 15%; cellulose, 7%; mineral and vitamin mix, 8%, wt/wt). Protein and starch content were altered, depending on age and level of protein supply. Casein was supplemented with limiting essential AA (Met + Cys, Thr, Trp) to the level recommended by the GfEH (1987). Correspondingly, CAS+ was supplemented with 1,15 g Met, 0.58 g Thr and 0.46 g Trp per 16 g N. Soy protein isolate was tested without (SPI-) supplementation. In the SPI-treatment the AA Lys, Met + Cys, Thr and Trp, provided 74, 58, 75 and 73% ofthe recommended level, respectively. For pre-calculation of ME values in diets (kJ/kg DM), the equation of HOFFMANN et al. (1993) was used. Effects of the dietary protein quality were compared at normal protein level (NP, experiment 1) and at low protein level (LP, 50% of NP, experiment 2).
Treatments and experimental procedures
In both experiments, trial periods were performed to the same pattern: 10 days of preperiod and 8 days of main period with determinations of energy and protein balance (CN balance), including 4 days of measurements of gaseous exchange in the respiration Chamber. During the main period pigs were kept in metabolic cages allowing daily collection of urine and faeces separately. The body weight of the animals was recorded weekly and health Status was monitored by daily measuring of rectal temperature. Each experiment was performed as a cross-over trail, i. e. after the 2" period the dietary proteins CAS+ and SPI-were replaced by one another. Pigs were given 21 days to adapt to the new experimental conditions and diets. Measurements of CN balance were done in all periods and determination of thyroid hormones in the 2 nd and 4 period. For thyroid hormone determination blood samples (10 mL/day) were taken during the main period prior to the morning feeding. Each sample was allowed to clot for 12 hours, followed by centrifugation with 3000 xg at 4°C. Aliquots of sera were taken and stored at -20°C until assayed for thyroid hormones.
Analytical methods
The energy content of feed, freeze-dried faeces and urine was determined with an adiabatic bomb calorimeter (C400; JANKE & KUNKEL GmbH, Staufen, Germany). The amino acids were analyzed by ion-exchange chromatography using an amino acid analyser BIOCHROM 20 (PHARMACIA-BIOTECH EUROPE GmbH, Freiburg, Germany). For all other analyses of feed, faeces and urine conventional methods ofthe Verband Deutscher landwirtschaftlicher Untersuchungs-und Forschungsanstalten (VDLUFA, 1988) were used. The energy balance was measured by indirect calorimetry based on the carbon and nitrogen balances. Energy retention was calculated by using the factors given by BROWER (1965) and HOFFMANN and SCHIEMANN (1980). Heat production was calculated as difference between ME intake and energy retention. Thyroid hormones, total thyroxine (T 4 ), total triiodothyronine (T 3 ) and the free forms, fT 4 and fT 3i were determined by radioimmunoassays (RIA-COAT, BYK-SANTEC DIAGNOSTICA GmbH & Co. KG, Dietzenbach, Germany).
Statistical analysis
Effects of protein sources within periods were evaluated by one-way ANOVA using SPSS (Statistical Package for the Social Science, Version 7.5, Chicago, 1997). All the results presented in tables are mean values with Standard deviations. Differences were considered to be significant at p < 0.05.
Results
Nutrients and energy of both diets were highly digestible (approximately 90%, data not shown). The apparent N digestibility was influenced by both the protein quality and quantity and ranged between 89.9 and 93.5 at NP and between 82.9 and 87.1% at the LP level, always with the lower values in the SPI-groups. With the exception of the 2 nd period of NP the differences in the intake of metabolizable energy between CAS+ and SPI-within periods were not significantly different (Table 1). Generally the intake of ME in SPI-fed pigs was slightly lower. Within the periods of NP the intake of protein per kg BW 0,62 was not significantly different. At LP the intake of protein per kg BW 062 within the periods was slightly but significantly lower for SPI-than for CAS+. Both, protein quality and quantity, affected the daily weight gain of pigs (Table 1). At NP the values for CAS+ fed pigs in the l st and 2 nd period were 504 and 567 g, respectively, which was in both periods approximately 90 g higher than the daily gain of the SPI-pigs. After cross-over feeding, the difference between CAS+ and SPIincreased to 130 g/d in the 3 rd period; in the 4^ period the difference was 86 g/d. In all periods of NP the differences were statistically significant. Reduction of the protein content to 50% (LP) generally resulted in a reduced growth Performance in both dietary groups. The daily gain of CAS+ fed pigs amounted from 302 g in the l sl to 494 g in the 4* period, and of SPI-fed pigs from 262 to 406 g, respectively. With the exception of the 3 rd period, after cross-over feeding, differences in daily gain between CAS+ and SPI-groups remained significant. The final BW within the same time period was approximately 90 kg at the NP and approximately 60 kg at the LP feeding level.
At almost the same ME intake levels, protein energy retention (PER) was markedly affected by protein quality and quantity. In the l st period of NP, feeding of SPIresulted in a significantly lower PER than feeding of CAS+ (205 vs. 279 kJ). The values of PER in the 2 nä period decreased for CAS+ by 27% and for SPI-by 18% compared to the l st period. However, the difference between both protein sources was significant. After cross-over feeding, the values of PER for SPI-remained constant but the value for CAS+ increased to 221 kJ. In the 4 th period the values of PER decreased further for both protein sources, however the differences between CAS+ and SPIremained significant. As the PER during growth decreased, the fat energy retention (FER) increased for CAS+ from 394 kJ in the l sl period to 715 kJ in the 4* period, and for SPI-from 447 to 642 kJ, respectively. Also, the total energy retention (ER) increased from 673 kJ for CAS+ and 651 kJ for SPI-in the l st period to 882 and 785 kJ in the 4 lh period, respectively. Correspondingly, heat production (HP) decreased for both treatments. Although in all periods of NP feeding of SPI-as compared to CAS+ resulted in a significantly lower PER, no significant differences in values of FER (exception of4 l period), as well as of HP, between CAS+ and SPI-were determined. In Table 2 the partition of ME into PER and FER as well as into ER (total efficiency of utilization of ME) and HP in % are calculated. With exception of the 4 th period, feeding of SPI-in comparison to CAS+ resulted in a significantly lower proportion of ME in PER; values of FER/ME and ER/ME as correspondingly HP/ME were not significantly different but always on an average slightly higher for SPI-. There was a decrease of PER/ME and an increase of FER/ME for both dietary groups during growth of the animals. In the periods 1 to 3, CAS+ fed pigs showed significantly higher values of PER/ER than SPI-fed pigs. During growth, as the values of PER/ER decreased the values of FER/ER increased. Additionally, the differences between CAS+ and SPI-diminished with increasing BW, however in the 4* period no significant differences were observed. The reduction of protein supply to 50% (LP) resulted in a similar HP than at NP (Table 1). At LP markedly lower PER (50 to 60%) and higher FER (30 to 40%) than at NP in both dietary groups were observed. During growth PER decreased slightly for CAS+ as well as for SPI-, however the differences between both protein sources remained significant. In all periods of LP for FER and HP no significant differences between CAS+ and SPI-were observed. The proportions of HP in ME (Table 2) between CAS+ and SPI-were not significantly different in all periods of LP; both dietary groups showed an average value of 60%. Feeding of SPI-resulted in a significantly lower proportion of PER in ME than feeding CAS+, FER/ME was not affected by the dietary treatment. There was a decrease of PER/ME and an increase of FER/ME for both dietary groups during growth of pigs. In all periods of LP CAS+ resulted in a significantly higher ratio of PER/ER and in a lower ratio of FER/ER than SPI-. During growth the values of PER/ER decreased for CAS+ from 19.2 to 16.0%; for SPI-the values remained constant at about 12%.
Thyroid hormones
Concentrations of thyroid hormones in the serum measured at both protein levels in the 2 nd and 4 lh fattening period are summarized in Table 3. Thyroid hormone levels were constant within periods, which allowed to calculate an average for each variable (T 4 , T3, fT 4 , fT 3 ) for each period. At NP, serum concentrations of thyroid hormones were not affected by the dietary protein quality. The ratio of T 3 /T 4 was not affected by the dietary protein quality or by the period. In the 4 th period similar thyroid hormone concentrations were determined as in the 2 nd period. At LP, serum concentrations of thyroid hormones were affected by the dietary protein quality. In the 2 nd period for SPI-in comparison to CAS+ not significantly but slightly higher concentrations of T 4 (56.28 vs. 50.84nmol/L) and significantly higher concentrations of fT 4 (16.15 vs. 11.91 pmol/L) were observed. In contrast, significantly lower values of T 3 (1.57vs. 2.32 nmol/L) and fT 3 (0.31 vs. 0.75 pmol/L) were determined compared to the CAS+. Similar observations were made in the 4 period. The reduction of protein supply to 50% (LP) caused a markedly higher T 4 concent- Protein sources were casein wilh amino acid supplementation (CAS+) and soy protein isolate without amino acid supplementation (SPI-). Experiments were cross over trials, i. e. after two periods dietary proteins CAS+ and SPI-were replaced by each other.lb Means with different superscripls within a period and within one column are significantly different (p < 0.05); 'J,, total thytoxine; Ti, total triiodothyronine; ITs.free Uiiodthyronine.ffi, free üiyroxine •One animal was taken out of experiments due to health problems; **Two animals were taken out of experiments due to health problems.ration (30%) for both protein qualities, and a higher T 3 concentration (30%) for CAS+. Consequently, the T 3 /T 4 ratio for CAS+ resulted in similar values of approximately 4.4% as at NP; the ratio for SPI-was decreased to 2.8% in the 2 nd and to 2.4% in the 4* period.
Discussion
It is known that thyroid hormones may play a role in mediating the thermogenic response to low protein diets. The decreased energetic efficiency observed in growing rats and pigs fed low protein diets is often associated with a marked rise in T 3 concentration (TULP et al., 1979, ROTHWELL et al" 1983, GURR et al., 1980), which is frequently interpreted as an adaptive diet-induced thermogenesis. Some energy dissipating Systems such as the hepatic mitochondrial ct-GP Shuttle (TYZBIR et al., 1981, SAWAYA and LUNN, 1985, KEAGY et al., 1987), as well as the thermogenic activity of brown adipose tissue (ROTWELL et al., 1983) are sensitive to the thyroid hormone Status. Because dietary proteins with a lower biological value induce an increase in T 4 levels, we propose that similar to protein deficient diets the excess of dietary energy, which can not be deposited as protein, could also be dissipated as heat through increased thermogenesis.
In the present study at LP the T 4 concentrations were slightly but the fT 4 concentrations were significantly 'higher in SPI-than in CAS+ fed pigs within both periods. These findings are in accordance with literature data (CREE and SCHALCH, 1985;FORSYTHE, 1986;BARTH et al., 1988;SCHOLZ-AHRENS et al., 1990;POTTER et al., 1996). AtNP, serum concentrations of all thyroid hormones were not affected by the dietary protein quality. The latter finding is surprising, because BARTH et al. (1988) and SCHOLZ-AHRENS et al. (1990) have shown that feeding pigs with diets similar in composition to the NP diets, providing 18, 52 and 30% of ME from protein, carbohydrate and fat respectively, resulted in significantly higher T 4 concentrations in SPI-fed pigs. Reasons for the contrary endocrine response may be due to different pig breeds, age and sex of animals. In the present study growing castrated male pigs of the German Landrace were fed isoenergetic diets, which provided for 2.5 times the maintenance requirement of ME. BARTH et al. (1988) and SCHOLZ-AHRENS et al. (1990) used adult female Göttingen miniature pigs fed isoenergetic diets at the maintenance requirement of ME. However, BARTH et al. (1988) obtained in experiments with growing minipigs fed 15% soy protein isolate or casein higher values of T 4 and fT 4 in the SPI-group, but the difference was statistically not significant. These findings correspond better to our findings at the NP level. The present results indicated that both CAS+ and SPI-fed pigs responded similarly to the low protein feeding regime by exhibiting an increase in total T 4 concentration of 30%. Interestingly, the T 3 and fT 3 levels were only increased in the CAS+ fed pigs. However, we expected results in the opposite direction. But recently, similar inverse relationships between T 3 concentration and casein diets have been reported by POTTER et al. (1996), who carried out experiments with hamsters. Our findings are partially in disagreement with other literature data (TULP et al. 1979, ROTHWELL et al., 1983), which showed that consumption of low protein diets is associated with an increase in T 3 concentration and mostly with no effect on the T 4 level. However, Interpretation is made difficult, if animals were fed diets to ad libitum intake. In contrast to our findings such data do not allow to conclude unambiguously whether protein supply is responsible for differences in thyroid hormone levels. It is possible that the increased T 3 level in LP diets was a response to an elevation in food and energy intake (DANFORTH et al., 1979) Our observations of increased serum T 4 concentrations are consistent with findings of ORIEN et al. (1979), who estimated elevations in T 4 concentrations in young rats fed protein restricted diets (8%) of a normal energy density in comparison to high protein diets (22%). Because T 4 is synthesised exclusively from the thyroid gland it may be suspected that the observed higher T 4 levels may indicate an increase in thyroid secretion induced by dietary protein deficiency. In contrast, ATINMO et al. (1978) observed lower serum T 4 concentrations in pigs after feeding low protein diets. Furthermore, the question arises why it comes to contrary endocrine response in T 4 concentration after feeding low protein diets. BERGNER (1989) found lower thyroid secretion rates (TSR) in pigs and rats after feeding proteins with a lower biological value. He concluded that the T 4 concentration in the blood depends on the thyroid secretion rates and the T 4 utilization of tissue. From this point of view it can not be concluded that higher serum T 4 concentrations exclusively derive from a higher TSR. The ratio T 3 /T 4? as a measure of the peripheral deiodination of T 4] should also be interpreted critically. NOWAK and SLEBODZINSKI (1986) proved that the greatest proportion of the daily T 3 production derived from peripheral monodeiodination of T 4 with values ranging between 70 and 80%. In the present study the T 4 levels at LP were increased in both dietary groups, the CAS+ groups resulted in similar values of T 3 /T 4 ratio at both protein levels. Apparently at LP, more T 3 is derived from T 4 in the periphery in CAS+ fed pigs, which is also covered by higher T 3 values. The T 3 levels for SPI-were similar at both protein levels with the consequence of a 50% reduction in the T 3 /T 4 ratio at LP. In general, a higher T 4 level is followed by a higher T 3 level (VOSBERG and WAGNER, 1991). A reason for the contrary endocrine response of SPI-at LP may be due to the lower protein quality of SPI-. As described by LAUTERIO and SCANES (1987), Phenylalanine and tyrosine deficiency have been found to decrease circulating T 3 concentrations. Since SPI-contains approximately 35% less of both AA than CAS+, the lack of an increase in T 3 concentration following an increased T 4 level in SPI-, in comparison to CAS+, may be explained by the deficiency of these two AA in SPI-at LP. In summary, our findings indicate that the circulating concentration of thyroid hormones depends primarily on the protein supply and after a decrease of supply to 50% secondly on the dietary protein quality. In general, long term feeding of dietary proteins with a lower biological value leads to significant lower growth Performance, i.e. protein deposition, than feeding dietary proteins with a higher value (SALTER et al., 1990;ROY et al., 1997;SEVE et al., 1997). The present results indicate changes in the same manner when feeding SPI-with the lower biological value in comparison to CAS+ in all periods at both protein levels. The excess of dietary energy, which cannot be deposited as protein, can either be used for fat deposition (KEAGY et al., 1987) or dissipated as heat (TULP et al" 1979;GURR et al., 1980). In the present study the total efficiency of utilization of ME (energy retention/ME intake) within periods at both protein levels was not significantly different between CAS+ and SPI-. Independent on the protein quality and quantity, pigs retained 38 to 42% of ME. The slightly lower efficiency of utilization of ME (36 vs. 40%) in the l st period of NP, in both feeding groups, was caused by the higher PER in relation to FER (see table 2). However, in general, our findings are in accordance with the data of a review presented by BERGNER and HOFFMANN (1996), who indicated values of 40 to 45% for growing finishing pigs. Under the conditions of feeding isoenergetic diets, which provided 2.5 times the maintenance requirement of metabolizable energy (1875 kJ ME/(kg BW 062 x d)), the pigs fed SPI-retained less protein energy per unit metabolic BW compared to pigs fed CAS+. However, with one exception, SPI-feeding neither resulted in a significantly higher FER nor in a significantly higher HP. There are several reasons for no observed differences in both of these parameters with regards to CAS+ and SPI-. First there is, on average, a 3% lower ME intake in SPI fed pigs. Secondly, the changes in PER between SPI-and CAS+ are significant, however, they are expressed as values with an average of 46 kJ/(BW 062 x d). When taking the value ofthe Standard deviation (on average 53 kJ/(BW 062 x d)) of HP into consideration, it is evident that the Standard deviation in HP is equal or even higher than the difference in PER between CAS+ and SPI-, therefore no significant differences in FER and HP could be expected. In addition the higher excess of ME in SPI-, in comparison to CAS+, seems to be divided into both FER and HP. Analogous experiments with growing rats fed low protein diets based on SPI-or CAS+ (KLEIN et al., 2000) are not consistent with our findings of a similar energetic efficiency in SPI-and CAS+ fed pigs. The SPI-fed rats retained both less protein energy and fat energy in comparison to CAS-fed rats. Several experimental results reported in the literature (e.g., MÜLLER and KIRCHGESSNER, 1979;SCHIEMANN et al., 1983;reviewed by KLEIN and HOFFMANN, 1989) demonstrated that the additional energy costs for protein deposition compared to fat deposition are relatively high. The protein deposition is connected to the processes of protein tumover, in which the additional energy costs are mainly caused by higher synthesis rates. MILLWARD et al. (1976) calculated a value of energy costs (ME) for synthesis, which is dissipated as heat, of 0.15 kJ/kJ and 3.6 kJ/g protein synthezised, respectively. This value is based on the generally accepted energy cost of 5 moles ATP per mole of amino acid incorporated in the peptide chain, and represents the minimum cost of protein synthesis. When considering the relatively high Standard deviation in HP, the question arises whether there were differences in the protein synthesis between SPI-and CAS-fed pigs, which produced measurable differences in HP. In the present study measurements of energy balance and protein tumover were carried out simultaneously (SAGGAU et al., 2000). The comparison between CAS+ and SPI-demonstrated that in CAS+ fed pigs, the higher protein deposition was realized by both a higher synthesis and degradation, e.g., enhanced protein tumover. Under consideration of calculated values for minimum energy costs (ME) of protein synthesis (MILLWARD et al., 1976) and the observed differences in protein synthesis between the both dietary groups, no measurable effect on HP can be expected between CAS+ and SPI-within the periods of both dietary levels. FÜLLER et al. (1987 a, b) also carried out trials with growing pigs fed daily a constant energy supply with a low in comparison to a high protein level and a Variation of protein quality by altering the level ofthe first limiting AA, lysine, at both protein supply levels. Their data support our conclusion that there was no change in heat production after feeding an AA deficient diet. As mentioned above, higher T 4 and fT 4 levels were estimated in SPI-than in CAS+ fed pigs at NP. In addition, both SPI-and CAS+ fed pigs responded similarly to LP by exhibiting an increase in total T 4 concentration of 30%. However we could not establish any relationship between increased thyroid hormone concentrations and heat production. HILLGARTNER and ROMSOS (1987) postulated that higher thyroid concentrations resulting from consumption of low protein diets are not directly responsible for activation of adaptive thermogenesis. HILLGARTNER and ROMSOS (1987) carried out experiments with rats fed diets containing 5, 8 or 22% CAS. The T 3 level was increased in the same manner when fed 5 or 8 in comparison to 22% CAS diets but the efficiency of energy retention was reduced in rats fed 5% CAS, whereas no significant change was observed in rats fed 8%. From these findings, the authors concluded that thyroid hormones play only a permissive role in dissipation of energy in growing rats. In a recent paper, MOROVAT and DAUNCEY (1998) analysed the thyroid hormone Status in growing pigs of the Large White breed after modification of feed intake. Pigs with a total T 4 concentration of 37.9 ± 3.9 nmol/L were described as euthyroid and those with a concentration of 70 nmol/L and higher were considered hyperthyroid. From these findings, we concluded that total T 4 concentration (50 to 56 nmol/L) at LP were slightly increased but still euthyroid. Furthermore, the question arises of how an excess of dietary energy can increase either fat deposition or heat production. GURR et al. (1980) demonstrated both metabolic pathways in growing pigs. There were, however, essential differences to our trials. GURR et al. fed either restricted amounts of a high protein diet (26%) or to ad libitum amounts of a low protein diet (2%) to pigs of 6 and 20 kg BW. In all cases, pigs fed low protein diets consumed approximately three times as much as pigs fed high protein diets. In contrast we fed similar restricted amounts of isoenergetic diets per unit metabolic body weight in all fattening periods of NP and LP which enabled a comparison due to protein intake quantity and/or quality differences alone. GURR et al. (1980) demonstrated that in 20 kg pigs, almost 70% ofthe energy excess caused by feeding the low protein diet was deposited in the carcass as fat, which is in accordance to our findings. In the 6 kg pigs fed low protein diets, changes in body energy content accounted for only a small fraction (27%) ofthe total energy intake, a large difference in energy expenditure was seen between these animals and the high protein group, which GURR et al. (1980) attributed to differences in dietary-indueed thermogenesis. In addition, possible metabolic indicators of diet induced thermogenesis were investigated in the group of 6 kg pigs. At the low protein supply these pigs showed higher plasma T 3 levels and hepatic mitochondrial a-glycerophosphate dehydrogenase activity. It must be taken into account that at LP the ratio of protein energy to total energy intake was drastically reduced and corresponded to a supply as in Kwashiorkor patients (COWARD and LUNN, 1981;SCHOPPE, 1988). In these metabolic Situation very young pigs seemed to dissipate the excess of energy mainly as heat through increased thermogenesis. It is apparent from the study of DANFORTH et al (1979) on humans that the intake of high caloric amounts, e.g.above the energetic requirement, also increases T 3 concentrations, which may be associated with an increased thermogenesis. ROTHWELL et al. (1983) speculated that the thermogenic activity of brown adipose tissue (BAT) may play an important role in diet-induced thermogenesis when low protein diets are fed to rats. The BAT contains a tissue-speeifie mitochondrial uncoupling protein (UCP), which is sensitive to thyroid hormones. KEAGY et al. (1987) postulated that protein-deficient chickens cannot dispose a surplus of dietary energy by this mechanism, because most birds apparently lack BAT. It is known, that pigs also do not contain BAT (TRAYHURN et al., 1989) and therefore they cannot use this mechanism. However, recently in white adipose tissue a similar tissue-speeifie mitochondrial uncoupling protein (UCP 2) was discovered (FLEURY et al., 1997).
Table 1
Energy and N balance data of growing pigs fed different dietary protein qualities and quantities (Means and Standard deviations).(Energie-und N-Bilanzdaten wachsender Schweine bei Fütterung unterschiedlicher Nahrungsproteinqualitäten und -quantitäten (Mittelwerte und Standardabweichungen)) Partition of metabolizable energy (ME) into protein energy retention (PER), fat energy retention (FER), total energy retention (ER) and heat production (HP) as Pwell as ratios of PER to ER and FER to ER in growing pigs fed different protein qualities and quantities (Means and Standard deviations).(Aufteilungderjrumsetzbaren Energie (ME) in Proteinenergieretention (PER), Fettenergieretention (FER), Gesamtenergieretention (ER) und Wärmeproduktion (HP) sowie die J Verhältnisse von PER zu ER und FER zu ER in wachsenden Schweinen bei der Fütterung unterschiedlicher Nahmngsproteinqualitäten und -quantitäten (Mittel-Protein sources were casein with amino acid supplementation (CAS+) and soy protein isolate without amino acid supplementation (SP1-). Experiments were cross over trials, i. e. after two periods dietary proteins CAS+ and SPI-were replaced by each other.*•Meanswithdifferent superscripts within a period and within one column are significanüy different (p <0.05).•Oneanimal was taken out of experiments due to health problems; «• Two animals were taken out of experiments due to health problems. SAGGAU et al.: Effects of dietary prolein quality on energy metabolism and thyroid hormone stalus in growing pigs Table3Concentration of thyroid hormones (T 4 , T 3 , fT 4 and fT 3 )' in the serum of growing pigs fed different dietary protein qualities and quantities (Means and Standard deviations).(Schilddrüsenhormonkonzentration (T 4 , T 3 , fT 4 and fT 3 )' im Serum wachsender Schweine bei Fütterung unterschiedlicher Nahrungsproteinqualitäten und
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-11-03T07:10:41.023Z", "created": "2000-10-10T00:00:00.000", "id": "39013110", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://aab.copernicus.org/articles/43/633/2000/aab-43-633-2000.pdf", "pdf_hash": "0ae2dc6bf2fd8f8201022e548168863439d73fcb", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:35581", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "0ae2dc6bf2fd8f8201022e548168863439d73fcb", "year": 2000}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Management of genetic diversity using gene dropping method based on pedigree information
Preservation of genetic diversity in populations is an important task to ensure a possible longterm response to selection in animal breeding. The purpose of this study was to consider how pedigree analysis and gene dropping method could be used for management plans in order to maintain genetic variation in a population under selection of Japanese quail. Therefore, the distributions of alleles frequencies originated from founders were estimated on an actual pedigree using gene dropping simulation software. Then, genetic contribution of founders to the current population, components such as the F-statistics and effective population size were estimated. The results show that from 156 founders there are only 64 of them (22 males and 42 females) in the last generation. The average genetic contribution of a founder male and female contributing to the last generation were 1.87 and 1.40 %, respectively. A total of 87 and 95 % of the genome in the last generation were constituted by 34 and 42 founders, respectively. The effective population size decreased as inbreeding increases. The allele frequency averaged over replicates agreed with the genetic contribution. Some useful information regarding the management of genetic diversity such as the probability of allele extinction, the probability of alleles surviving at a critically low frequency and risk of future allele extinction were derived by using these distributions. Results show that pedigree analysis and gene dropping are valuable tools in optimizing decisions to preserve genetic variability.
Introduction
Along with the enhancement of genetic progress, preservation of genetic diversity in populations is important in animal breeding. This is to ensure that a long-term response to selection for traits of interest is achievable (Barker 2001). Because of the potentially deleterious effects of loss of genetic diversity in small selected populations (Ralls et al. 1986), pedigree analysis is often recommended to allow to design breeding plans to minimize the variation in family size and maximize the genetically effective population size relative to the census population size (Lacy 1989). The expected loss of genetic variation from the populations can be minimized by increasing the number of descendants of poorly represented founders (Foose et al. 1986). Although the mean of the distribution of allele frequency for a founder should coincide with the genetic contribution computed from pedigree analysis, the probability of allele extinction can be obtained only through the gene dropping simulation (Honda et al. 2002). Gene dropping is a simulation procedure in which two unique alleles are assigned to each founder and the genotypes of all descendants along the actual pedigree are generated following the Mendelian segregation rules (MacCluer et al. 1986). However, application of this method rests on certain assumptions such as 50:50 transmission probabilities, no mutation and no migration. The method is also flexible to extend several loci and to measure the change in linkage disequilibrium (Baes & Reinsch 2008). This allows simulating a genome for each animal where the number of repetitions refers to the number of unlinked loci (Suwanlee et al. 2007). This technique provides considerably more information about population structure than is available from calculations of the proportionate contributions of the founders (MacCluer et al. 1986). Furthermore, parameters derived from probabilities of gene origin are very useful to describe a population structure after a small number of generations, whereas the use of effective population size and inbreeding for monitoring the genetic variability over a long time period is very sensitive to the quality of available pedigree information (Pérez Torrecillas et al. 2002). The effective population size is a key parameter in conservation and population genetics because of its direct relationship with the level of inbreeding, fitness and the amount of genetic variation lost due to random genetic drift (Caballero & Toro 2000). The purpose of this study was to consider how pedigree analysis and gene dropping methods can be used for management plans for maintaining genetic variation in a selected quail population. Therefore, by using gene dropping simulation software (Lange et al. 2001) on an actual pedigree, the distributions of frequencies of alleles originating from founders were estimated at the first and secondly, genetic contribution of founders to the current population, inbreeding and effective population size were estimated. From the distributions of frequencies of alleles, useful information for the management of genetic diversity, such as the probability of allele extinction and the probability of alleles surviving at a critically low frequency, were derived.
Material and methods
The experimental Japanese quail population (Coturnix coturnix) originated from a commercial farm in Yazd city (Iran). The population had not been selected for any traits before the start of the experiment. A total of 126 birds (generation 0) were randomly selected from the population and allowed to reproduce with the sex ratio of 1:2 (male:female) to establish a selection line. In the next generation, 30 birds more were introduced (unknown parents) into the population. These birds had then been selected for 4 wk body weight (Khaldari et al. 2010). The number of male and female birds and contributing founders in each generation are presented in Table 1. The gene dropping simulation applied is illustrated in Figure 1, in which the process of one trial of the simulation is presented with a simple pedigree. To obtain a reliable distribution of allele frequencies in the reference population, the process was replicated 1 000 times using the Mendel software (Lange et al. 2001). The probability of extinction of alleles originating from a founder (Pr (lost)) was calculated from the proportion of replicates in which both alleles derived from the founder did not segregate in the reference population. Similarly, the probability of alleles being at high risk of extinction (Pr (risk)) was obtained by the proportion of replicates in which allele frequency (q) was within the range of 0<q<0.01. The upper limit (0.01) of the range was chosen following MacCluer et al. (1986). Finally, the probability of alleles surviving at a critically low frequency, conditional on the founder alleles being retained in the reference population, was computed as: Allocation of unique hypothetical alleles (G) to founders (F) and genotype assign to descendants (A, B, C, D) by Mendelian segregation of founder alleles.
Figure 1 Illustration of gene dropping simulation
This conditional probability is an indicator for the risk of future allele extinction. It should be noted that these three probabilities are conditional on the pedigree structure (Honda et al. 2002).
The contribution of founders and the inbreeding coefficients (F) for all the animals in the pedigree and F-statistics were calculated using the Eva-Inbred software 1.3 (Berg 2010). The F-statistics (F IT , F ST and F IS ) were estimated in each generation to assess the amount of inbreeding and the population structure. The coefficient F IT is the average inbreeding coefficient and F ST is the inbreeding coefficient expected under random mating. The latter was estimated as the average kinship between the sires and dams (parents) of the generation. The third coefficient F IS is the deviation from random mating obtained by the formula of Wright (Wright 1969) as: The actual inbreeding (F IT ) exceeds the level expected under random mating (F ST ) when F IS >0, implying that mating among more closely related parents than expected is predominant or that the population is partitioned into subpopulations and mating is more or less restricted within each subpopulation. In contrast, avoidance of inbreeding or mating between subpopulations is predominant in populations with F IS <0. The effective size of the population was estimated from the increasing rate of F ST per generation (Caballero & Hill 1992, Wright 1984). The generation rate of inbreeding (ΔF ST,g ) of F ST was first computed as ΔF ST,t = (F ST,t − F ST,t−1 )/(1 − F ST,t−1 ), where F ST,t−1 and F ST,t are the coefficients of F ST in two successive generations. The effective population size was then computed as 1/(2ΔF ST,t ) (Nomura et al. 2001). Finally, we reached to the following formula for calculating the average allele frequencies for each founder: where sTA j is the sum of the total alleles of founder j, nR=1 000 is the number of replicates, N is the number of individuals in the last generation and p i and q i are the allele frequencies for the two alleles of founder j in the last generation of replicate i.
Results
Descriptive statistics for the number of dropped alleles from base population into each generation are summarized in Table 2. Although the number of alleles retained in each reference population varied among replicates of simulation, the consistently small coefficient of variation implies that the average over replicates is a good indicator for the number of retained alleles. The number of birds, inbred birds, F-statistics and effective population size are presented in Table 3.
The average genetic contribution of founders to the last generation is presented in Figure 2. The results revealed that only 64 founders consisting of 22 males and 42 females contributed their genetic material to the last generation. The average genetic contribution of a male and female founder to the last generation was 1.87 and 1.40 %, respectively. A total of 87 % of the genome in the last generation has been contributed by 34 founders, with 13 ∑ 1000 i=1 males and 21 females contributing 37 and 50 %, respectively. The relative contribution of the 22 male and 42 female founders were 41.1 and 58.9 %, respectively. A more detailed description of extreme founders contributing to the last generation is presented in Table 4. It can be seen that the genetic contribution based on the pedigree is a good estimate of the realized contribution. It is also evident that all founders still segregating have a considerable risk of being lost in the future, irrespective of their risk of being lost in the previous four generations.
Retained alleles and founders
Variable contributions of dropped founders into the last generation (Figure 2) show that 92 founders (59 %) have no descendants. Furthermore, about 87 and 95 % of the alleles in the last generation is contributed by only 34 and 42 founders, respectively. After increasing the average number of retained alleles in generation 1 (due to additional founders), there was a rapid decrease of alleles, which declined to 60 alleles in the last generation. These 60 segregating alleles could be assigned to the 128 alleles of the 64 founders contributing to the last generation. Although the number of alleles retained varied among replicates, the small coefficient of variation implies that the average over replicates is a good indicator for the number of retained alleles. The major reason for decreasing the founder' contribution and subsequently the average number of retained alleles and also increasing of coefficient of variation is the selection of progeny of superior founders as parents of the next generation. The rapid loss of alleles in early generations has also been reported (Honda et al. 2002, Rodrigáñez et al. 1998).
The proportion of founders in the category of less than 10 % risk of being lost (0-0.10)decreased from 0.89 in the base population to 0.15 in the last generation. In contrast, proportion of founders with a 50-100 % probability of being lost increased from 0.11 to 0.26. Generally, most of the founders still contributing to the last population were at risk of being lost in future generations.
Expectedly, the allele frequency averaged over replicates agreed with the genetic contribution. The small coefficients of variation imply that allele frequencies had a little variation around the average. For the 64 founders contributing to the last generation, founders with higher genetic contribution have lower probabilities of allele extinction (Pr(lost)) and probability of allele extinction increased as genetic contribution decreased (Table 4). In the management of genetic diversity, Pr (lost) gives useful information, for example the low probability of allele extinction of the founder implies that alleles can be surely transmitted to the progenies (Trinderup et al. 1999).
The probability of allele extinction of the three founders with the highest genetic contributions was near zero but their surviving alleles had relatively high risks of future extinction (0.28, 0.24 and 0.37). In contrast, the probability of allele extinction of founders with the three lowest contributions were high (>0.75)while their surviving alleles had a relatively low risk of future extinction. The high probability of allele extinction and the low risk of future extinction (Pr(risk|survive)) of alleles indicated that the alleles had passed a strong drift in the early generations. In contrast, the low probability of allele extinction and relative high risk of future extinction state that genetic contribution of founders are not fully informative for the distribution of allele frequency. This is in agreement with the results from Honda et al. (2002). It is also evident from comparing two founders of 263 and 505, each in which Pr (lost)=0 for both of them and the risk of future extinction is the same but a genetic contribution of the former is about 2.5 times higher than that of the latter. Nevertheless, all founders have a significant probability of future extinction.
inbreeding and effective population size
Results from Table 3 show that the effective population size decreases as inbreeding increases. However, the disproportionate contributions of the founders or ancestors is caused by many factors, such as population structure, mating policy, selection etc., but estimates of effective population size based on the increase or decrease in inbreeding would accurately reflect the genetic history of the population, namely the size of their founder population, their mating policy or bottlenecks due to abusive use of reproductive individuals. All these phenomena influence the pedigree of the individual and therefore are reflected in the increase or decrease of inbreeding (Gutie et al. 2008). In this study, the rate of inbreeding was 0.33 % in generation 2 up to 1.1 % in generation 4. This result is in agreement with Brah et al. (2001) while it is in opposite with Biedermann et al. (2009) that reported 4.62 % and 7.12 % at generation 2 and 3, respectively for a population of white park cattle. These high figures of inbreeding are due to the small number of population (11 males and 33 females).
The effective population size is a very important parameter to monitor breeds and breeding programmes, as it affects the inbreeding depression and the loss of genetic variability. So the maintenance of genetic variation in breeding programmes is important for short-and long-term selection responses as well as for conservation purposes (Hill 2000). The large number of founders in the reference population and the low effective size in Table 2 is showing that some founders could be used more intensively than others, which is always a function of loss of genetic diversity. This is in agreement with a report of Vostrý et al. (2011) where 98, 139 and 168 founders were in the total reference population for three breeds of horse but the effective numbers of founders were only 43.14, 79.11 and 52.21, respectively. Meuwissen & Woolliams (1994) suggested that if the effective population size is in the range of 31-250, it would prevent the decline of fitness owing to inbreeding depression. Goddard & Smith (1990) suggested 40 as a minimum effective size.
A population with unequal representation of founders will contain less genetic variability than a population with the same number of founders, in which the founders have made equal contributions to future generations (Lacy 1989). This loss of variation is reflected by the loss of heterozygosity and the loss of allelic variants. Lower heterozygosity often results in lower average fitness of individuals (inbreeding depression) whereas a lack of allelic variants prevents long-term adaptive response to selection (Falconer et al. 1996).
We can estimate the distribution of allele frequencies and confidence intervals for the allele frequency by applying the gene dropping approach. This estimation can be valuable for observing trends in active breeding populations as have been shown by Manatrinon et al. (2009). Thus, for a sustainable development of response to selection, preservation of alleles from the lowest contributed founders should be an urgent task.
It can be concluded that gene dropping contributes to a detailed description of the genetic variation in a population and the risk of future loss of genetic variation. Thus, it is a valuable tool to optimize decisions to preserve genetic variability.
Figure 2
Figure 2 Genetic contributions (%) of 64 retained founders in the last generation
Table 2
Descriptive statistics for the number of dropped alleles per replicate and the total number of alleles assigned to founders in each reference population(Na) N: number of birds, Ne: effective population size
Table 4
Genetic contribution based on pedigree, average allele frequency, probability of being lost and the risk of being in future generations of founders with extreme genetic contributions
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-09-05T11:05:16.690Z", "created": "2013-10-10T00:00:00.000", "id": "52241065", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.arch-anim-breed.net/56/518/2013/aab-56-518-2013.pdf", "pdf_hash": "34b48a22075bbbd1a49b74efc32445a67e89772b", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:35596", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "43dd1975193830b9243c8eef672db134e42e871d", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effect of Salinity Stress on Physiological Changes in Winter and Spring Wheat
: Salinity is a leading threat to crop growth throughout the world. Salt stress induces altered physiological processes and several inhibitory effects on the growth of cereals, including wheat ( Triticum aestivum L.). In this study, we determined the effects of salinity on five spring and five winter wheat genotypes seedlings. We evaluated the salt stress on root and shoot growth attributes, i.e., root length (RL), shoot length (SL), the relative growth rate of root length (RGR-RL), and shoot length (RGR-SL). The ionic content of the leaves was also measured. Physiological traits were also assessed, including stomatal conductance ( gs) , chlorophyll content index (CCI), and light-adapted leaf chlorophyll fluorescence, i.e., the quantum yield of photosystem II ( Fv (cid:48) /Fm (cid:48) ) and instantaneous chlorophyll fluorescence ( Ft ). Physiological and growth performance under salt stress (0, 100, and 200 mol/L) were explored at the seedling stage. The analysis showed that spring wheat accumulated low Na+ and high K+ in leaf blades compared with winter wheat. Among the genotypes, Sakha 8, S-24, W4909, and W4910 performed better and had improved physiological attributes ( gs , Fv (cid:48) /Fm (cid:48) , and Ft ) and seedling growth traits (RL, SL, RGR-SL, and RGR-RL), which were strongly linked with proper Na + and K + discrimination in leaves and the CCI in leaves. The identified genotypes could represent valuable resources for genetic improvement programs to provide a greater understanding of plant tolerance to salt stress.
Introduction
Plants respond to environmental changes by altering metabolism, growth, and development. When climate changes are rapid, plants perceive them as stress. Among the abiotic stresses, those that most affect the productivity of agricultural crops are: extreme temperatures, osmotic stress, drought, and salinity. Many environmental conditions can lead to water stress in plants. For example, the high concentrations of salt in saline habitats germplasm in wheat. Since wheat (Triticum spp.) is a major food crop, the development and identification of salt-tolerant wheat cultivars is an important research purpose.
Fluorescence of chlorophyll (e.g., Fv /Fm , Ft) and gas exchange have been considered important physiological indicators for screening the tolerance of different cultures. There are two important exit photosystems (PSI and PSII) in the plant. PSII is found to be more prone to the hazardous effects of salinity [26]. Measuring chlorophyll fluorescence is a good indicator of salt effects in the photosynthetic apparatus [27]. Consequently, it is important to evaluate the relationship between the efficiency of PSII and CO 2 assimilation in the leaves as the measurement of fluorescence detects the differences in the response of plants to abiotic stresses by evaluating their tolerance. The use of morphological traits, along with physiological tolerance and their relationship with salinity tolerance indices, are applicable and considered sufficient to be exploited as selection criteria in the breeding of salt-tolerant germplasm [27].
Wheat is also called a salt excluder, which means it mitigates salinity stress by excluding Na + from the shoot as much as possible [28,29]. The capacity of bread wheat to exclude Na + is much stronger than that of durum wheat genotypes [10]. Moreover, salt tolerance is a polygenic trait, and its expression can be influenced by genetic, environmental, and physiological factors. In fact, in the same species it is possible to select a salt-tolerant genotype [30], suggesting that this potential may be improved through conventional breeding approaches. Furthermore, little work has been carried out to examine physiological differences in spring and winter genotypes under salt stress. In the present study, we used five different spring wheat and five durum wheat genotypes to test the different responses and adaptations to salt stress. Some of these, such as S-24, were selected for their well-known salt-tolerance; therefore, it was used as reference one [31]. The genotypes used were evaluated for the effect of salinity on some key physiological and morphological traits. The identified genotypes could represent valuable resources for genetic improvement programs to provide greater understanding of plant tolerance to salt stress, supporting agricultural production on salinized soils irrigated with brackish water.
Response of Wheat Genotypes against Different Salinity Levels
Significant differences (p ≤ 0.05) were recorded among the spring wheat genotypes (G), winter wheat genotypes, and the total wheat genotypes ( Table 1) in terms of their interactions (G * S) at different NaCl salt stress levels (S) for ionic, physiological, and seedling growth traits (Table 1). Table 1. Mean square values (p < 0.05) for root length (RL), shoot length (RL), the relative growth rate of root length (RGR-RL), the relative growth rate of shoot length (RGR-SL), leaf Na + and K + concentration (mg g −1 dry wt), K + /Na + ratio, stomatal conductance gs (mol/m 2 s), maximum quantum efficiency of PSII (Fv /Fm ), instantaneous chlorophyll fluorescence (F t ), chlorophyll content index (CCI) of spring wheat genotypes, spring wheat genotypes, and ten wheat genotypes grown under various NaCl stress levels.
Value of % Control
Significant variation (p ≤ 0.05) was observed between the spring wheat and winter wheat for ionic, seedling growth, and physiological traits (Figure 1a-l). The value of % control of ionic, physiological, and growth traits was decreased by increasing the salt stress (Figure 1a-l). Spring wheat showed a maximum value of % control for ionic (Na + , K + and K + /Na + ratio; Figure 1a-c), growth (RGR-SL), and physiological traits (gs, Ft, QY, and CCI; Figure 1i-l) compared with winter wheat.
Value of % Control
Significant variation (p ≤ 0.05) was observed between the spring wheat and winter wheat for ionic, seedling growth, and physiological traits (Figure 1a-l). The value of % control of ionic, physiological, and growth traits was decreased by increasing the salt stress (Figure 1a-l). Spring wheat showed a maximum value of % control for ionic (Na + , K + and K + /Na + ratio; Figure 1a-c), growth (RGR-SL), and physiological traits (gs, Ft, QY, and CCI; Figure 1i-l) compared with winter wheat. Figure 1. Effect of salt stress on winter and spring wheat. Diagrams are based on value of % control (salt-treated/control * 100) of Na + in leaves (a), K + in leaves (b), K + /Na + ratio (c), leaf senescence % (d), shoot length (SL) (e), root length (RL) (f), relative growth rate of root length (RGR-RL) (g), relative growth rate of shoot length (RGR-SL) (h), stomatal conductance (gs) (i), instantaneous chlorophyll fluorescence (Ft; j), quantum yield of photosystem II (Fv′/Fm′; k), and chlorophyll content index (CCI; l).
Leaf Na + and K + Contents
Significant variations in Na + and K + content were identified between the leaves of genotypes treated with salinity stress (Table 1; Figure 2a,b). Na + concentration in leaves was increased in all wheat genotypes by increasing the salinity stress (Figure 2a,b). At 100 mol/L salinity stress, the minimum Na + content in leaves was found in W4909 (8.51 mg g −1 DW; SW) followed by S-24 (8.59 mg g −1 DW; SW), W4910 (9.09 mg g −1 DW; SW), and Sakha 8 (11.03 mg g −1 DW; SW), respectively ( Figure 2a). Genotypes W4910 (11.40 mg g −1 DW; SW), W4909 (15.48 mg g −1 DW; SW), Sakha 8 (15.58 mg g −1 DW; SW), and S-24 (11.40 mg g −1 DW; SW) performed better at accumulating low leaf Na + at 200 mol/L salt stress and were considered to be salt tolerant. K + concentrations in leaves were decreased in all wheat genotypes by increasing the salinity stress ( Figure 2b). The maximum K + concentration and K + /Na + ratio were observed in Sakha 8 (23.67 mg g −1 DW; SW), followed by S-24 (21. (Figure 2b). Furthermore, the K + /Na + ratio was also decreased by increasing the salt stress in both wheat genotypes (Figure 2c).
Shoot and Root Growth
Significant responses (p < 0.05) were observed for growth-related attributes among the genotypes for salt tolerance. Seedlings SL and RL were decreased by increasing the salt stress in all wheat genotypes (Figure 3a
The senescence of the leaves also increased in wheat genotypes by increasing the salinity stress, as shown in Figure 4c
Physiological Traits
The CCI in all wheat genotypes decreased with increasing levels of salt stress (Figure 4b Chlorophyll fluorescence, e.g., Fv′/Fm′ and Ft values, declined under salt stress. There was a significant difference (p < 0.05) in chlorophyll fluorescence concerning genotypes, salinity level, and the interaction between genotypes × salinity level (Figure 5b,c).
A significant decrease in Ft was observed in all genotypes by increasing the salt stress level (Figure 5b Chlorophyll fluorescence, e.g., Fv /Fm and Ft values, declined under salt stress. There was a significant difference (p < 0.05) in chlorophyll fluorescence concerning genotypes, salinity level, and the interaction between genotypes × salinity level (Figure 5b,c).
A significant decrease in Ft was observed in all genotypes by increasing the salt stress level (Figure 5b
Trait Correlations
A highly significant correlation was observed among the various traits under salt stress ( Table 2). All traits were positively correlated except Na + and the senescence of leaves, which were negatively correlated with all other attributes (Table 2). Na + was positively correlated with the senescence of leaves (Table 2).
Discussion
Salt stress induces a number of negative effects including physiological and biochemical changes in plants which manifest as a reduction in plant biomass and crop yield. Different plants have different tolerance levels, as do most cereals, including wheat [17]. In the present study, five genotypes of winter wheat and five genotypes of summer wheat were considered in order to carry out a comparative study on the salt tolerance of new genotypes, with the exception of the S-24 genotype already extensively studied by Ashraf [31], that is better suited to grow on salinized soils. Overall, the results showed different responses between summer and winter genotypes in terms of physiological traits (Figure 1). Under salinity stress, all genotypes accumulated a higher Na + content in their leaves compared with non-stress conditions (Figure 2a). However, the sodium uptake in leaves was different in spring and winter wheat (Figure 1a). Winter wheat accumulated more sodium com-pared with spring wheat (Figure 1a). Most of the salt-excluder genotypes were previously recognized as salt-tolerant by many scientists [32][33][34], which is further confirmed by this study's results (Figures 1a and 2a). The salt-tolerant genotypes may possess a better ability to maintain low Na + in their leaves, as reported by Elkelish et al. [35]. Saddiq et al. [32] reported that tolerant genotypes preferred to accumulate low Na and high K in their leaves, which was also observed in this study (Figure 1a,b and Figure 2a,b). Munns and Tester [10] reported that the removal of Na + from the cytoplasm into the apoplast is due to the salt-inducible enzyme Na + /H + antiporter located at the plasma membrane. Moreover, Na + accumulation in wheat is controlled by Nax1 and Nax2 genes, located on 2A and 5A chromosomes, respectively [36,37], which are being used as molecular marker cultivars in a breeding program. TNHX1, TNHX2, and TVP1 (vascular Na + /H + antiporter) are responsible for improved seedling shoot growth by generating the pH gradient and facilitating sodium sequestration into values under salt stress [38]. Furthermore, salt-tolerant genotypes could have a sophisticated K + regulation system, such as two-pore K + channels and a shaker type, as described by Shabala and Pottosin [39], and on-selective cation channels, which aid the permeability of K + and transporters (HKT, KUP/HAK/KT, and K + /H + ). An inverse relationship exists between Na + and K + ions due to direct competition for ions in plant absorption [40].
The plants' growth performance was also decreased under salt stress conditions (Figure 1e-g, Figure 3a-c, and Figure 4a). All genotypes had lower shoot lengths and root lengths in salt-stressed conditions compared with controls (Figures 3a-c and 4a). Nevertheless, spring wheat genotypes improved their RGR-SL compared with winter wheat genotypes (Figure 1g). Janmohammadi et al. [41] reported that winter wheat had a lower root length than spring wheat under abiotic stress (e.g., cold stress), ultimately affecting the wheat's growth performance. NaCl stress induced a significant reduction in plant height, root length, and dry weight of roots and shoots in winter wheat [42]. Qiong et al. [43] reported that salinity significantly increased Na accumulation in winter wheat, which significantly reduced shoot dry weight and plant height. Na remarkably reduced the accumulation of K + , K + /Na + ratio, as well soluble proteins and proline. Brestic et al. [44] reported that chlorophyll fluorescence is a more effective method for screening PSII thermostability in winter wheat genotypes. A high concentration of salt in the soil causes water stress, which leads to a significant decrease in the yield of many crops worldwide. Zivcak et al. [45] reported that the photosynthesis efficiency of PSI of winter wheat was decreased by increasing the water stress. Damage caused by salt stress was more prominent at the donor side, rather than the acceptor side of PSII [46]. Munns and Tester [10] reported that the accumulation of Na + at toxic concentrations in the leaf negatively affects the photosynthetic mechanism, resulting in a lower intake of carbohydrates to the young leaf, reducing root and shoot growth. Thus, spring wheat was considered a tolerant crop, with a greater supply of assimilates from leaves to growing parts, e.g., root and shoot length (Figure 3a-b). This might be linked with prolonged retention of chlorophyll in the leaves of spring wheat (Figures 1l and 5b), which could stamp out Na + from leaves, and thereby prevent Na + from reaching toxic levels [41]. Poor performance in terms of growth might be linked with high cell membrane injury and senescence of leaves due to Na + toxicity in growing embryos [4,39,47,48], and this suggestion is supported by the present study (Figures 1d and 4c).
Salt stress has an adverse impact on photosynthesis by destroying chlorophyll pigments and inhibiting the PSII activity. In this study, photosynthesis efficiency declined in both winter and spring wheat genotypes under salt-stress conditions (Figure 5b,c), but winter wheat was more affected (Figure 1j,k). In fact, under saline stress, stomal closing results in a reduction in the photosynthetic rate of the plant. CO 2 assimilation in leaves, the efficiency of PSII, and their relationship allow fluorescence to be used to screen salt-tolerant germplasm against abiotic stresses [49]. Kanwal and his coworkers [26] evaluated the effects of salt stress on newly licensed wheat cultivars using gas exchange parameters and chlorophyll fluorescence. The results reported a smaller reduction in plant biomass in cultivars S-24, Saher-226, and FSD-2008.
Measuring chlorophyll fluorescence is an excellent indicator to quantify salt-induced destruction in the photosynthetic apparatus [50]. Damage to photosystem II has been studied using this technique. Reactive oxygen species (ROS) degrade various proteins (a membrane linker protein, chlorophyll protein) that are necessary for the hooking of phycobilisomes to thylakoids [35,46]. ROS burst destroys thylakoid membranes, resulting in modulations in membrane protein profiles, which leads to decreased activity of the oxygen-evolving complex (OEC) of PS II and increases the working of PS I. Salt-tolerant plants grown under a salt regime downregulate PS II in order to improve the quantum efficiency of excitation energy (Fv /Fm ) [50], as found in this study (Figure 1j,k). The maximum quantum yield of PSII, i.e., Fv /Fm , is an important parameter to discriminate wheat genotypes. Of the different physiological attributes, stomatal conductance and the chlorophyll content index have been reported to be of prime importance in screening crop plants for salt tolerance. Generally, salt stress is known to cause a marked reduction in stomatal conductance and the chlorophyll content index [35,51], as found in this study (Figure 1i,l; Figures 4b and 5a). ROS are regarded as the main source of structural damages under abiotic stresses such as drought, salinity, and heat [52]. ROS are highly cytotoxic and can seriously react with vital biomolecules such as lipids, proteins, nucleic acid, and disturb normal metabolic pathways [52,53]. It has been well documented that during salinity stress, somatically stressed plants reduced CO 2 assimilation due to the closing of stomatal pores, which generate ROS in the plant leaves [54]. In this study, spring wheat exhibited comparatively lower reductions in stomatal conductance and chlorophyll content index compared with winter wheat under salt stress (Figure 1i,l). The reduction in winter wheat might have been due to lower root water potential and the transport of plant hormone ABA from the root into different plant organs, thereby inducing stomatal closure [55]. Compared with spring wheat, winter wheat was more affected (Figure 1), which might be strongly linked with high Na levels in their leaves. The toxic concentration of Na + in leaves encourages the reduction of stomatal conductance in wheat by limiting photosynthesis efficiency [32,47]. Abiotic stress conditions caused by exposure to salinity, drought, heat, and waterlogging cause the stressed plant to produce ROS. The plant also produces antioxidants, flavonoids, and secondary metabolites that detoxify the ROS, thus protecting the plant from abnormal conditions, i.e., abiotic stress [52,53,56]. Therefore, tolerant genotypes prefer to accumulate high K + instead of Na + [57]. In this study, the influx of K + was higher in spring wheat compared with winter wheat, helping to mitigate the salinity stress. Over time, salinity causes Na + toxicity in leaves [10]. Therefore, controlling the transport of Na + in the plant through the exclusion of Na + from mesophyll cells is an important and reliable trait used to improve the salinity tolerance in many crops, i.e., durum wheat [58] and bread wheat [32,59].
Germplasm Collection
Seeds of five winter and five spring wheat genotypes (Table 3) were obtained from the USDA-ARS National Small Grains Collection, Aberdeen, ID, USA.
Hydroponic Culture
A germplasm nursery (5 spring and 5 winter wheat genotypes) was raised in November 2015 in a growth chamber by sowing 50 seeds in 8 cm × 6 cm sand-filled polythene bags at 50% relative humidity and a light intensity of 400 mol m −2 s −1 . Plants were grown with a 14-h day length and with a 20 • C/17 • C day/night temperature cycle. Fifteen plants per genotype, replicated three times, were transplanted at the two-leaf stage into hydroponic tubs filled with 50 L of aerated half-strength Hoagland solution, which was changed fortnightly [60]. Seedling root length and shoot length were also recorded before being transplanted. The experimental design was a completely randomized design (CRD) factorial with three replications. Subsequently, commercial-grade salt was added in 50 mol/L increments twice daily to create different NaCl salt stress levels (0, 100, and 200 mol/L) to avoid osmotic shock.
Determination of Leaf Na + and K + Concentrations
After applying the salt in hydroponic culture, the expanded leaves that emerged under stress conditions were collected and put into the oven for drying. Leaf dry weight was determined. Dried leaves were put into falcon tubes filled with 25 mL of 1% HNO 3 solution for digestion on a hot plate at 85 • C for 4 h. One milliliter was taken from the digested solution, and a volume of 10 mL was prepared to measure the K + and Na + concentration in the leaf samples using a flame photometer (Sherwood, U. K., Model 360) [58,61].
Morphological Traits
After 10 days in a saline environment, the performance of seedlings was assessed based on morphological traits such as seedling root length (RL), shoot length (SL), the relative growth rate of root length (RGR-RL), and the relative growth rate of shoot length (RGR-SL). The relative growth rate was calculated using the formula of Gardener et al. [62].
Chlorophyll Index and Stomatal Conductance
From the seedlings in a saline environment, the topmost fully expanded leaf was used to determine the chlorophyll index using a chlorophyll meter (Model Spad-502) [63]. Stomatal conductance was measured using a leaf photometer (Model Sc −1 ).
Chlorophyll Fluorescence
The data for the chlorophyll fluorescence were recorded based on Baker [64] and Krame et al. [65] nomenclature. Chlorophyll fluorescence parameters, i.e., instantaneous chlorophyll fluorescence (Ft) and quantum yield of photosystem II (QY), were recorded by using the portable fluorescence meter, FluorPen FP 110 (Photon systems instruments, Czech Republic). The FluorPen FP 110 was equipped with a blue LED emitter (470 nm) optically filtered and precisely focused on delivering light intensities of up to 3000 µmol m −2 s −1 to measure plant tissues. QY is a measure of the Photosystem II efficiency. QY is equivalent to Fv /Fm and F 0 is equivalent to Ft in a light-adapted leaf. Quantum yield of PSII (Fv /Fm ) was calculated as
Leaf Senescence
Three random plants in each treatment were tagged. At harvesting time, the total number of leaves and the number of green and senesced leaves were counted. A leaf was considered senesced if less than half of its area remained green.
Statistical Analysis
Quantitative observations of experiments were uploaded in SAS 9.4 (Texas A&M University, College Station, TX, USA) software to deduce the results in the form of variance analysis (ANOVA) for spring wheat genotypes, winter wheat genotypes, and all wheat genotypes. Data are presented in Table 1 with critical values to compare treatment means using the LSD test at the 5% probability level. The Statistix 8.1 package was also used to find correlations among the spring wheat genotypes, winter wheat genotypes, and all wheat genotypes for various growth, ionic, and physiological attributes ( Table 2).
Conclusions
In this study, physiological comparisons of wheat genotypes under salt regimes Sakha 8, S-24, W4909, and W4910 performed better compared with PI 94341, TX12M 4713, and TX12M 4637, depicted by improved seedling growth, CCI, which was linked with better physiology traits, i.e., Fv /Fm ; Ft and gs due to preferential K + uptake and translocation to leaves. The identified plant material can be a source for more deeper insight into determining the genes responsible for enhanced salt tolerance in wheat.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-07-26T00:05:46.787Z", "created": "2021-06-11T00:00:00.000", "id": "236291156", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.mdpi.com/2073-4395/11/6/1193/pdf?version=1623723205", "pdf_hash": "ba90bb3055a7f118336da42a2fbf2c77ab6bfa76", "pdf_src": "ScienceParsePlus", "provenance": "peS2o-0075.json.gz:37952", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "2ed4905c0c83e83b28c008ecebbef934387eff9b", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effect of Moringa olifera Leaf Extract on Growth and Productivity of Three Cereal Forages
Moringa olifera leaf extract at different concentrations were used to evaluate their effects on three cereal forages (Sorghum bicolor L. Moench, Penisetum typhoideum Rich and Sorghum Sudanese) grown under stress environment of soil and water salinity in an arid environment. Three independent experiments carried out at King Abdulaziz University Experimental Station, Makkah province during 2015 and 2016. Treatments consisted of four concentrations of Moringa olifera leaf extract: (C1 = 1 ml of juice + 10 ml of distilled water ,C2 = 1 ml of juice + 20 ml of distilled water, C3 = 1 ml of juice + 30 ml of distilled water and C4 = 1 ml of juice + 40 ml of distilled water in addition to distilled water alone as a control). Results showed that the higher concentration C1 contained the highest amount of inorganic elements and growth hormones compared to other concentrations. This in turn reflected in significant higher growth and forage yields of the three forage crops tested. Significant increases in growth and forage yields in both seasons reported for the higher concentration C1 compared to others. Fresh and dry forage yields of Millet, during 2016, increased by the higher concentration over the control treatment by 17.67 and 4.87%, respectively. Results indicated the effectiveness of Moringa leaf extract in improving growth and increasing productivity of cereal forages under harsh environment of salinity and aridity.
The origin of Moringa olifera tree is the Hamalanian Mountains in the Indian Continent (Osman & Abohassan, 2015). The geographic distribution of Moringa extends from Asia, India, Arabian Pensiula, and Africa to South and Central America. The tree is rich in its nutritive values for human and animals as well as containing many medicinal and chemical substances for other uses to be called the miracle tree (Osman & Abuhassan, 2015). This feature coupled with its geographical distribution gives the tree a special importance as its growth cover places that are highly populated with high poverty rates (Osman & Abohassan, 2015). Among the different usages of Moringa is the leaf extract as a growth hormone promotor for many crops (Price, 2007;Muhamman et al., 2013;Amirigbal et al., 2014).
The use of Moringa leaf extracts as a possible plant enhancer can provide a relatively environmentally friendly, easily accessible and affordable means of increasing crop yields to meet the growing demand for food all over the world, considering the global increasing population rate with its threatening hunger waves. Extracts from fresh Moringa leaves could be used to produce an effective plant growth promoter increasing yield by 25-30% for nearly any crop (Price, 2007). Amirigbal et al. (2014) reported that three sprays of Moringa and brassica leaf extracts significantly affected plant height, number of branches per plant, number of pods per plant, number of seeds per pod seed yield as well as biological yield of canola as compared to their sole application.crops: Mvumi et al. (2012) and Emongor (2015) on onion and kidney beans, Muhamman et al. (2013) on tomato, Abdalla (2013) on rocket, andMohammed et al. (2013) on onions.
The significance of Moringa leaf juice lies in the Zeatin. Zeatin is one of the phytohormones that are the major components of oil and protein structure (Mir et al., 2009). Zeatin is a part of the compound cytokinins, which stimulate cell division, growing cell tissue, delay the process of senescence and ageing in plant tissue and promote nutrient partitioning and uptake (Emongor, 2002;Andrews, 2006;Taiz & Zieger, 2002).
Traditionally, there are five groups of growth regulators including auxins, gibberellins, abscisic acid, ethylene and cytokinins (Prosecus, 2006). Cytokinins enhance food production, and Zeatin is one of the most common forms of naturally occurring cytokinins in plants. Moringa leaves gathered from various parts of the world were found to have high Zeatin concentrations of between five and 200 μg/g of leaves (Alawady, 2003).
Cereal crops (grain and forages) are important sources of carbohydrates and fibers for humans and animals. Their production, area wise and quantities, exceeded any other agronomic crops. For people living in rural areas and in most developing countries enhancing crop production via high cost inorganic fertilizers are beyond their financial capabilities. Therefore, the objective of this study was to look into alternatives that are cost effective, available and environmentally friendly. All these features are strongly true in Moringa trees, and considering its wide spread natural growth in Africa, Asia, South America and many other parts of the world, which makes it a higher candidate for such research. Yasmeen et al. (2013) reported that Moringa leaf extract when it applied for drought or salt stressed plants modified plant phenotypic response to positively affect growth and productivity with alteration in metabolic processes. Moreover, environmental stresses of soil and water salinities under arid lands, where this research was conducted, was another objective as to evaluate Moringa leaf extracts on performance of the tested crops under such environment.
Materials and Methods
Moringa trees, from which leaves were collected for the extraction of juice, were seven years old growing in the same experimental farm in which the field experiments were conducted.
Moringa Leaf Extract
Moringa olifera leaf extract was prepared after collection of fresh green leaves using a conventional electric mixer, afterwards The mixture was passed through a cloth sieve to separate the juice from the residue, The juice was collected into 500 liter flask and four different concentrations were then prepared as follows: 1) 1 ml from juice mixed with 10 ml of distilled water (1:10 by volume) -(C1); 2) 1 ml of juice mix with 20 ml of distilled water (1:20) -(C2); 3) 1 ml of juice mix with 30 ml of distilled water (1:30) -(C3); 4) 1 ml of juice mix with 40 ml of distilled water (1:40) -(C4); 5) Distilled water only as a control -(C5).
These four concentrations in addition to the control sprayed on three cereal crops at a rate of 25 ml/plant at an interval of two weeks following emergence. The cereal crops were Sudan grass, forage sorghum, and millet, and each crop was planted as an independent field experiment.
A randomized complete block design with three replicates was used for each experiment; each replicate contained five randomly assigned treatments with plot size (experimental unit) of 1 × 1 m for each treatment. The three forage crops were planted in continuous rows 80-cm apart and watered via perforated plastic pipes along the row. Borehole was the source of irrigation water that contained 3000 TDS (ppm) and the soil was a saline one. Measurements of growth parameters were taken at forage harvest. Ten randomly selected plants from middle row were used for each growth parameter measured. For yield parameter, the entire plot (1 × 1 m) was used to estimate forage fresh and dry yields.
The following growth and yield parameters taken during the course of the study, which lasted for two years: plant height, leaf area, number of leaves/plant, stem diameter, fresh and dry yields.
Hormonal Analysis
Juice extract was prepared from green Moringa leaves and green twigs after crushed by a conventional kitchen mixer. The juice separated from residue using a cotton cloth as a filter. The same treatments described above were used for hormonal analysis: Afterwards each concentration (C1 to C5) mixed with cold redistilled 95% ethanol, kept in a dark bottle, and deep-freeze overnight before assay started. The fraction of the ethanol extract was carried out according to the method described by Wasfy et al. (1974). The Acidic fraction contain the acidic hormones (IAA, GA, and ABA) while the aqueous fraction comprised the cytokinins. The growth promoters (auxins, gibberellins and cytokinins) and the growth inhibitors ABA quantified using high performance liquid chromatography (HPLC) according to the method of Muller and Hilgenberg (1986).
Inorganic Minerals
Mineral elements extracted from leaf tissues according to Chapman and Pratt (1961), Phosphorus determined according to the method described by Humphries (1956) while Potassium determined photometrically according to Williams and Twine (1961), Calcium and Magnesium determined by atomic absorption spectrophotometer according to AOAC (1984) and total nitrogen was determined by Micro-kejeldhal, Tector model 1026 after digestion in sulphuric acid (Horwiz, 2002).
Soil and Water Analysis of the Experimental Site
Soil chemical analysis of the experimental site, in top 30 cm depth, showed a soil pH of 8.25, an Ec of 1.61 ds -1 , OM of 0.1% and N, P, K, Ca, Mg, and Na of 0.32, 0.129, 2.5, 3.6, 6.3 and 16.8 mg kg -1 , respectively. Irrigation water (which was a borehole water), analysis contained 3000 TDS (ppm) with Nacl the dominant salt.
Results
Inorganic contents of leaf extract are presented in Table 1. The concentration of all inorganic contents dropped with dilution of the juice. The control treatment (C5) contained negligible amounts of the inorganic compounds, as it was distilled water. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water.
Similar to inorganic contents, the hormonal concentration dropped with increase in dilution rate, while no hormones in the control treatment (distilled water) as shown in Table 2. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water.
Effect of Treatments on Growth and Yield of Sudan Grass
Results of different concentrations of Moringa leaf extract on Sudan grass are presented in Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
The effect of different concentrations of Moringa juice on fresh and dry yields of Sudan grass is presented in Table 4. Significant differences in fresh and dry yields of Sudan grass were reported for the different concentrations in comparison to the control (C5) in both seasons. The higher concentration C1 always recorded higher fresh and dry yields in both seasons, whereas the control C5 resulted in the lowest fresh and dry yield. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
Effect of Treatments on Growth and Yield of Forage Sorghum
Results of the effect of Moringa Juice extract on growth parameters of forage sorghum are presented in Table 5. Moringa leaf juice extract had a significant effect on growth parameters of forage Sorghum in both seasons except pant height and stem thickness in 2015 and number of leaves per plant in 2016. Taller, thicker and leafy Sorghum forage plants were recorded for the higher concentration C1 compare to other concentrations and the control, which recorded the lowest height, thickness and number of leaves., 2015Mar., 2016Apr., 2015Mar., 2016Apr., 2015Mar., 2016Apr., 2016Mar., 2016Apr., 2015Mar., 2016 C1 125.27a 98.53a 6.12a 0.80a 6.86a 5.66a 5.94a 6.30a 1.94a 1.85a C2 122.76a 93.76b 6.07a 0.76a 6.76a 5.66a 5.83a 6.18b 1.94a 1.83a C3 120.93a 92.91b 6.00a 0.76a 6.55a 5.53a 5.72a 5.83c Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
Effect of treatments on fresh and dry yields of forage Sorghum are presented in Table 6. Significant (P < 0.05) differences on fresh and dry yields, because of Moringa leaf extract, were recorded during both seasons. Similar to Sudan grass results, higher fresh and dry yields were recorded for the higher concentration C1 and the lowest yield was recorded for the control. It is worth mentioning here that higher yields recorded during second season of 2016 compared to 2015 regardless of treatments used. This might be due to weather variations (especially rain and temperature) during both seasons. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
Effect of Treatments on Growth and Yield of Millet
Results of the effect of treatments on growth and yield parameters of millet are presented in Tables 7 and 8, respectively. Significant differences for growth parameters of plant height stem diameter and number of leaves per plant were reported in both seasons except April of 2016 for number of leaves per plant (Table 7). The tallest, thickest and leafy plants recorded for the higher concentration C1 except for plant height in April 2015 and March 2016, when taller plants recorded for the Control C5 and the concentration C2, respectively. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
Effects of treatments on fresh and dry yields of millet are presented in Table 8. Fresh and dry yields of millet were not significantly affected by treatments in April 2015. However, significant differences were recorded for both fresh and dry yields during 2016. Again, variations in weather conditions could be an explanation for this. Fresh and dry yields, during 2016, increased by the higher concentration over the control treatment by 17.67 and 4.87%, respectively. Note. C1 = 1 ml of Moringa juice with 10 ml of distilled water; C2 = 1 ml of Moringa juice with 20 ml of distilled water; C3 = 1 ml of Moringa juice with 30 ml of distilled water; C4 = 1 ml of Moringa juice with 40 ml of distilled water; C5 = only distilled water); Values with same letters are not significantly different from each other; LSD = least significant difference; CV = coefficient of variation.
Effect of Moringa Leaf Extract on Growth Parameters
Chemical and hormonal analysis of Moringa leaf extract clearly showed that the higher concentration treatment (C1) revealed the highest inorganic contents compared to other concentrations. Furthermore, the higher concentration (C1) showed the higher concentration of the hormones, especially cytokinins, compared to other concentrations. Consequently, it expected to affect growth attributes of plant height. Stem diameter and number of leaves in a positive way. Moringa leaf juice is rich with growth hormones, especially Zeatin, that has been reported to increase the crop yield in the range of 10 to 45% (Muhammad, 2014). Moringa leaf juice also contains micronutrients in sufficient amounts and suitable proportions that increase the growth and yield of a variety of crops ranging from cereals to oil crops, from fiber to sugar crops and from forages to tuber crops (Price, 2007;Muhamman et al., 2013;Amirigbal et al., 2014). Rehman et al. (2017) reported that Moringa leaf extract when applied to wheat plants increased plant height, number of tillers, increased grain yield and delayed leaf senescence. They related that to Moringa leaf extract being rich in Zeatin, a cytokinins maintained the green photosynthetic area, therefore contributed to higher grain yield. It should be recalled that the three cereals forages were grown under stress environment of water and soil salinities. Yasmeen et al. (2013) reported that Moringa leaf extract when applied on for drought or salt stressed plants modified plant phenotypic response positively affect growth and productivity with alteration in metabolic processes
Effect of Moringa Leaf Extract on Yield
Forage yield largely determined by growth attributes of plant height, stem diameter, size, and number of leaves carried by the plant, that resemble the resultant of yield. As shown from the results, forage yield increased significantly by the higher concentration (C1) of Moringa Juice extracts in both seasons compared to other concentrations. Several researchers found similar results with different crops: Mvumi et al. (2012) and Emongor (2015) on onion and kidney beans, Muhamman et al. (2013) on tomato, Abdalla (2013) on rocket, andMohammed et al. (2013) on onions. As cytokinins are considered to be regulators of leaf senescence (Davis, 2007), therefore we hypothesized that with rich in Zeatin type of cytokinins and other regulators (Rady & Mohamed, 2015), Moringa leaf extract can play a role to maintain photosynthetic area by delaying senescence and affecting source-sink strength to increase yield.
Conclusion
It can be concluded from the results of this study that Marina leaf extract without or with little dilution can increase growth and productivity of cereal forages grown in stressed environment. The spread of this tree in the Southern Hemisphere characterized by human explosions, poverty and aridity gives the tree special importance to be called the miracle tree. We recommend that under aridity, where salt stress prevails, use of Moringa leaf extract be used to replace inorganic expensive and environmentally polluting fertilizers.
Table 1 .
Inorganic contents of Moringa (mg/kg dry weight)
Table 2 .
Hormonal contents of Moringa extract (mg/kg fresh weight) Table 3 for both seasons 2015 (April) and 2016 (February and April). All growth parameters of Sudan grass were significantly (P < 0.05) affected by the different concentrations of the Moringa Juice extract in both seasons, with exception of plant height in 2015 and stem thickness in 2016. Taller, thicker and leafy Sudan grass plants were recorded for the higher concentration C1 compared to other concentrations. Table3. Effect of different concentrations of Moringa leaf extract on growth parameters of Sudan grass
Table 4 .
Effect of different concentrations of Moringa leaf extract on fresh and dry yields Sudan grass
Table 5 .
Effect of different concentrations of Moringa leaf extract on Moringa leaf extracts growth parameters of forage Sorghum
Table 6 .
Effect of different concentrations of Moringa leaf extract on fresh and dry yields of forage Sorghum
Table 7 .
Effect of different concentrations of Moringa leaf extract on growth parameters of millet
Table 8 .
Effect of different concentrations of Moringa leaf extract on forage fresh and dry yields of millet
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-05T19:54:32.335Z", "created": "2017-06-07T00:00:00.000", "id": "55122061", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.ccsenet.org/journal/index.php/jas/article/download/67834/37421", "pdf_hash": "dec39afce65240aa28463e744bfdcf11283f5dc5", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:30690", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "dec39afce65240aa28463e744bfdcf11283f5dc5", "year": 2017}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
TRUE SHALLOT SEED PRODUCTION OF LOWLAND SHALLOT (BIRU LANCOR VARIETIES) UNDER THE APPLICATION OF SEAWEED EXTRACT AND N FERTILIZER
The major problem on the improving shallot production is due to availability of quality tuber seeds requied by farmers in sufficient quantities. One effort that can be done in order to fulfill the seed production is to use TSS or True Seed of Shallot technology. For this reason, efforts are needed to improve nitrogen fertilizer efficiency following the application of seaweed extract along with different source of N fertilization. The research was carried out in the village of Sidomulyo, Batu, with an altitude of 923 m asl with planting material from seed growers in Probolinggo who have experienced on shalor seed production with seed a dormancy period of approximately 2 months. Seaweed Extract (RL) uses Citorin and ammonium nitrate using Calcium Ammonium Nitrate Fertilizer. For seeds to be planted, vernalization is carried out by inserting seeds into the refrigerator at ± 10C for 3-4 weeks. The study began in September 2018 and ends in January 2019. The study used a Factorial Randomized Group Design of 1 factor, namely the dose of Ammonium Nitrate with several concentrations of seaweed extract (RL). The dose of Ammonium Nitrate consists of 0% (0 kg / ha), 50% (from the total N dose of 178 kg / ha and 100% (as much as 178 kg / ha). The dosage of N 178 kg per hectare comes from the calculation of N content in doses recommendation is NPK 600 kg / ha (NPK fertilizer contains 15% N), ZA 200 kg / ha (ZA fertilizer contains N 26%), Urea 100 kg/ha (urea fertilizer containing N 46%), and KCl 150 kg/ha Seaweed extract with a concentration of 0 ppm, 120 ppm, 240 ppm, 360 ppm and 480 ppm. Thus the treatment is as follows: (1) Control = 0% N and 0 ppm RL; (2) N1 RL 120 ppm (50% dose N and 120 ppm RL); (3) N1 RL 240 ppm (50% N and 240 ppm RL); (4) N1 RL 360 ppm (50% dose N and 360 ppm RL); (5) N1 RL 480 ppm (50% dose N and 480 ppm RL); (6) N2 RL 120 ppm (100% N and 120 ppm RL (7) N2 RL 240 ppm (100% N and 240 ppm RL (8) N2 RL 360 ppm (100% N and 360 ppm RL); (9) N2 RL 480 ppm (100% dose N and 480 ppm RL). The variables observed were plant height, number of leaves, number of tillers and number of tubers per plant, 50% bloom time, number and height of stem, root dry weight, canopy dry weight, number of flowers per stem, number of kernels per stem, percentage of flowers being kernels, weight of 1 seed, seed weight per 1000 m, mini tuber production. Data from the observations were analyzed using ANOVA and if there was a significant effect followed by LSD at the level of 5% to see the differences between treatments.
105.000 ton for fulfilling exports. When the average productivity of shallots is projected to reach 10.22 tons ha -1 , then in 2017 there will be around 121.749 ha of harvested area. Referring to the 2012 harvest area, which was equal to 99.519 ha, the fulfillment of the demand for shallots in 2017 requires an additional expansion of the harvested area of around 22.230 ha. An additional area of 22.230 ha requires additional tuber seeds (1.5 ton ha -1 ) of 33.345 tons so that the total need for 2017 tuber seeds should be in the range of 138.245 ton. However, the quality tuber seeds needed by farmers are not sufficient in quantity and those availability at the market, due to achieving rapid grow of the shallot production it is required to optimize seed tuber production Rosliani et al., 2016).
The availability of tuber seeds is predicted to only be able to meet 75.87%. If the tuber demand are replaced with TSS (True Shallot Seed) (5 kg/ha) then the requrement for this in 2017 is 111 ton (processed BPS data, 2014). Red shallot (Allium ascalonicum L.) is one of the important vegetable species which is a national superior commodity (Rosliani et al., 2013;Fritsch and Fiesen, 2002;Sopha et al., 2014). As a result, farmers meet the needs of seeds by producing their own seeds. The use of seeds themselves which is carried out continuously can cause a decrease in productivity and is prone to root tuber diseases such as Fusarium and Colletotrichum (Sumarni and Rosliani, 2010;Rosliani, 2013). Some other problems faced in the production and use of tuber seeds are more expensive especially when the tuber seed stock is limited, requires greater storage space, higher transportation costs due to volume, and tuber seed production ratio is lower than seed production ratio. The average tuber production ratio is 1: 10 while the seed production ratio can reach 1: 200. The onion seed dormancy period is 2-3 months and if stored for a longer period (> 3 months) there will be a decrease in quality. The advantage of using tuber seeds is that they do not require pollination and complicated technology to replant because the bulb size is quite large (Sumarni and Rosliani, 2010;Rosliani, 2013).
In order to meet the demand of shallot seeds, use the TSS method or botanical seeds. TSS seeds that are an alternative to tuber seeds have several advantages, among others, are healthier seeds, have a higher seed production ratio than tuber production and have a longer dormancy period of more than 2 years . TSS production still faces several challenges, among others, the percentage of flowering and seed formation is still low. Therefore the production of red onion TSS is a very interesting study material because TSS can be an alternative to meet the demand of seeds at the farmer level, thus opening opportunities in efforts to increase national shallot production. Khokar (2014) stated that flowering stimulation in shallots in TSS seed production is influenced by many factors starting from the vernalization process, bulb size, environmental conditions after planting and varieties.. Biostimulants are known as ingredients and/or microorganisms that can increase absorption of nutrient absorption by rooting plants, especially nitrogen. At present the use of biostimulant seaweed extract on shallots is still limited. Farmers generally use GA 3 . Seaweed extract has several advantages compared to GA 3 because in addition to containing gibberelin there is also a cytokinin which, among others, functions to accelerate flowering and increase uniformity of flowering time in addition to its function in increasing fertilizer efficiency. Research on the nutrient requirements of Nitrogen, especially Ammonium Nitrate (NH 4 NO 3 ) in TSS production is still not widely used. The aim of the study was to study the potential of ammonium nitrate and seaweed nitrogen in shallot seed production.
The Ministry of Agriculture has issued a package of TSS production technology consisting of components of site selection technology, vernalization, fertilization, and the application of GA 3 and BAP biostimulants The solution to increase the percentage of seed formation is by using the vernalization technique. Regarding the vernalization technology component, research that looks at the mechanism of shallot seed production through vernalization techniques is still very limited (Wu et al., 2016). The use of 10 o C vernalization temperatures in the highlands in Bima varieties has provided information that TSS production can reach 8.12 grams per 12 plants . Besides that the fertilizer component is an important strategy in increasing shallot production. However, irrational use of N and carried out continuously will cause an imbalance of nutrients in the soil and degradation of soil fertility which has an impact on decreasing yields. Biostimulant applications such as the application of Growth Regulating Substances (ZPT) and growth stimulants such as Benzyl Amino Purine (BAP) and Giberellic Acid (GA 3 ) are also important components that influence TSS production. It's just that a number of previous studies have shown that the effect of the application of GA 3 growth regulators is still varied or unstable, increasing the productivity of shallots. The results of the study showed no interaction between varieties (Pancasona and Mentes) and the way GA 3 was applied to plant growth, flowering and TSS yield of shallots.
In view of this, the improvement of TSS seed production technology components is a very interesting study material, among others studies on tuber vernalization aspects (storage of tubers at low temperatures), application of N sources in the form of ammonium, nitrate and ammonium nitrate for N availability evaluation. the use of biostimulant using alternative ingredients such as seaweed extract (Euchema spinossum) so that it can be seen the effectiveness in increasing productivity of the results of onion TSS. The objective of this study was to evaluate the response of shallot varieties to the type and dose of biostimulant and the dosage and source of N fertilizer as a component of TSS production technology in increasing the speed of flowering initiation and yield of TSS seed varieties of specific location, namely Blue Lancor variety.
METHODS OF RESEARCH
The study was conducted in Sidomulyo Village, Batu City with an altitude of 923 m above sea level. The study was performed in a plastic house (greenhouse) starting at September 2018 to January 2019. This location was approximately 30 miles from the central city of Malang, with geograhic position at 7 o 51'12.29" S and 112 o 31'25.62" E.
The research will be carried out using onion seeds which previously have been vernalized at 10 o C for 2-4 weeks. Plastic houses (greenhouses) are made with an area of approximately 150-300 m 2 . Silver black (SB) plastic mulch is used to cover beds, biostimulants in the form of seaweed extract (using Citorin), Calcium Ammonium Nitrate fertilizer as a source of Ammonium Nitrate, SP-36 as a source of P and KCl for K. N,P, and K fertilizers are given in 2 periods, starting from the planting period to optimal vegetative growth and the period of flowering formation to harvest. Fertilizer is being applied according to the recommendation level, which were equal to 600 kg ha -1 NPK, 200 kg ha -1 Urea and 100 kg ha -1 ZA (Ministry of Agriculture, 2014). The second period, fertilizer was given in the form of Ammonium nitrate, SP-36 and KCl equivalent to a dose of 100 kg ha -1 NPK and boron as much as 3 kg ha -1 . Pesticides was used for controlling plant pest organisms, manure or petroganics were applied for fulfilling basic fertilizers. Boron fertilizer as much as 3 kg ha -1 and additional SP-36, KCl, and Ammonium nitrate fertilizer as much as 100 kg ha -1 to support flower growth. The supporting equipment for research were used including ruler, raffia, measuring cup, micro pipette, analytic scales, minimum-maximum thermometer, hand sprayer and oven, pollinator insects that help pollinate and tray processes for harvesting and bamboo stick to support flower stalks (belalo) during flowering to seed harvest.
The study was used Randomized Complete Block Design (RCBD), whereas the application of Nitrogen Ammonium Nitrate under the combination of Seaweed Extract (RL) and this were repeated 3 times. The list of treatments is given in Table 1. The use of UV plastic shade on plastic housing or greenhouse was carried out during the research in an effort to minimize crop failure. The structure UV plastic shade uses a bamboo frame and the shape of a semi-circular roof with a circle peak height of approximately 4 meters and a height of the right and left side of the shade of 2 meters. Around the house the plastic is installed with paranet to control pest and disease outbreak.
The study used experimental plots in the form of beds with a size of approximately 2 m x 1 m. The distance between beds in the replication is 30 cm and the distance between beds between replications is 50 cm. The land is processed perfectly by hoeing and leveling the soil surface. After that, manure is given at a dose of 20 tons per hectare or approximately 0.5 kg per bed.
Silver black plastic mulch is installed along the beds by installing wooden pegs around the beds. Installation of mulch is done a week before planting. Mulch used is with a width of 120 cm plastic size x length of the bed.
Planting holes are made by heating iron rings by burning wood charcoal on top. This tool is specifically designed for making planting holes above plastic mulch. Making a planting hole by observing the spacing of 15 x 20 cm, which is 15 cm between lines and 20 cm in distance so that there are approximately 5 x 10 plants = 50 plants per bed.
Planting is done on seeds that have been vernalized for 2-4 weeks at 10 o C. Before being planted, first the seed bulbs are sprayed with a 2 ml per liter biostimulant solution according to the concentration of the treatment then dried. After that the bulb seeds are cut 1/3 the end for the bulbs ready to be planted. Planting by making a planting hole 1-2 cm deep and the tubers covered with a mixture of soil and manure.
Watering is done twice a day in the morning before sunrise and evening or according to land conditions and weather in the field. Watering uses a stringy so that it does not damage the tuber seeds and the seeds are not thrown from the planting hole.
In the rainy season, it is estimated that more diseases attack the onion plants compared to pests. The main disease is fusarium which can cause plants to not grow normally and must be eradicated so as not to spread to other plants around it. Pest and disease control is controlled by the application of pesticides as recommended. To prevent severe attacks, monitoring is carried out every day morning and evening together with watering activities.
Fertilization with a distance of 5 cm from the base of the plant. After that watering is done. P and K fertilizers are in accordance with the recommended dosage which is equivalent to 600 kg NPK ha -1 , 200 kg ha -1 Urea, and 100 kg ha -1 KCl. Fertilization is done twice, namely when the plants are 15 and 30 days after planting. When the plants are out, SP-36, KCl, and Ammonium nitrate fertilizers are added at a dose of 100 kg / ha for plant maintenance in supplying the nutritional needs of plants starting from flowering to harvesting seeds. The time of application of additional fertilizers according to the conditions of the plants in the field, carried out up to 5 times at 10-day intervals and added boron with a dose of 3 kg ha -1 to help improve the ability of plants in seed formation.
Mounting the support to support the erection of the umbel or the main flower stem that appears. In 1 clump of plants usually grow 2-3 tubers. If no attachment is installed, the umbel will break or collapse and die. Installation is done when the plants are 35-40 days after planting.
Biostimulant is applied when the plant is approaching the tuber formation phase and the flower has not yet come out. This phase is estimated when the plants are 25-45 days after planting. Biostimulant application 3 times at 7 days interval. The spray volume is 300 liters per hectare so that the volume of spray per plant is 1.2 milliliters.
Insect pollinator intervention to help pollinate the shallot seed production. Insect pollinators use the intervention of green flies (Calliphora vomitoria) Harvesting of seeds is done when plants start 88 days after planting and are harvested three times until the plants are 99 days after planting with harvest intervals 3-4 days depending on plant conditions and weather on the land. Mini bulbs are tubers produced by shallots after the seeds are harvested.
Measurement were made in the vegetative phase, generative planting. Variables and time of observation are as follows: Observations were made by calculating the number of flower stalks that appeared with flowers that had been fully bloomed in all plants in each treatment plot. Observations were made at 63 HST.
It is done by calculating the shallot seed weight produced. Observations were made by weighing the seeds produced in all kernels in each treatment plot. Observations are made at harvest time, after finishing processing the seeds. The stages are as follows: the harvested flowers are dried, separated from the stover, and manually extracted seeds in the kernel, then weighed using a scale.
Observations were carried out by weighing the harvested tubers which at the same time as the shallot seed harvest, collecting from the tubers weight produced per plot of the treatments.
Root dry weight and above ground biomass dry weight were carried out by destructive sampling and weighing the harvested their fresh weight before placing into the oven overnight at 60 o C to derive dry weight for the next following days measurtement.
The data obtained were analyzed by ANOVA and if there was a significant effect followed by Fisher LSD analysis (5%) to find out the differences between treatments.
RESULTS OF STUDY
The results the analysis of variance showed that the treatment had a significant effect on time of flower with 75% of flower in full blooming period and stalk height per plant (P<0.05) but not significantly effected on the number of stalk observed at 63 DAP (days after planted) ( Table 2). The growing time required for flower being appeared with a 75% of flower in full blooming period at control treatment was 50.67 DAP, which was not significantly different to those N1 treatment at all RL concentrations (120, 240, 360 and 480 ppm), except for N2 treatment. This meant that the application of higher dose of 178 kg ha -1 N-NH 4 . NO 3 (N2) exaggerating time for shallot for producing flower slower than low dose application In addition, the treatment of control was not significantly different to all N1 treatment (lower dose of N fertilizer).
In term of flower stalk height, the greatest was obtained from the treatment of N1 RL 360 ppm, which is not significantly different to N1 RL 480 ppm and N2 RL 240 ppm treatments. In comparison to control, generally the height of the stalk at the N1 dose treatment increased with the addition of RL concentrations except at 480 ppm but on the contrary, at the higher N2 dose the stalk height decreased with increasing RL concentration so that N2 480 ppm treatment produced the lowest stalk height. The treatment of N1 RL 240 ppm producing the lowest height of stalk which was not significantly diiferent to N1 RL 120 ppm, N2 RL 120 ppm and N2 RL 480 ppm.
The results of the variance analysis showed that the treatment had no effect on the variable number of flowers per flower and the percentage of flowers into capsules (P<0.05) but it was significantly affected the variable number of capsules per stalk (Table 3). The highest number of capsules per stalk was found in the treatment of N2 RL 360 ppm as many as 16.69 capsules per stalks in which it was significantly different to controls which only able to produce capsules as much as 6.60 capsules per stalks. However there was no significantly different on the number of capsules per stalk between N2 RL 360 ppm and lower application of RL (N2 RL 240 ppm) or even higher (N2 RL 480 ppm). The average number of capsules per stalk at lower N fertilizer (N1) under various concentrations of RL was contributed to the increasing of those value by 70.15% compared to controls. In addition, the application of higher dose of N fertilizer (N2) at all concentrations of RL (120, 240, 360 and 480) was resulting in the raising of an average number of capsules per stalk by 133%. The number of capsule per stalk in the treatment of N2 at various concentrations of RL (120, 240, 360, and 480 ppm) was higher than that those treatment of N1. The results of the variance analysis showed that the treatment did not significantly influence the weight per 1 seed but it was significantly affected the seed weight in each 1,000 m 2 and tuber weight (Table 4).
In general, seed weight in all treatment were significantly higher than those of control (P<0.05). The lowest seed weight per 1,000 m 2 is in the control (578.67 grams), whilst the highest was found under the treatment of N2 RL 360 ppm (983.33 grams). Among the treatments, N2 RL 360 ppm produced highest seed weight (983,33 g per 1000 m 2 ), in which those value are almost twice than the control, and it was significantly different to other treatment, before it was drop to 895,67 g per 1000 m 2 at the treatment of N2 RL 480 ppm. The average seed weight in treatment N1 (at all concentrations of RL 120, 240, 360, 480 ppm) was 748.84 g in which it was lower than those average seed weight of N2 (at all concentrations of RL 120, 240, 360, 480 ppm), reached 867.46 g. This means that there was an increasing on seed weight by 29.41% and 49.90% compare to control treatment, respectively.
In contrast, in term of tuber weight, the highest was detected under N1 RL 240 ppm treatments (60 g per 2 m 2 ) which was not significantly different to the treatment of N1 RL 480 ppm, N2 RL 240 ppm and N2 RL 480 ppm, producing tuber weight at 57.67, 56.33 and 57.33 g per 2 m 2 , respectively.
The average tuber weight produced in the N1 treatment at all RL concentrations was 58.00 g while for the N2 treatment in all concentrations produced an average tuber weight at of 53.08 g per 2 m 2 . This acccounrted for the increasing of tuber weight at 81.25% for low fertilizer dose (N1) and 65.88% for high dose N fertilizer (N2) treatment.
The results of analysis of variance showed that the treatment significantly affected the root dry weight and aboveground biomass dry weight P (<0.05) ( Table 5). The lowest root dry weight were found in control (0.09 g/plant) and the highest was treated with N1 RL 480 ppm (0.28 g/plant) which was not significantly different to N1 RL 120 ppm. The higher dose of fertilizer (N2) did not influence root dry weight since there was no significantly different to all RL concentrations (120, 240, 360 and 480 ppm). Generally, the additional of N fertilizer increase root dry weight and above ground biomass at the range of 20 to 100 % in all treatments compare to control, eventhough there were no clear evidence that under higher concentration of N fertilizer (N2) was given better result. The lowest aboveground biomaas weight was detected in the control which is not significantly different to N1 RL 240 ppm, N1 RL 360 ppm, N2 RL 120 ppm and N2 RL 240 ppm. The treatment of low N fertilizer (N1) and high N fertilizer (N2) at all concentrations of RL caused an average increasing in dry weight by 85.80% and 87% respectively. The dry weight of the aboveground biomass has the similar pattern to root dry weight affected by those treatment.
DISCUSSION OF RESULTS
Related to the time the flower stalk to be appeared, Rosliani et al. (2016) have conducted in-depth research related to the flowering phase of shallots for seed production. The results of the study showed that flowering time with 75% of flower in full blooming occurred at 62-66 DAP while in this study the flowering time were found at 50.33 -54.67 DAP, which meant the period was coming earlier. Harvesting time in Rosliani et al. (2016) up to 107 days, whlist on this experiment were reduced to 99 days. This is due to the effect of treatment which adding seaweed extract, therefore the period of flowering was changes. In addition there was also a differences on the locataion, which from the climatic, geographic and also evelation perspective was different. The difference of shallot flowering to be appeared with 75% of flower in full blooming period between low dose fertilizer application (N1) and high dose fertilizer application (N2) was due to a shortage of N in the plant, therefore under N1 or even in control becomes flowering faster. The results of this study in line with Gebretsadik In term of number of capsule per stalk which is the lowest to be found in control is also being affected by the low N suppliey to this treatment. The increasing number of capsule per talk were between 70 to 133 % under the treatment of low application of N fertilizer (N1) and high dose N fertilizr (N2), respectively. This research is in line with (Gustfson, 2010) that the fruit and seed resistance is higher in plants with higher nitrogen doses because nitrogen can increase phospor and potassium uptake coupled with research conducted by Du Jurdin (2015) that the addition of biostimulants increases the effectiveness absorption of nutrients by roots to N, P and K. In this research research, ammonium nitrate was used, where nitrate itself is a common form of nitrogen and influences plant regulation in various aspects of plant development. Nitrates provide nutrients and are reported to affect the growth of seeds, roots and leaves, root architecture, flowering time, branch formation, and plant aging, and affect crop yields (Crawford and Forde, 2002;Guiboileau et al., 2012;Stitt, 1999;Vidal et al., 2014). Meanwhile, flowering is the transition between the vegetative phase and plant reproductive growth so that this flowering period is a critical period in the role of future generations of plant sustainability and is very influential on fertility or the ability to support plant growth (Srikanth and Schmid, 2011).
In this study, flowering and seeding occurred in the rainy season at the end of November 2018. According to Rosliani et al. (2016) that during the rainy season pollinating insect activity decrease when compared to its activities in the dry season, especially in Apis serana and Vespidae insects. The emergence of 75% of flowers in full blooming period of all treatments were 52.85 days which is between December (weeks 3-4), in which by this period it rains frequently, whereas the average amount of rainfall in December is at 4.84 mm. According to Rosliani et al. (2005) the planting time affects flowering and seed production. The dry season is the right time for flowering and onion seed production (Rosliani et al., 2005). Furthermore, as stated by Rosliani et al. (2016) which mentioned that the production of shallots seeds should produced during the dry season because at that time, besides the high activity of pollinating insects, there was also a low attack of pests and diseases. This could be potentially reducing seed yields. On this experiment, the main diseases were fusarium and caterpillar (Agrotis sp.). To anticipate the disease attack, prevention and control of disturbing organisms has been carried out regularly once a week with the application of fungicides and insecticides in accordance with the recommendations. The main pests that arise when plant growth are controlled by the Furadan application. To anticipate rain water exposure at the study site, plastic houses were used because of planting shallots to ensure that the treatment was not washed away by rainwater. Thus, plants can carry out vegetative growth until they succeed in entering the generative period and seed formation.
Instead of genetic and endoegnious factors of the palnt itself such varieties Rikanth and Schmid (2011) stated that flowering time can be influenced by various environmental factors. To increase yield, research needs to be done by under a higher altitude. This study was conducted at an altitude of 923 m above sea level while according to ) that the production of optimal shallot seeds is cultivated at altitudes above 1,000 m above sea level. This is because flowering onions requires a low temperature of 7-12 o C to induce flowering and 12-18 o C to increase the size and time of flowering. The novelties of this study was the succefullnes on producing lower altitude of shallot (Biru Lancor varieties) seeds under various different treatments to dtecet the the different level of N fertilizer ans seaweed extract.
Observation to the weight of 100 seeds is very influential on overall seed yield. The higher the weight of 100 seeds, the higher the yield of seeds produced per unit area. However, there is no significantly effect from all treatments in this tudy. Lack of water during seed filling or seeding phase can reduce yields as a result of reduced seed size (Akil et al., 2007). Adding to this, total seed yields were calculated per 1000 m 2 area. The results of the seeds in this study showed that the average seed yield on N1 was 748,835 g while the average yield on N2 was 867.46 g in which thos value are significantly different compare to control treatment. Average seed yields of seeds per 1000 m 2 can reach 11.529 g (Rosliani et al., 2018), which meant higher than those value compare to the seed yield this study. The differences may due to the difference on shallot varieties and those geographical positions. The effect of the addition of seweed extract as biostimulant successfully increases the seed yield in the area in 1000 m 2 . The results of this study are in line with the research conducted by Du Jurdin (2015) on soybean plants which suggested that biostimulant had a significant effect on the number of crop pods, number of seeds per pod, number of branches, seed harvest. This is due to an increase in the absorption of N, P and K (Rathore, 2015) The observation of tubers weight at 99 DAP showed that the lowest results were in the control (32 grams per plot size 2 m 2 ) which was significantly different from all other treatments. The average tuber weight produced in the N1 treatment at all RL concentrations were at 58.00 grams while the N2 treatment in all concentrations produced a tuber average of 53.08. When compared with the control, tuber weight of N1 at all concentrations increased by 81.25% while in N2 treatment the increase was only 65.88%. The results of this study are in line with the research conducted by Gebretsadik and Dechassa (2018) that the lower the nitrogen, the higher the formation of tubers were produced.
The lowest dry weight of shoots and roots were detected in the control, which is equal to 0.81 g an 0.09 g per plant which were significantly different from all other treatments particularly when it was compared to those of N2 treatment (high application of N fertilizer). When it was compared to controls, the treatment of N1 at all RL concentrations caused an average increasing of dry weight at 85.80% while treatment N2 at all concentrations resulted in an average increase in hihger dry weight at 87.65%. This variable is closely related to plant height which has a pattern that is almost the same where the treatment of N2 doses produces a higher plant height compared to treatment N1. The dry weight of roots and leaves was measured at tuber harvesting 56 DAP. At the time of harvest, root and canopy formation is under maximum condition. Bertoni (1992) states that the dry weight of plant roots increases rapidly until the beginning of tuber formation and then slows down during the bulb enlargement phase.
Bertoni's (1992) study provides information that the dry weight of leaves and roots is known to be almost the same in all treatments of nitrate levels but the levels of nitrate in roots are known to increase significantly until near harvesting period. This result is along with the Jurdin et al. (2015) statement who also reported that at harvest time the absorption of nitrate in roots is known to be high. The addition of biostimulant applications to nitrogen nitrogen treatment increases the efficiency of absorption of nutrients including the mobilization and uptake of nutrients from the soil, transportation, storage and assimilation. The absorption of nutrients in these plants is also influenced by the density of plant roots. The results of this study indicate that crown and accrual dry weight were highest in the N2 RL 480 treatment but were not different from the other treatments. In biostimulants, there are cytokinins, gibberellins and auxins. According to Aryanti (2012) that auxin increases the content of organic and inorganic substances in cells. These substances are converted into proteins, nucleic acids, Polysaccharides, and other molecular complexes. These compounds will form tissues and organs so that the wet weight and dry weight increase. Auxin can also increase plant osmosis pressure and softened cell walls which can increase water absorption and nutrients.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-07-26T08:08:04.949Z", "created": "2019-05-29T00:00:00.000", "id": "198272588", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://doi.org/10.18551/rjoas.2019-06.41", "pdf_hash": "2986cfae5e5c345171b4220242158e38347bef0b", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:31558", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "a59e79a507b97b2bdb840ca23a7df6c6f40ec148", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers : a Review
I Universidade Federal de Santa Maria, Departamento de Zootecnia da UFSM/CESNORS, Av. Independência, 3751, Vista Alegre, Palmeira das Missões, Rio Grande do Sul, Brasil II Faculdade de Engenharia Agrícola, Universidade Estadual de Campinas – UNICAMP, Avenida Candido Rondon, 501, Cidade Universitária Zeferino Vaz, Distrito de Barão Geraldo, Bairro Cidade Universitária, Campinas, São Paulo, Brasil. III Student of the Post-Graduation Program in Animal Science, Faculdade de Ciências Agrárias, Universidade Federal da Grande Dourados. Rodovia Dourados à Itahum, km 12. Caixa Postal 533. CEP: 79804-970 – Dourados – MS.
INtRODUCtION
Chicken egg production has experienced remarkable development in the last few years compared with previous decades and with the other sectors of the Brazilian poultry industry. According to the Brazilian Institute of Geography and Statistics (IBGE, 2012), 2.563 billion egg dozens were produced in 2011, representing a 4.3% increase relative to 2010. In addition, the total number of hens housed in 2011 was 118.915 million, which corresponded to a 3.2% increase compared with 2010. Egg production and the number of hens housed are expected to increase in 2012 due to the recent opening of new markets for the exports of poultry products.
In commercial egg production systems in Brazil, birds are reared in open-sided houses, benefitting from the existing natural environmental resources, as the country is located in a tropical region, with long photoperiods. Most commercial layer houses are equipped with ventilation/exhaustion fans, side curtains, foggers, etc., to provide adequate environmental control with aim of achieving efficient bird performance. However, the technological development of the houses of the last decades did not take into account the possible limitations of the use of electric energy, considering the global trend of saving natural and non-renewable resources.
Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers: a Review
Layers have been submitted to genetic improvement to produce more eggs at a lighter body weight and with lower feed intake. As a result, egg operations need to face the challenges of supplying the high nutritional requirements of layers and of designing management practices adapted to the increasingly automated and environmentally controlled facilities and to high stocking densities. According to Pavan (2005), both in Brazil and in the USA, there is a trend of housing more hens per cage in order to reduce production costs.
An interesting aspect of the physiology of egglaying poultry is that they do not require long and continuous periods of light. This phenomenon is called "subjective day", which indicates that adult hens in lay ignore periods of dark between the 14-16 hours of light stimulation. Subjective day is the period during which the bird is awake and physiologically active, even if it is in the dark. This allows the use of intermittent lighting programs for laying hens, which are programs that include more than one period of light (photophase) and one period of dark (scotophase) within a 24-h cycle (Gewehr & Freitas, 2007;Freitas et al., 2010).
In Europe and in the USA, such programs have been tested and are widely applied for layers maintained in closed sheds. The aim is to reduce the use of artificial lighting while maintaining bird performance and concentrating egg-laying time (Ernst et al., 1987;Boshouwers & Nicaise, 1993;Sauveur, 1996). Artificial lighting is a tool commonly applied to delay or to stimulate egg synthesis, as the circulating levels of the luteinizing hormone (LH) and of the follicle-stimulating hormone (FSH) increase within a single day of exposure to long photoperiods.
In Brazil, this issue is not frequently discussed because daily artificial lighting is applied for shorter periods in the open-sided houses compared with countries that rear layers in closed sheds. Therefore, research on this matter, using multidisciplinary analyses involving animal performance, environment, and electric energy consumption, and considering genetics, nutrition, and management practices, should be developed in Brazil.
Because electric energy derives from natural and non-renewable resources, its consumption should be carefully considered in animal production systems. In this context, all studies on the rational use of energy in such systems are significant, as the preservation of natural resources is essential for the sustainable development of the production chain.
This literature review aimed at analyzing the utilization of artificial lighting in commercial egg production. The effects of the different artificial lighting programs on layer performance and egg quality reported in Brazilian and foreign research studies are discussed, as well as studies on light perception by poultry and the physiological basis of the stimulation of the poultry reproductive system by light.
Light perception
Poultry perceive light through photoreceptors that transform the energy contained in photons in biological signals. In the eye, the energy of the photons is transformed by photosensitive pigments in the cones and rods of the retina, and transmitted through neurons to the brain, where the signal is integrated in an image.
However, light perception for reproductive processes does not depend on eye photoreceptors. It was demonstrated that photoreceptors in the hypothalamus are biological transformers that convert photon energy into neural impulses. These impulses affect the endocrine system that controls ovarian activity in females, and therefore, their reproductive and behavioral functions and secondary sexual characteristics (Morris, 1973;Etches, 1994).
According to Sauveur (1996), as opposed to mammals, the transcranial route is more important than the ocular route for perception of light information in birds. That author showed that darkening the head of sparrows with black paint blocked their sexual response, while depriving the eye of access to light did not have any effect.
However, considering that there are physiological differences among bird species, specific studies on this subject with egg-laying chickens are required, as they may reveal sexual characteristics that may pass unnoticed by the farmers.
Circadian rhythm
Ovulation depends on an endogenous mechanism that is closely related to external factors. The synchronization of these factors is called circadian rhythm and allows ovulation to occur regularly during lay. Layers use circadian rhythms to perceive the duration of the day and they are most sensitive to light between 11 and 15 hours after the light is turned on. During this photosensitive phase, a neural-hormonal mechanism controls the reproductive functions (Boni & Paes, 1999).
Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers: a Review
The circadian rhythm, also called biorhythm, is the physiological control of the metabolic activities of an individual by the light. Under natural lighting conditions, physiology changes during the 24 hours of the day (Freitas et al., 2003). According to Sesti & Ito (2000), chickens are not constantly stimulated during the entire photoperiod, but only on two important times of this period. They are initially sensitized when the lights are turned on and then 11-15 later. This is called the photosensitive phase and will determine if the bird will perceive the day as long or short. It must be noted that birds are not stimulated when days are short, whereas long days will trigger and maintain the hormonal flow that controls ovulation.
Considering that the stimulation of reproduction requires a light period, studies on artificial lighting management to delay or to stimulate gonad activity are becoming increasingly important. Different lighting management practices have been applied to delay oviposition in broiler breeders maintained in dark house systems, and to stimulate egg laying in commercial layers submitted to continuous light. According to Boni & Paes (1999), the main objective of housing broiler breeders during the development phase in dark houses is to prevent the release of sexual hormones (around 16 and 22 weeks of age) at the same time as growth hormone release, that is, before body development is completed.
Lighting programs for layers
Lighting programs used in layer production are classified according to photoperiod into hemeral and ahemeral. Hemera means day in Greek. Hemeral lighting programs consist of 24-h periods divided in light phases (photoperiod or photophase) and dark phases (scotoperiod or scotophase). Hemeral programs are used in open-sided houses, which take advantage of natural light (Campos, 2000).
Hemeral programs are classified as continuous or intermittent. In continuous programs, artificial lighting complements natural lighting to stablish a long and continuous photoperiod, while intermittent programs alternate light (photophases) with dark periods (scotophases). When the number of hours of photophases and scotophases is similar, intermittent programs are called symmetrical, and asymmetrical when different photophases and scotophases are applied. Asymmetrical hemeral programs have been used in egg production for energy-saving purposes (Rowland, 1985). Such programs allow workers to perform their activities inside the house during working hours (during the day).
Ahemeral lighting programs are used to improve eggshell quality and egg size, without affecting egg production. They are applied in environmentallycontrolled facilities mainly in Europe and in the United States (Ernst et al., 1987). Such programs may increase labor costs because daily tasks need to be performed outside regulated working hours, which, however, could be compensated by the production of larger eggs. Ahemeral programs apply photophases and scotophases that are shorter or longer, but not equal to 24h, and may continuous or intermittent (Etches, 1996).
Some intermittent programs have specific designations: the Cornell program and the biomittent program. In the Cornell program, 2 hours of light (2L), 4 hours of dark (4D), 8 hours of light (8L), and 10 hours of dark (10D) are supplied (2L:4D:8L:10D). It was developed by Tienhoven and Ostrander (1976) at Cornell University. The bird interprets this program as 14L:10D, ignoring the period of four hours of dark and considering a night period of 10 hours. The program was created to allow farmers to perform their eight hours of activities during the natural photoperiod.
The biomittent lighting program consists of fractioning the time of alternate light and dark cycles (25%L:75%D). According to Morris & Butler (1995), the objectives of the programs are to increase egg size and to improve eggshell quality. In the biomittent program, only 15 min of light are supplied per hour during the stimulation period, which may be interesting as it reduces lighting in 75% and improves feed efficiency in 5-7%. However, studies have shown that egg size is reduced in 0.5-1% when this program is applied (Rowland, 1985). Morris et al. (1988) demonstrated that the Cornell program reduces electric energy consumption and feed intake and promotes higher egg production. On the other hand, despite reducing feed intake, egg size and weight are also reduced when the biomittent program is applied before hens are 22 weeks old (Morris et al., 1990).
Light intensity
Egg production is reduced if light intensity is insufficient (Ostrander & Turner, 1962). Therefore, a luxmeter should be used to precisely determine light intensity in layer houses. The minimum light intensity required for maximum egg production is 5.38 lux in dark houses for layers (Skouglund et al., 1975). According to Cotta (2002), in open-sided houses, 10 lux are needed at cage or bird's head height. This
Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers: a Review
recommendation is accepted by the Illuminating Engineering Society of North America (IESNA, 2001), which performed a study that determined that a daily period of 14 hours of light for optimal egg production and that more than 17 hours of light negatively affect egg production. Because the visual system of chickens respond to light radiation within the visible spectrum range of 664-740 nm, the lamps used in layer houses should emit light within this range. The IESNA ( 2001) also recommends 10 lux as minimum light intensity for egg production. Light intensities higher than 10 lux do not bring any additional benefits and in fact may have negative effects on egg production as they may favor aggressive behaviors, hyperactivity, and cannibalism among hens.
Application of lighting programs
Both photoperiod and light intensity affect egg production. Lighting programs are fully integrated in modern layer management, with clear effects on performance. However, the light spectrum of each type of lamp needs to be considered, independently of lamp type (incandescent, fluorescent, sodium vapor, etc.), as mentioned by Etches (1996). Birds perceive not only the colors visible to humans, but also those in the shorter and longer ends of the spectrum. Birds are particularly sensitive to ultraviolet light, producing more reproductive hormones (Boni & Paes, 1999); however, not all lamps used for the artificial lighting in poultry production supply this wavelength.
Artificially changing the photoperiod is one of the most powerful management tools available for breeding poultry. It may delay or advance the onset of lay, synchronize egg-laying time, and influence egg production rate, eggshell quality, feed efficiency, and egg size (Etches, 1994). The reproductive system is not stimulated when days are short. On the other hand, long photoperiods stimulate the sexual function of layers and increase egg production. Long days are those which photoperiod is longer than 12 hours (Etches, 1996).
Several hypotheses were proposed in the past to try to explain the relationship between the reproduction cycle of poultry and photoperiod, but according to Sauveur (1996), none was proven. Rowland (1985) highlights two theories proposed for commercial poultry. The photoinductive theory assumes the presence of an endogenous rhythm (biological clock) with a cycle of around one day that is called circadian rhythm. The natural variation of the daily photoperiod acts as a conditioning factor of this clock, having a synchronizing role. The second theory is called photosensitive theory and assumes the presence of a model of external coincidence. Cycles of light and dark "train" the body to produce hormones during a specific period. The theory tries to demonstrate that birds are not equally sensitive to light during the day, and present maximum sensitivity 10-15 hours before dawn. Therefore, only long days are photostimulating (Sauveur, 1996). The second hypothesis, according to Malpaux et al. (1996), allows including periods of dark during the light phase, as these periods of dark are ignored by the birds. These hypotheses are not mutually exclusive. The model of external coincidence offers a better explanation because it was demonstrated in several experiments (Rowland, 1985), but neither theory make any restrictions as to the use of intermittent lighting for the stimulation of reproduction.
One of the most interesting physiological phenomena of breeding poultry is that they do not require long and continuous periods of light. Egg production can be stimulated by intermittent lighting programs (cycles of light and of dark). This lighting management is based on the concept of subjective day, which is the period during which the bird is awake and physiologically active, despite being in the dark. This concept allows including periods of dark during artificial lighting periods to stimulate egg production, shortening the artificial lighting period, but has no negative effects on performance.
Intermittent programs are based on the concept of subjective day. This theory assumes that mature laying hens in lay that were previously trained to a continuous and long photoperiod require only the information that their biological day is starting or ending, and ignore periods of dark within the time required to stimulate lay. This information can be given by a mere flash of light, and the bird thereafter ignores intermediate periods of dark. Understanding this phenomenon allows the application of intermittent lighting programs for layers. The concept of subjective day is illustrated in studies carried out with chickens and later with quails by Bacon & Nestor (1975). This theory was globally accepted after the 1980s, and was imprecisely called lighting program applied at the end of the laying cycle to improve eggshell quality (Sauveur, 1996). An intermittent lighting program may be defined as a program that includes more than one period of light (photophase) and one period of dark (scotophase) ina 24-h cycle. Evaluating the effect of intermittent lighting programs on layer performance, Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers: a Review Morris (1973) concluded that they can be applied in commercial egg production, and should be further explored. Freitas et al. (2005) compared the effects of an intermittent lighting program, natural lighting on days of increasing lighting, and a continuous lighting program on the performance of layers at the end of the laying cycle. The authors concluded that the intermittent lighting program or only natural lighting on days of increasing lighting can be applied in opensided houses to maintain performance.
Gewehr & Freitas ( 2007) published a literature review on the utilization of intermittent lighting programs for layers maintained in open-sided houses, discussing different lighting programs, the physiological basis of light stimulation of reproduction, and research results. They concluded that the results obtained until then indicated that intermittent lighting does not negatively affect the performance of commercial layers.
Further research is needed to elucidate some aspects of the physiological processes involved in egg synthesis; however, studies have shown that intermittent lighting programs do not interfere with the physiology of egg synthesis.
Lighting of layer houses using lightemitting diode (LED) technology Rozenboim et al. (1998) evaluated a new lighting system using monochromatic light for layers. In total, 45 layers were distributed in 15 cages per room (3 per cage) and submitted to three treatments: 0.1 and 0.01 W.m -2 light intensity using LED lamps or 0.1 W.m -2 using compact fluorescent lamps (PL or control treatment). In each LED room, three wavelengths were tested: 560 nm (n=9), 660 nm (n=9), 880 nm (n=6), as well as 660 using intermittent lighting (15 min light and 45 min dark) (n=9). In the room with PL lamps, birds were exposed to 12 h of light and 12 h of dark. At 21 weeks of age, the light period was increased to 12.75 h using 5.5 h with LED lamps and 7.25 h with LP for groups 1 and 2, and group 3 received 12.75 h of PL lighting. Up to 28 weeks of age, lightning was increased by 0.5 hour per week of light for all three groups until 16 h of light in week 28. Egg production and feed intake data were collected daily, and eggs for egg component analysis were collected weekly for 10 weeks. The authors concluded that, when optimal light spectrum is used, light intensity can be dramatically reduced, causing significant reduction in feed intake. A significant reduction in egg production, however, was observed in all groups exposed to 880 nm. This lighting method may benefit farmers as it allows reducing production costs. The obtained feed intake and electric energy cost reductions may increase net income in 20-30%, depending on feed costs. It was also observed that 0.01 W.m -2 light intensity significantly reduced feed intake, independently of light spectrum. In this experiment, egg weight was not affected by light color or intensity. Er et al. (2007) evaluated the effect of LED lighting on the egg quality of commercial layers. Hy-line layers were exposed between 19 and 52 weeks to 15-lux light intensity for 16 hours daily, using blue, red, green LED or incandescent lamps placed on top of each cage. The results showed that layers exposed to incandescent light laid heavier eggs than those exposed to red LED lamps, which produced the lightest eggs. Blue LED light reduced egg length and width, and changed egg shape, which became progressively round as hens aged. Red LED light reduced egg width, and egg shape became longer as hens aged. Egg weight tended to decrease with age when hens were exposed to blue and red light compared with green and incandescent lights. The quality of the eggs laid by layers exposed to green light were the most affected and was better compared with the other treatments. Xie et al. (2008) evaluated the effect of monochromatic LED lighting (red, green, white, or blue) on the immunity of broilers reared up to seven weeks and did not find any differences in antibody titers among birds exposed to blue, green, and white light. The authors concluded that blue and green LED light had a stronger effect in terms of enhancing the immune response compared with red LED light, and that blue LED light may alleviate the response to stress of broilers. Borille et al. (2008) compared the performance of commercial layers submitted to artificial lighting by LED lamps with different colors and conventional incandescent lamps. In total, 360 ISA Brown layers, 56-week-old at the start of the experiment, were exposed to six light sources: blue, yellow, green, red, and white LED lamps and incandescent 40-W lamps. A 17-h of continuous lighting program was adopted, and birds were fed a diet based on corn and soybean meal. Egg production was significantly different among the treatments, and the best results were obtained under red and white LED and incandescent lamps. The authors also observed that egg weight, feed intake, and internal egg quality (albumen height, specific gravity, and Haugh units) were not influenced Influence of Artificial Lighting on the Performance and Egg Quality of Commercial Layers: a Review by light source, and concluded that the replacement of incandescent lamps by red or white LED lamps does not have any negative effect on the egg production of commercial layers. Chen et al. (2008), studying the effect of artificial lighting on broiler performance and myofiber growth, reared 276 broilers under artificial lighting using red, green, blue, or white LED lamps. In the starter phase (0-26 days), broilers maintained under green monochromatic light had the best performance, whereas in the next phase (27-49 days), those reared under blue light presented better performance. The authors demonstrated that blue and green light promoted myofiber growth due to more effective stimulation of testosterone secretion.
fINAl CONSIDeRAtIONS
This review showed that layers are indeed photostimulated with more than 12 hours of light, independently of the artificial lighting program (continuous or intermittent) applied.
Results demonstrate that artificial lighting programs influence egg production, but not egg quality parameters. Literature suggests that egg-laying species show the same behavior relative to the physiology of egg synthesis, ovulation, oviposition, and ovulation cycle.
Intermittent programs are good alternative for lighting of layers maintained in open-sided houses, which are typically used in Brazil for commercial egg production, because such programs provide better cost-benefit ratio.
It is well established in literature (Mobarkey et al., 2010) that birds perceive light through both the transcranial and ocular routes. However, the transcranial route is more important for the stimulation of reproduction in commercial layers (Foss et al., 1972;Rocha, 2008;Oishi & Lauber, 1973). On the other hand, endogenous hormonal rhythms may be synchronized by ocular route, indicating to the birds the times to sleep, to awake, and to feed (Jácome, 2009;Rocha, 2008). Considering the need of a light period and the perception of light through the cranial route to stimulate reproduction, studies on the regulation of artificial lighting to delay or to activate egg laying are increasingly important. Their results can make significant contributions to improve the management of commercial layers and broiler breeders during the rearing and development phases.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-08-16T03:05:33.529Z", "created": "2014-12-01T00:00:00.000", "id": "38394323", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/rbca/a/47gSSD7krCVCJKZgyvG884x/?format=pdf&lang=en", "pdf_hash": "5bbd7c563224cbb7741a2d6806cd98312647b223", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:35151", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "5bbd7c563224cbb7741a2d6806cd98312647b223", "year": 2014}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Influence of drying on the chemical composition and bioactivity of Piper aduncum (Piperaceae) essential oil against Aedes aegypti (Diptera: Culicidae)
Piper species are producers of essential oils with high yield and promising chemical composition for both perfumery and the pharmaceutical industry. They present bioactivity against pathogens and against insect pests, whether agricultural or medical, such as Aedes aegypti, for example, a vector of arboviruses with a high incidence in tropical and subtropical regions. In this study, an investigation was carried out to elucidate the chemical composition of essential oils from the leaves and inflorescences of Piper aduncum collected in the state of Goiás, Brazil. Evaluating the interference of the drying process on yield, chemical composition and larvicide potential against Ae. aegypti. Leaves and inflorescences of P. aduncum were collected in the rural area of the municipality of Iporá-GO. Fresh and dried samples were processed separately and subjected to hydrodistillation for two hours. The oil obtained was qualitatively evaluated by gas-coupled chromatography and mass spectrometry. Greater yield was observed in samples submitted to the drying process. Oils obtained from fresh samples had a higher percentage of monoterpene hydrocarbons. Variation was observed between the major components of samples of fresh leaves and inflorescences, with eupatoriochrome being the major component in dried samples. Larvicidal activity against Ae. aegypti was considered promising (LC50<100μg/mL) in all samples. The results obtained showed a chemical composition different from that generally presented by P. aduncum. This reinforces the idea of intraspecific variability of essential oils and the need for chemical evaluation between samples even if they belong to the same species.
Introduction
The family Piperaceae is represented by plants of herbaceous, shrub sizes or small trees (Gogosz et al., 2012), with specimens distributed in tropical and subtropical regions of the planet (Judd et al., 2009). They are part of the group of angiosperms, which generally present in their phytochemical composition alkaloids, cardiac glycosides, coumarins, flavonoids, saponins, and triterpenes associated with different biological and pharmacological activities (Albuquerque et al., 2020). The genus Piper is considered the largest genus of the family, encompassing more than 700 described species, of which approximately 170 are native to Brazilian biomes (Sousa et al, 2008). Distinct species of Piper are used for medicinal purposes as anti-inflammatory, anxiolytic, anticonvulsant, sedative, antidiarrheal, and in urinary disorders (Oliveira et al., 2012).
Piperaceans are also essential oils producers. Some constituents common to essential oils of Piper are: safrole, predominant in the species Piper hispidinervum; ɣ-Terpinene and ρ-Cymene in Piper marginatum; safrole and dillapiole in Piper aduncum (Pereira Filho et al., 2021;Dos-Santos et al., 2018;Star et al., 2006). These constituents are known to be bioactive against bacteria, fungi, protozoa, and insects (Carballo-Arce et al., 2019). Dillapiole for example showed insecticidal potential against agricultural pests such as Sitophilus zeamais (Coleoptera: Curculionidae), Spodoptera frugiperda, and Helicoverpa armigera (Lepidoptera: Noctuidae), and urban pests such as Aedes aegypti (Diptera: Culicidae) (Durofil et al., Research, Society andDevelopment, v. 10, n. 8, e46810817397, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: [URL]8.17397 3 2021; Morais et al., 2007;Estrela et al., 2006). This constituent is a potent inhibitor of cytochrome activity (P450 and CYP34A), acting synergistically with pyrethroids and causing insect mortality including strains resistant to current insecticides (Carballo-Arce et al., 2019). Inhibition of these cytochromes in the insect reduces the ability to excrete xenobiotics present in food, where the death of the insect occurs through the accumulation of toxic substances located in the digestive tract (Li et al., 2007). Essential oils are produced by aromatic plants for the purpose of protecting the species against various environmental parameters. The use of the term "essential" for volatile oils occurs because they have odorous components in their composition (Simões, 2017). In general, volatile oils have very unstable characteristics, especially in the presence of light, air, heat, moisture, and metals (Simões, 2010). According to the material acquisition condition, qualitative and quantitative variation is a parameter to be analyzed. In addition to genetic, phenological, edaphic, and climatic factors (Gobbo-Neto & Lopes, 2007; Morais, 2009), the chemical composition of essential oils can be variable depending on the extraction technique and factors such as the time and weather condition at the moment of collection, as well as the processing conditions of the plant material before extraction. These factors can modify the chemical profile of the oil obtained, modify the biosynthetic pathway of the plant, stimulating or inhibiting the production of compounds, as well as interfering with the yield of the process .
P. aduncum has the essential oil with the highest percentage of dillapiole and, therefore, is a promising species for the research and development of bioinseticides (Gainza et al., 2016). It is popularly known as monkey-pepper, long pepper, false jaborandi. It is distributed throughout Latin America, from sea level to high altitudes (Fazolin et al., 2006). Medicinal use of the plant is reported in popular preparations as a digestive stimulant, sedative, diuretic, antidiarrheal, antimalarial agent, and even as an insect repellent (Pohlit et al 2006). There are many studies involving the biopotential of P. aduncum in different Brazilian biomes and in other countries of the American continent.
This research aimed to evaluate the chemical composition of essential oil of leaves and inflorescences of P. aduncum from Cerrado of Goiás (Goiás savanna), comparing samples that went through the drying process with fresh samples and the bioactivity of these oils against larvae of Ae. aegypti to investigate possible variations in both chemical composition and insecticide potential, depending on the variables observed in the acquisition process.
Botanical material harvesting and obtaining the essential oil
Samples of P. aduncum were obtained in the rural area of the municipality of Iporá-GO (6°29'45.96"S and 51°10'46.87"O -828 m) in the morning period in the first half of March 2021, end of the wet season. The region is climatically classified as humid subtropical (Alves & Biudes, 2008). Leaves and inflorescences of various shrubs of the same locality were collected. Botanical material was identified by the biologist Camila Aline Romano, MSc. In the laboratory, the samples were sanitized and screened. Part of the material was dehydrated in a forced air convection oven at 37°C for 24 hours. Another part was immediately undergone to the extraction process.
To obtain the essential oil, fresh and dehydrated leaves and inflorescences were crushed in a conventional blender and subjected to hydrodistillation in a Clevenger apparatus for two hours (Farmacopeia Brasileira, 2019). The essential oil obtained was desiccated with sodium sulfate anhydrous and stored in amber vials under refrigeration at -22°C for further chemical evaluation and bioassays.
Gas chromatography-mass spectrometry (GC-MS)
Essential oil of P. aduncum was subjected to gas chromatography-mass spectrometry (GC-MS) in a Shimadzu GC-MS QP2010A chromatograph equipped with a DB-5 fused-silica capillary column (30 m × 0.25 mm ID × 0.25 µm and 5%-Phenyl-Methylpolysiloxane) and ramp programmed as follows: 60-240°C at 3°C/min, then at 280°C at 10°C/min, ending with 10min at 280°C. Helium was used as a carrier with a flow of 1mL/min and the injection port at 225°C. Operating conditions of the mass spectrometer were: interface temperature 240°C; electron ionization at 70 eV with scanning mass range of 40-350 m/z and sampling of 1 scan/s. Constituents were identified by comparing their retention indices (Dool;Krstz, 1963) to nalkanes C9-C26 and mass spectra with data from the literature (ADAMS 2007) and digital library (NIST, 1998).
Larvicidal activity against Aedes aegypti
The essential oils of fresh leaves, dehydrated leaves, fresh inflorescences, and dehydrated inflorescences were tested against third instar larvae of Ae. aegypti. Bioassays were performed in a heated chamber with temperature of 28°C±1°C, relative humidity of 85%±5% and photoperiod of approximately 12 hours. An aliquot of essential oils was solubilized with surfactant polysorbate 80 (v/v) and distilled water to produce a solution at a concentration of 100 µg/mL, with which test solutions were prepared in serial dilutions of 100 to 10 µg/mL. A total of 20 larvae were exposed to the test solutions. A solution of water and surfactant was used as negative control, temephos solution (Abate ® -Basf Chemical Company) at 0,012 µg/mL was used as a positive control. For each bioassay, three replicates were performed at various times (WHO, 2005).
Mortality events were verified after 24 hours of exposure. The larvae that did not respond to the mechanical stimulation and that presented blackening of the body and the cephalic capsule were considered dead. The data obtained in the larvicide assays were submitted to the nonlinear regression method of Probit to determine the LC of 50 and 90% mortality.
Analyses were performed with the software Statistica Version 12.0 (StatSoft 2013).
Results and Discussion
The extractive process resulted in essential oils of yellowish color, and gently sweetish odor. The yield of the extractive process varied between 0.219% for fresh samples and 0.886% for dehydrated samples. Chromatographic analysis revealed 50 substances, presented in Table 1. There was variability in the composition of essential oils that went through the drying process. Samples of fresh hydrodistillated leaves and inflorescences showed a higher percentage of monoterpene hydrocarbons. On the other hand, dehydrated leaves and inflorescences presented a higher percentage of sesquiterpene hydrocarbons ( Figure 1).
The extractive process showed higher yield in dehydrated samples. However, in general, the samples had lower yield than that observed in the literature. The samples of this study were collected during rainy periods. Souza et al. (2018) found lower yield values in the extractive process of essential oil of Spiranthera odoratissima obtained in Cerrado of Goiás (Goiás savanna). Lemos et al. (2012) observed the variation in the composition of essential oil of Melaleuca alternifolia subjected to different drying temperatures, suggesting that the increase in temperature favors the oxidation of monoterpenes such as α-Pinene, as well as the volatilization of these compounds.
aduncum collected in different locations in the Amazon region of Brazil, they showed that the oil has a high yield (2.5 to 3.5%) and presence of the phenylpropanoid dillapiole as the majority component, presenting a variation between 30-97% among the samples evaluated. The present study showed a different chemical composition, to those already present in the literature.
Eupatoriochromene, a chromene highly found in P. aduncum (Taher et al, 2020) was detected in all samples, being the majority in dehydrated samples of leaves and inflorescences. This compound is bioactive against microorganisms and insects (Taher et al, 2020;Torres et al, 2017). Studies carried out a few decades ago already showed the larvicidal potential of eupatoriochromene on culicides, especially linked to the ability of interfering in the hormonal regulation of immatures and in the process of ecdysis (Klocke et al, 1985;Proksch & Rodrígues, 1983). Research, Society andDevelopment, v. 10, n. 8, e46810817397, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: [URL]8.17397 In the bioassays for investigation of larvicidal activity all samples showed promising larvicidal activity (LC50<100 µg/mL) ( Table 2). Although fresh leaf essential oil had a lethal effect in lower concentrations, there was no significant difference between the lethal concentrations obtained through the bioassay. Thus, it can be proposed that the oils from dehydrated leaves are more promising because they present bioactivity like the other samples tested, however present higher yield. Oliveira et al. (2012) also investigated the larvicidal activity of essential oil of P. aduncum. The authors found values of LC 50=289.9 µg / mL, concentration higher than that considered promising and when evaluated 1,8-Cineole, they did not observe a lethal effect. Studies with the essential oil of fruits of P. aduncum presented LC50= 30.29 µg/mL, where the composition revealed β-Pinene and limonene as majority compounds (Costa et al., 2010). In the present study, the essential oil of the inflorescences of P. aduncum presented limonene in its composition. All investigated oils presented as majority compounds with larvicidal activity investigated and proven in previous studies, such as germacrene D (Lima et al., 2019) and the eupatoriochromene (Torres et al., 2017). Thus, although dillapiole is not present in abundant content in some samples, bioactivity remained promising, due to the action of other substances with equal potential, or the synergism and enhancement that the composition can promote.
Conclusion
The present study investigated the effect of drying on the chemical composition of essential oil of leaves and inflorescences of P. aduncum obtained in the Midwest of Goiás, in addition to its bioactivity against larvae of Ae. aegypti. The drying process proved to be efficient to increase the yield of the extractive process in relation to the fresh one. It was also possible to observe that the amount of monoterpenes in the dehydrated samples was lower, due to their volatilization. The chemical composition of essential oils was relatively different from most studies with P. aduncum where dillapiole and safrole are majority constituents. Despite the differences in the composition of the oils, the larvicidal bioactivity remained in promising concentrations, mainly because in all samples there is a predominance of substances with already known insecticidal potential, such as eupatoriochromene and germacrene D. These results reinforce the need to evaluate the chemical constitution of essential oils whenever the conditions of collection, processing and extraction vary.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-09-27T20:56:06.521Z", "created": "2021-07-16T00:00:00.000", "id": "237645925", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://rsdjournal.org/index.php/rsd/article/download/17397/15690", "pdf_hash": "2fa675f1cb4db682dfc1ff1bc9533ec1f81193c3", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:36608", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "db5b9cfb7eadc80f7b45de21faee9422f7b38210", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
High grain quality accessions within a maize drought tolerant core collection
Maize (Zea mays L.) landraces are an important source of genes for improving commercial germplasm. Today, drought tolerance and grain quality are major challenges in maize cultivation due to climatic changes and population growth. The Maize Research Institute genebank has a drought tolerant collection, which includes 13 landraces (from the former Yugoslavia) and 12 introduced populations (from different countries). These accessions were analyzed for protein, oil, starch and tryptophan contents, in order to identify drought tolerant accessions with high grain quality. Also, simple sequence repeat (SSR) analysis with specific primers for opaque2 recessive allele (o2) was carried out. All analyzed accessions showed high levels of protein. Oil content ranged from 3.75 % to 5.40 % and starch content from 67.5 % to 71.30 %. Average protein content was not different (p < 0.01) between landraces and introduced populations. Starch and oil contents were higher in introduced populations at 0.84 % and 0.39 %, respectively (p < 0.01). Twenty-three accessions had high levels of tryptophan content. A high percentage of kernel type 1 and 2 indicated the presence of endosperm hardness modifier genes. Recessive o2 allele was found in most of the accessions. Absence of o2 in some high tryptophan accessions indicated action of another mutation. In two high tryptophan accessions an unknown band was detected. Absence of negative correlations between proteins, tryptophan and oil makes certain accessions suitable for use in the simultaneous improvement of target genotypes for these traits. Identified drought tolerant, high quality accessions can be used in breeding programs aimed at nutritional improvement of maize grown under drought conditions.
Introduction
It has been predicted that by 2020 the demand for maize (Zea mays L.) in developing countries will overtake the demand for both wheat and rice -by 2025 maize is likely to become the crop with the greatest production globally (Rosegrant et al., 2008).
Maize cultivation has been primarily aimed at increasing yield potential and stability under different environmental conditions. However, the quality of maize kernel composition is extremely important for human and animal nutrition and chemical characterization of the grain should become an inevitable component of maize cultivation programs. The relative chemical composition (e.g.protein, oil, and starch contents) in grains defines the purpose for which maize will be used (White et al., 2001;Baye et al., 2006). For example, quality protein maize (QPM) has increased amounts of essential amino acids lysine and tryptophan, and thus increases the nutritional value for protein-deficient populations (Krivanek et al., 2007), while in animal feed significantly increases the rate of pig weight gain (Sofi et al., 2009).
Redesigning maize for improved kernel qualities may require introduction of novel alleles not presently found in commercial maize germplasm. Teosinte and landraces harbor phenotypic variation that can facilitate genetic dissection of kernel traits and grain quality, ultimately leading to improvement via traditional plant cultivation and/or genetic engineering (Flint-Garcia et al., 2009). Using teosinte and landrace accessions as sourc-es of genetic variation for maize breeding is especially significant for genes that have limited or no variation remaining in modern inbreds (Wright et al., 2005;Yamasaki et al., 2005).
Maize Research Institute Zemun Polje (MRI) has a gene bank which conserves 2,217 landraces from the former Yugoslavia and 3,258 accessions from 40 countries. Surveying for drought tolerant accessions has resulted in the establishment of a drought tolerant core collection (Babic et al., 2011). Herein, the results of grain quality analysis of accessions from this core collection are presented. This study was aimed to: i) identify potential accessions with alleles for both drought tolerance and grain quality (macronutrient content, tryptophan content and kernel type) and ii) identify accessions that carry the recessive opaque2 allele for high tryptophan content, with specific simple sequence repeat (SSR) primers.
The twenty-five accessions from the elite drought tolerant core were multiplied (via pair crossing, i.e. fullsibling) in two replications, according to a completely randomized block design (RCBD), in 2009. At least 80 ears were collected per multiplied population. Each multiplied population was dried and shelled and samples of kernels from the centre of the ear were prepared for grain quality analysis.
NIR analysis
Near Infrared Spectroscopy (NIR) analysis was used for starch, oil and protein content analysis. The samples of non-destructed seed (300 g per sample) from each replicate of multiplied seed were scanned twice. Spectra were collected between 800 nm and 1,100 nm.
Tryptophan and QI analysis
Genotypes were represented with two samples, each consisting of 30 randomly chosen kernels from ten competitive plants. Kernels were dried in a thermostat at 65 °C overnight (16-18 h) and milled in a Cyclone sample mill. The flour was defatted by hexane treatment for four hours in a Soxhlet extractor.
Tryptophan content was determined using the colorimetric method (Nurit et al., 2009). The color was developed in a reaction of flour hydrolysate (obtained by overnight digestion with papain solution at 65 °C) with 2 mL of reagent containing 56 mg of Fe 3+ dissolved in 1 L of glacial acetic acid and 2 mL of 15 M H 2 SO 4 . After incubation at 65 °C for 15 min, absorbance was read at 560 nm. Tryptophan content was calculated using a standard calibration curve, developed with the known amounts of tryptophan, ranging from 0 to 30 μg mL −1 .
In order to obtain accurate quality index (QI) values, protein content was also determined in the samples used for tryptophan analysis by the standard Kjeldahl method based on nitrogen determination (Vivek et al., 2008). The protein was estimated from the nitrogen value as % protein = % nitrogen × 6.25 (conversion factor for maize).
Quality index (QI), i.e. tryptophan to protein ratio in the sample, was calculated as QI = 100 × tryptophan content in the sample / protein content in the sample
Identification of kernel types
For each genotype 200 kernels were visually scored using a light table (Vivek et al., 2008). The scoring scale defined kernel types, from type 1 (completely translucent, with no opaqueness) to type 5 (completely opaque). Kernels with 25 % opaqueness were scored as type 2, while types 3 and 4 were 50 % and 75 % opaque kernels, respectively.
Statistical analysis
All chemical analyses were performed in two replicates and the results were statistically analyzed. One factorial analysis of variance (ANOVA) for trials set up according to the RCB design was carried out. Fisher's least significant difference (LSD) test was used to test the significance of differences between the observed means. All statistical analyses were done using MSTAT-C software. Students' t-tests between average values of landraces and introduced populations were performed to determine statistical differences in the chemical composition and kernel type of the two groups of accessions. Pearson's correlation coefficient was used for determining correlations between protein, starch, oil, tryptophan, QI and kernel types, as well as between protein contents as determined by the NIR and Kjeldah method.
SSR analysis
SSR analysis of the 25 accessions was carried out using a DNA-pooled sampling strategy (bulk analysis). Each accession was represented by 30 individual seeds, which were grounded into flour. For bulk analysis equal amounts of the flour per accession were pooled into a single sample. Also, a subset of six accessions was analyzed for individual seeds, in order to get insight into distribution of the o2 allele within an accession. The choice of accessions was based on electrophoregram profiles of bulk analysis and tryptophan content.
PCR (polymerase chain reaction) amplification was performed on the DNA isolated from 25 bulked populations and 180 individual samples (30 seeds from six accessions) according to Rogers and Bendich (1988), with two primers specific for opaque2 gene -phi057 and umc1066. The amplification reaction was carried out in 20 μL reaction volume containing 1x enzyme buffer, 3 mM MgCl 2 , 200 μM dNTPs, 0.25 μM primers, 1.25 U Taq polymerase and 50 ng of DNA. The amplification profiles were as follows: an initial denaturation at 94 °C/2 min, followed by 40 cycles each of denaturation at 94 °C/1 min, annealing at 60° C/2 min and extension at 72 °C/2 min, with final extension at 72 °C/10 min. Amplified fragments were separated on 12 % polyacrylamide gels. After electrophoresis, gels were stained with ethidium bromide, photographed and analyzed using the BioDocAnalyze (Biometra) gel documentation system.
Biochemical analyses
NIR analysis revealed high levels of protein (from 10.90 % to 13.85 %) in all accessions (Table 1). Starch levels were in the range from 67.5 % to 71.30 % and content over 70 % was detected only among five introduced populations (IP3, IP4, IP10, IP11 and IP12). Oil content was in the range from low (3.75 %) to high (5.40 %). In one landrace (L1) and three introduced populations (IP2, IP3 and IP8) oil content was over 5.0 %. Average protein content in landraces and introduced populations was Sci. Agric.v.71, n.5, p.402-409, September/October 2014 12.07 % and 12.12 %, respectively, and the difference between them was not significant. On the other hand, starch and oil contents were higher in introduced populations at 0.84 % and 0.39 %, respectively (p < 0.01).
Protein content determined by the Kjeldahl method was also high in all accessions (11.20 % to 14.29 %) and on average higher when compared to the NIR meth-od, but the difference was not significant. Twenty-three accessions had tryptophan content over 0.070 % (0.070 % to 0.088 %). One landrace (L7) and one introduced population (IP6) had the lowest values for tryptophan (0.065 % and 0.069 %, respectively). Most of the accessions had medium QI values (0.60 to 0.69). Four landraces and five introduced populations showed QI below 2. Protein content values determined by both the NIR and Kjeldahl method were used for correlation calculations. Pearson's correlation coefficient for these two methods was 0.751 and was significant at p < 0.01 (not presented in the table).
Proteins determined by NIR showed negative correlation with starch and QI (p < 0.01). Likewise, proteins determined with Kjeldahl method showed the same negative correlations, but at p < 0.05 and p < 0.01, respectively. Positive correlations were determined between tryptophan content and QI (p < 0.01), as well as between tryptophan content and kernel type (p < 0.05). Also, QI showed positive correlation with kernel type (p < 0.05).
Identification of kernel types
Kernel type scores of the analyzed accessions were high (representing hard endosperm), in the range from 1.25 to 1.98 for landraces and 1.24 to 2.26 for introduced populations (Table 1). No differences between landraces and introduced populations were found. The most frequent were kernel type 1 (over 40 %) and 2 (over 30 %), making a total of 84 % within landraces and 82 % within introduced populations. Kernel type 3 was present in 15 % and 17 %, respectively (Figure 1). Types 1, 2 and 3 were detected in all accessions. Kernel type 4 was found in three landraces (L4 -4 %, L10 -2 % and L11-10 %) and two introduced populations (IP8 -2 % and IP11 -2 %). Completely opaque kernels (type 5) were not found in any of the accessions analyzed.
SSR analysis
Marker phi057 was monomorphic for the accessions analyzed and could not be used for identification of opaque2 recessive allele (o2). Umc1066 was polymorphic and this marker identified accessions which harbor o2. Illustration of homozygous and heterozygous accessions for opaque2 identified with bulk analysis is given in Figure 2. Opaque2 recessive allele was found in all accessions, except L7, IP2 and IP6 (L7 and IP6 are not illustrated in Figure 2). This locus was homozygous recessive (o2o2) in L1, L3, L5, L8 and L9 accessions and heterozygous (O2o2) in the remaining seven landrace accessions. Among the introduced populations, opaque2 was found to be homozygous recessive in IP7, heterozygous in nine and homozygous dominant (O2O2) in IP6 accessions. IP2 accession was heterozygous with dominant O2 and an unknown allele (UA) positioned between o2 and O2 bands. The two O2O2 accessions had the lowest tryptophan content (0.065 % and 0.069 %) (Figure 3). In all landrace o2o2 accessions tryptophan content was in the range from 0.072 % to 0.086 % and in all O2o2 accessions from 0.072 % to 0.081 %. Tryptophan content in the o2o2 introduced population (IP7) was 0.072 % and in the O2o2 populations in the range from 0.071 % to 0.088 %. In the IP2 accession in which an unknown allele was found, tryptophan content was 0.088 %.
Based on the allele profiles of the bulked samples and their tryptophan content, six accessions were chosen for the SSR analysis of individual kernels. Two homozygous recessive (o2o2) accessions L1 and IP7 had lower tryptophan content than L2, L10 and IP1 heterozygous (O2o2) accessions (0.072 % and 0.072 % vs. 0.077 %, 0.081 % and 0.084 %, respectively). On the other hand, IP2 accession had one of the highest tryptophan contents (0.088 %), despite having an allele that was unknown rather than the recessive o2 allele. Dominant homozygous individuals were found in L1 and IP7 accessions (Table 3). All individuals of L1 and 96 % individuals of IP7 accessions were homozygous recessive. In IP7 4 % were heterozygous for opaque2 locus, although the band characteristic for normal maize was not found in the bulked sample. One fifth to one quarter of the individuals were dominant homozygous for opaque2 in L2, L10 and IP1. Most of the samples were heterozygous, while recessive homozygous o2 allele was detected in 10 %, 25 % and 37 % of individuals, respectively. In 10 % of IP1 accession an unknown allele (UA), absent in the bulk and positioned between the o2 and O2 bands, was detected.
Recessive homozygous o2o2 individuals were not found in the IP2 accession. However, 8 % of the individuals were O2o2 heterozygous, although o2 allele was not detected in the bulked sample. An unknown band (also present in the bulk) was detected in 43.5 % of individuals (40 % in combination with O2 and 3.5 % as homozygous UAUA). This band is located between the o2 and O2 bands (Figure 2), at the same position as the unknown band in IP1 accession (not presented).
Discussion
The values for typical kernel composition for the commodity yellow dent maize on a dry matter basis are 71.7 % starch, 9.5 % protein, 4.3 % oil, 1.4 % ash and 2.6 % sugar (Watson, 2003). Compared with these data, all accessions analyzed by from MRI drought tolerant core collection had high protein contents (over 10.90 %), while high oil content (over 4.45 %) was found in four landraces and nine introduced populations. Analysis of 11 Mexican landraces, 32 white and six yellow hybrids revealed that landraces had higher average protein and oil contents as compared to hybrids -average protein and oil contents were 11.1 % and 4.9 % for the landraces, 9.7 % and 3.7 % for white hybrids and 9.8 % and 3.2 % for yellow hybrids (Vásquez-Carrillo et al., 2011). Similar values for these macronutrients were also found among 1245 accessions from different parts of the world (Berardo et al., 2009). Starch content among MRI accessions was low, except in the five introduced populations in which it was at the hybrid level (over 70 %). Lower starch content compared to hybrids was also found in accessions from the Gemplasm Enhancement of Maize (GEM) project (Singh et al., 2001).
Protein content was also determined by the Kjeldahl method, assuming that this method would be more sensitive than NIR analysis. Pearson's correlation coefficient for these two methods was positive at p < 0.01. Higher correlation for protein content (as well as for tryptophan and lysine contents) calculated by NIR and the colorimetric method was obtained in another experiment (Rosales et al., 2011), in which it was concluded that NIRS models can be used to support QPM breeding where a very high number of of samples have to be analyzed in a short time. However, for small scale analysis as the one performed in our work, the colorimetric method is more appropriate, since, in order to achieve a high level of confidence, NIR requires a development of the calibration curve that is based on a large and diverse set of samples and validation chemical analysis methods.
For many purposes not only the concentration but also the structure of macronutrients is important (Pollak and Scott, 2005). In this study, analysis of tryptophan content, one of the essential amino acids, was carried out. The limit values recommended for QPM selection are 0.075 % for tryptophan and 0.8 % for QI (Vivek et al., 2008). High tryptophan content was detected in the majority of the accessions. Fifteen accessions had a tryptophan content over the limit, eight in the range from 0.070 % to 0.074 % and two slightly below 0.070 %.
Tryptophan content in normal maize is usually under 0.060 %. In Ignjatovic-Micic et al. (2009) the tryptophan content in B73 and Oh43 inbred lines was 0.059 % and 0.054 %, respectively. Similar values of 0.059 % for white hybrid ZP74b and 0.052 % for yellow hybrid ZP636, as well as a somewhat higher content of 0.065 % in yellow hybrid ZP434 were obtained in a biochemical analysis of maize hybrids (Zilic et al., 2011). Across different genetic backgrounds lower limits of tryptophan in o2o2 maize overlap with its upper limits in normal maize. To confirm if the accessions with ambiguous tryptophan content were opaque2 genotypes SSR analysis with specific marker umc1066 was carried out. Two accessions with tryptophan content below 0.070 were found to lack o2 allele. Moreover, they had QI typical for normal maize. Presence of recessive o2 allele was detected in the five accessions with tryptophan content in the range from 0.070 % to 0.074 %, confirming that they were opaque2 genotypes. On the other hand, IP2 was one of the two accessions with the highest tryptophan content and highest QI, but within this population o2 was detected only in 8 % of the heterozygous samples. As tryptophan content is not elevated when o2 is not homozygous recessive, these results point to the presence of some other high lysine/tryptophan mutation (e.g.floury2 or opaque7), which, however, with different modes of action result in high tryptophan/lysine genotypes (Schmidt et al., 1987;Coleman et al., 1997;Miclaus et al., 2011).
Analysis of individuals within opaque2 heterozygous accessions (determined by bulk sample analysis) revealed different percentages of the presence of o2o2, from approximately one tenth to one third. High tryptophan content within these accessions, despite low presence of recessive homozygous opaque2 individuals, indicates that these individual plants had much higher tryptophan content than that measured for the whole population. It could also indicate the possible presence of amino-acid modifier/enhancer genes, considering that some landraces/introduced populations that were completely homozygous recessive for opaque2 had lower values of tryptophan content. Several enhancer genes have been identified (Wang et al., 2001;Wu et al., 2002) that affect the relative levels of lysine and tryptophan content in the grain endosperm (Moro et al., 1996).
In some cases bulk analysis did not detect O2 (IP7) or o2 (IP2) alleles, although they were detected in a small percent as O2o2 heterozygous loci in the analysis of individual kernels. This could be explained by the fact that alleles present in small percent can be missed in bulk analysis (Reif et al., 2005). In IP1 and IP2 accessions an unknown band (allele) was detected, positioned between the o2 and O2 bands. Both populations originate from Iran indicating that this band could be connected with their geographic origin. Most probably this allele is of dominant action, as it was present only as the heterozygous O2UA locus in IP1 and also the homozygous UAUA individuals were detected in a small percentage (3.5 %) of the IP2 accession. Since those two accessions had the highest tryptophan levels, there could be two possible explanations. The first one is that it was the consequence of some other gene(s) beside o2 (flowery2 or opaque7). The second one is that UA allele increases the tryptophan level, and if such is the case, this allele should be over-dominant to O2 allele. It would be worthwhile testing if UA has any influence on an increase in tryptophan content and its sequence should be identified.
High kernel type scores, absence of completely opaque kernels and presence of over 80 % of hard endosperm kernels were found in o2o2 accessions. The soft, opaque endosperm in o2o2 genotypes is the consequence of an o2 allele pleiotropic effect which has an influence on zein synthesis. However, a distinct genetic system comprising minor modifying genes (endosperm hardness modifier genes) influence gamma zein production and convert soft (opaque) into hard (vitreous) endosperm -such o2o2 grains with hard endosperm having Sci. Agric.v.71, n.5, p.402-409, September/October 2014 approximately double the amount of gamma zein in the endosperm as compared to the o2-only mutants (Danson et al., 2006, Wallace et al., 1990). The presence of a high percentage of hard endosperm kernels in the o2o2 accessions from the drought tolerant core clearly implies a presence of the endosperm hardness modifier genes.
Kernel type 3 was found in a small percentage (app.5 %) of both O2O2 genotypes. Some other opaque or floury gene could be present in these accessions, considering that according to the LSD test their tryptophan content was no different (p > 0.01) from some high tryptophan accessions. Also, the possibility of some other opaque or floury gene(s) action in the IP2 accession that has already been described is further enhanced by the presence of opaque kernels. Kernel types 3 and 4 were found in O2o2 accessions, as could have been expected, as bulk samples encompassed homozygous recessive individuals.
Conclusion
The accessions can be used as sources of new desirable alleles for food and feed improvement. Moreover, several accessions (e.g. L3, IP2 and IP8) have been identified that have high contents of protein, oil and tryptophan concurrently. Some of the high tryptophan accessions did not have the recessive opaque2 mutation, thus indicating the presence of some other mutation, such as fluory2 or opaque7. The drought tolerant core collection was selected in field trials performed in both sub-tropical and temperate zones. Hence, these genotypes could be used for breeding maize adapted to both environments.
Table 1 −
Oil, starch, protein and tryptophan contents, quality index (QI) and kernel type scores of the landraces and introduced populations from the drought tolerant core collection. The results are presented as a percentage of dry matter.
Table 2 −
Correlations between macronutrients (protein, starch and oil), tryptophan, quality index (QI) and kernel types determined by Pearson's correlation coefficient.
Table 3 −
Opaque2 profiles (percentages of homozygous and heterozygous individuals) obtained with umc 1066 SSR marker on the individual kernels of the six accessions.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-01-30T01:01:40.090Z", "created": "2014-10-01T00:00:00.000", "id": "41451131", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/sa/a/S3d9YqMsB9CvggPML559yLS/?format=pdf&lang=en", "pdf_hash": "6fc6bbd6e351315850970205d71c9e4f46127622", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:38296", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "6fc6bbd6e351315850970205d71c9e4f46127622", "year": 2014}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Seed yield and agronomic performance of seven improved cowpea (Vigna unguiculata L.) varieties in Ghana
Cowpea is well adapted to environmental conditions that affect crop production such as drought, high temperatures and other biotic stresses compared with other crops. Notwithstanding, growth and development of many cowpea cultivars are affected by drought and high temperatures, especially during floral development. This is because cowpea cultivars tend to have narrow range of adaptation as cultivars developed for one zone usually are not very productive in other zones. A study on the yield and growth performance of seven cowpea varieties was conducted during the 2012 major and minor rainy seasons at the CSIR-Crops Research Institute, Kwadaso-Kumasi, Ghana to compare the performance of the seasonal variation on each variety. These improved varieties Nhyira, Tona, Asetenapa, Asomdwe, Hewale, Videza and IT 89KD374-57 were evaluated using a randomized complete block design and replicated three times. The results showed that varieties Hewale, Videza and Nhyira gave higher seed yields, whereas IT 89KD374-57 and Asetenapa had lower seed yields. Nhyira and Hewale gave comparatively better seed yields under both conditions. Hewale was the highest seedyielding genotype under both major and minor raining season. Cowpea production could be a profitable agribusiness for cowpea growers in Ghana considering the higher returns in terms of grain yield obtained in both seasons.
INTRODUCTION
Cowpea (Vigna unguiculata (L.) Walp) is one of the most ancient human food sources and has probably been used as a crop plant since Neolithic times (Summerfield et al., 1974). Cowpea is grown extensively in 16 African countries, with the continent producing two-thirds of the world total (Winrock, 1992). The crop is of major importance to the livelihoods of millions of people in the tropics. For resource-poor small-holder farmers, the crop serves as food, animal feed, cash and manure. Going beyond its importance for food and feed, cowpea can be regarded as a pivot of sustainable farming in regions characterized by systems of farming that make limited use of purchased inputs like inorganic fertilizer. The crop can fix about 240 kg ha -1 of atmospheric nitrogen and make available about 60 to 70 kg ha -1 nitrogen for succeeding crops grown in rotation with it (CRI, 2006;Aikins and Afuakwa, 2008).
Cowpea is well adapted to environmental conditions that affect crop production such as drought, high temperatures and other biotic stresses compared with other crops (Martin et al., 1991). The aforesaid growth and development of many cowpea cultivars are affected by drought and high temperatures, especially during floral development (Dadson et al., 2005).
In Ghana, cowpea covers 156,000 ha (IITA, 1993). The yields of the crop, however, are among the lowest in the world, averaging 310 kg/ha (Ofosu-Budu et al., 2007). Meanwhile, the crop is one of the widely cultivated legumes, mainly in the savannah and transition zones of Ghana (CRI, 2006). Hence, efforts have been made to improve cowpea production in all agro-ecological zones of Ghana through various means including the introduction of new varieties such as those used in this study.
In recent years, several studies have evaluated the performance of cowpea genotypes in several ecological zones of Ghana. In selecting appropriate genotypes for different agro-ecological environments, it is important to know how various soils and climatic factors affect the growth and development of these new varieties in order to interpret the observed yields under these environments.
Appropriate agronomic practices to improve the performance of new varieties of improved and dualpurpose cowpea under different agro-ecological zones are generally important for breeding and production purposes. Yield and growth performance could be increased through the evaluation of all these varieties under different agro-ecological zones for a better understanding of their morphological, physiological and biochemical response to the environment. This underscores the importance of evaluating the agronomic performance of cowpea varieties as a food security crop under the current and foreseeable future scenarios. The objective was to evaluate key yield related parameters among seven cowpea genotypes in the Forest zone of Ghana.
Study site
The study was carried out at the research field of the CSIR-Crops Research Institute, Kwadaso-Kumasi, Ghana. The area has a bimodal rainfall pattern, the major season (April to July) with maximum rainfall normally in June and the minor season (September to November) with the maximum normally in October. Figure 1 shows the rainfall amount and distribution during the study year of 2012. The area receives a mean annual rainfall of 1500 mm with an average monthly temperature range of 24 to 28°C.
Soil analysis
Soil samples were randomly collected before planting from four different cores at 0 to 15 cm and 15 to 30 cm for determination of soil physical and chemical properties using soil auger. The soil is a sandy loam classified as Ferric Acrisol (FAO, 1990), equivalent to Typic Haplustult in the USDA soil classification system. Table 1 gives the initial soil analysis.
Experimental materials, design and treatments
Cowpea varieties Tona, Nhyira, Asetenapa and newly developed genotypes; Asomdwe, Hewale, Videza and IT 89KD374-57 used in this study were collected from the CSIR-Crops Research Institute, Kwadaso-Kumasi, Ghana. Seeds of the varieties were planted during the major and minor seasons after the experimental sites have been disc-ploughed and discharrowed. The seeds were sown in March and August, 2012 for the major and minor seasons respectively. Each of the genotypes was grown into a six-row plot of 3.0 × 5.0 m with a spacing of 50 and 20 cm between and within rows respectively. The experimental design was a randomized complete block design with three replications. Two inner rows were harvested to determine the final seed yield. Other parameters such as plant height, root length, nodule count and number of branches were measured on 10 randomly selected plants from each plot.
Data collection
Using the stratified sampling method, plants from an area of 1 m 2 were carefully uprooted from each plot. In all, 21 plots were sampled and for each sample, the roots were cut using a pair of secateurs, placed in an envelope and labelled before they were sent to the laboratory for dry matter analysis. The roots were carefully washed to remove attached soil. Fresh weight, root length and dry matter were taken from the sampled plants. Plants were placed in an oven maintained at 80°C for 48 h. Samples from the 21 plots were sent to the laboratory for dry matter analysis (Table 2).
Statistical analysis
Data were subjected to analysis of variance (ANOVA) using the Genstat Discovery 4th Edition statistical package with subsequent mean separation using LSD at 5% level of significance.
RESULTS
Results of the seasonal variation studies on the agronomic and yield performance of the improved cowpea varieties are presented in Tables 3 and 4. Significant differences (p<0.05) were observed among the varieties for grain and fodder production ( Dry matter yields were highest in Hewale (30.6 g) and Asetenapa (32.3 g) compared to the other varieties for the major and minor seasons respectively (Table 3). Hundred seed weight (HSW) ranged from 14.1 to 19.2 g with a mean of 15.8 g for the minor season and 16.2 to 25.1 g for the major season. The highest 100 seed weight was obtained from Videza for both the major and minor growing season (Table 3). Plant height was higher during the long rainy season compared to the short rainy season (Table 3).
Tap root length was significantly higher in Nhyira (21.8 cm) and least in IT89KD374-57 (15.3 cm) in the major season (Table 4). Nhyira again maintained the greatest root length in the minor season (19.0 cm) however this was only significantly different from IT89KD374-57 which had the least root length ((13.7 cm) (p<0.05) ( Table 4). Nodule count ranged from 15 to 20 and 12 to 20 for the major and minor seasons respectively. Videza recorded the highest nodule count for both the major and minor seasons (20.3 and 20.0 respectively), (p=0.05).
Nhyira produced the highest stem diameter (0.86 and 0.88 mm) for the minor and major seasons respectively with Asomdwe recording the lowest (0.56 and 0.72 cm) for the minor and major seasons respectively (Table 4). Number of branches per plant (NBP) among the cowpea varieties ranged from 5.0 to 8.0 and 5.0 to 7.0 for the minor and major seasons respectively (Table 4). Videza attained the highest number of branches in both the major and minor season (Table 4). First pod height for the minor season was not significantly different among the seven cowpea varieties (Figure 2). The major season produced first pod heights which were higher than the minor season. IT 89KD374-57 produced the highest first pod height (64 cm) and Nhyira producing the lowest FPH (52 cm).
On the other hand, final plant height at harvest was significantly different among the seven varieties for both seasons (Table 3). Videza recorded the highest height plant height of 82.3 cm whiles Tona recorded the lowest plant height of 54.2 cm for the major season (Figures 3 to 4).
DISCUSSION
The results of this study showed that rainfall is enough to support the growth and yield of cowpea in Kwadaso-Kumasi, Ghana, if other enabling factors are present in both the minor and major rainfall seasons. However, the cowpea varieties used in this study responded differently to the prevailing soil and climatic conditions. The highest yield of Videza was related to the continuous water supply. The results indicates that Nhyira and Videza will be more profitable than the other varieties in the minor and major seasons respectively and could serve as an alternative crop because of its desirable attributes and resistance to major biotic and abiotic constraints.
Videza, Hewale and Asomdwe in this study gave lower seed yield under short raining season than the seed yield of the same variety grown under long raining season (Table 2). This is because cowpea cultivars tend to have a narrow range of adaptation, as cultivars developed for one zone with distinct climatic factors usually are not very productive in other zones with different climatic factors (Hall et al., 2003). Irrespective of the potentials of these varieties as a drought resistant crop, failure of rainfall or lack of irrigation is a frequent cause of shortfall in production especially in Ghana where cowpea production is primarily grown in dry areas. Drought could be considered as an important factor among several seed yield-reducing factors. Clearly, there is a potential for further increase in seed yield by planting high-yielding genotypes, providing optimum irrigation, adding fertilizers (Quin, 1997), planting early and spraying with suitable insecticides. Therefore, selection of cowpea genotypes that have higher tolerance to drought is needed to obtain higher and more stable seed yields and in this regard Nhyira, Tona and Hewale appear to have some drought tolerance potential due to their higher yield during the minor season.
IT89KD-374-57 was observed to be low yielding due to its production of fewer nodules and dry matter. Low production of nodules means less nitrogen fixation by the variety. Production of relatively more leaves and branches with erect leave architecture in most cases reflect higher light interception and more photo-assimilate production that may result in increase yield.
In the development and growth of most cowpea varieties in the Sub Saharan Africa, yield and seed development require the production of assimilates in leaves, translocation of these assimilates to the fruits, unloading of assimilates from phloem of the seed coat into cells of cotyledons and synthesis of the various seed storage compounds. Yield losses resulting from water stress are generally associated with decreases in the activity of these physiological factors and dry matter production. Among the varieties that provided the highest biological yield under short raining season conditions in 2012 were Asetenapa and Hewale, whereas under the long raining season conditions the highest biological yield was provided by Hewale and Tona. The growth habits of these genotypes were bushy, erect or semi-erect, characteristic which can be used as a cover crop as well as for grain.
As observed in the grain yield, the biomass of the cowpea varied between the two seasons but the magnitude of the variation within the minor season was less than that observed in the major season. Differences in the seasonal fodder production of some varieties were not significantly different which showed that apart from rainfall some climatic factors like sun radiation might have influenced biomass production. Differences in day length between the major season and minor season in Ghana are that significant to affect crop production (Berchie et al., 2013).
The significant differences observed with the dry matter showed that attainment of reproductive phase was a varietal characteristic related to the genetic constitution of the varieties. Dry matter production in the minor season was more affected by genetic composition of the variety than the seasonal variation. This perhaps may be due to the ability of cowpea to survive under extreme water limiting conditions and could respond against the later drought. This was mainly achieved by slowing growth and reducing transpiration, as reported by Vianello and Sobrado (1991) that drought stress during vegetative stage provides diminution of the growth in most crop leaves and stems.
In this study, nodule numbers usually were lower at harvest than at earlier stages. The decline in number was especially noteworthy for nodules from taproots. Varietal differences account for nodule differences since the pattern of nodulation, most often, reflects the physical distribution of the root system in the soil. As reported by Hansen (1994), nodulation capacity is known to vary between and within legume species rather than rainfall variations as observed in this study. Varieties producing more nodules possess the capacity to fix nitrogen into the soil. However, genotypic effects on determinants of N 2 fixation resulting from nodulation are known to be complex. Lawn et al. (1974) suggested that the control of soybean nodule initiation occurs primarily in the root itself, but the control of nodule fresh weight occurs solely in the shoot and is related to the supply of assimilates.
The result of this study has shown better crop performance in terms of vegetative and grain yield during the long rainfall season than the short rainfall season. The reason could be attributed to relatively higher rainfall and milder temperature experienced during the production season of major rainfall. According to the annual report of the Science Daily (2008), plants growing under water limiting conditions tend to grow taller in an effort to scramble for below nutrients around the growth environment. These present results are consistent with previous study on cowpea by Hayatu and Mukhtar (2010), who reported that the results for plant height at final harvest showed that, increases in plant height under both moderate and severe water stress were recorded at the expense of seed yield in IT00K-835-45 and IT98K-819-118.
Plant population is reported to have effect on stem diameter, however, the results obtained from this study may be attributed to the better soil moisture availability, decreased plant competition and increased light penetration through plant canopy at such low plant population. The variation in stem diameter among cultivars might be due to genotypic differences.
Conclusion
From the results obtained in this study, it could be concluded that the performance of the three local and four improved varieties in terms of yield was higher in the major than minor season. Hewale and Videza are more suitable for high rainfall areas whereas Nhyira and Tona will be more productive and profitable in the drier areas. This study supports the clarion call that cowpea should become a successful legume crop for dry regions of Ghana.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-22T01:17:56.512Z", "created": "2015-01-22T00:00:00.000", "id": "91075559", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://academicjournals.org/journal/AJAR/article-full-text-pdf/75EB46C49673", "pdf_hash": "01c7f9fefacca611913ca1bedb10ae09dcdb14b7", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:38690", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "01c7f9fefacca611913ca1bedb10ae09dcdb14b7", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Challenges in Cocoa Pollination: The Case of Côte d’Ivoire
Cocoa (Theobroma cacao L.) is mainly pollinated by ceratopogonid midges (Forcipomyia spp.). However, other insect species will also pollinate cocoa flowers when these midges are scarce. In Côte d'Ivoire, inadequate pest control practices (insecticide spraying, mostly against the mirids Distantiella theobromae and Sahlbergella singularis) and landscape degradation as a result of deforestation and cocoa monoculture, have decreased overall pollinator population levels and, as a result, pollination services to cocoa trees. The current low average Ivorian cocoa yield of 538 kg per ha (in 2016) is the result of global agricultural mismanagement (deteriorated soils, lack of fertilizers, inadequate or absent pest control, absence of shade trees and intercrops). However, there is also an evidence of a pollination gap that could cause low cocoa yield. More research is needed to understand: (i) which agro-ecological efforts to enhance cocoa pollination can improve yield, and (ii) which strategies are effective in enhancing cocoa pollination. In this chapter, we briefly describe the cocoa sector. Next, the cocoa flower and pollinator biology and phenology are presented, followed by an overview of current environmental and management constraints to cocoa pollination in the context of Côte d'Ivoire, the largest cocoa producer in the world. We conclude with exploring possibilities to enhance pollination in the Ivorian small-scale cocoa sector.
Introduction
Non-bee insect pollinators play a significant role in global crop production [1]. Cocoa (Theobroma cacao L.) is one of the 13 most important commercial crops in the world. It entirely depends on insects for pollination and successful production [2]. In cocoa, pollination is almost exclusively performed by ceratopogonid midges (Order Diptera) from genus Forcipomyia [3,4]. In 2016, global cocoa production was 4.472 million tons of dry beans, of which 2.655 million tons (59%) were produced in West Africa and 1.472 million tons (33%) in Côte d'Ivoire [5].
In 2016, global average cocoa yield was 438 kg dry beans per ha and per year (480 kg per ha in West Africa), whereas it was shown in research stations that cocoa bean yield could attain up to 2000 kg per ha and per year [6].
Cocoa yield remains under the latter potential level due to: (i) an inadequate cropping system mainly consisting of full-sun monocultures without shade, leading to soil erosion, nutrient depletion, water shortages, weed growth, and increased pest and disease outbreaks [7][8][9][10][11]; and (ii) inadequate pest and disease management [12,13]; both leading to (iii) below-optimum pollinator population levels. Earlier research showed that increasing pollination, either manually [14] or indirectly by improving breeding opportunities for pollinating midges [14][15][16], had a significant impact on cocoa yield compared with normal agricultural practices control plots.
Over the past 50 years, cocoa demand has consistently increased annually by some 2.5% [17]. Demand continues to rise, particularly as a result of newly emerging chocolate markets such as China and India [18]. However, cocoa production levels have decreased by 3-5% over the past five years (compared to 2012 levels), leading to unstable prices because of market shortages. Cocoa production could expand through increasing global cocoa acreage (as it has always been done in the past). However, this is not a sustainable solution as it is mostly achieved at the cost of deforestation in tropical areas [19]. The other, more sustainable, approach is increasing productivity per ha. The latter strategy not only increases overall cocoa production without further deforestation, but can also increase income of cocoa farmers who nowadays often leave the cocoa production sector because of its low profitability [9].
In this chapter, we explore how cocoa farmers-besides by improving soil conditions, adequate pruning and integrated pest and disease management-can increase cocoa yield by increasing pollination intensity of their cocoa trees. We will first present the global cocoa sector and then focus on Côte d'Ivoire, the leading cocoa-producing country in the world. Next, we give background information on the biology and phenology of both cocoa flowers and pollinating midges (Forcipomyia spp.), followed by a discussion of pollinator-reducing factors (environmental and managerial) in Côte d'Ivoire. We conclude by discussing some options for relieving the constraints on cocoa pollination.
The cocoa crop sector 2.1. Global production and economic value
Although global yearly cocoa production quantities are below those of other tropical commodity crops such as sugar cane, rice, soybean, oil palm, cassava or banana, it is a unique crop because more than 90% of its production comes from small-scale farmers (each with a cultivation area not larger than 3 ha) [5,17]. As such, cocoa provides a livelihood to around 4.5 million farming families. Globally, 14 million people work in cocoa production [17]. In 2013, the total chocolate confectionary retail consumption had a value of 109,992 million USD [20]. Cocoa is produced on around 10 million ha, which is just 0.7% of the total global arable land, but 7% of the global permanent crop area. As a result, cocoa cultivation, and particularly cocoa agroforestry systems, play an important role in carbon sequestration and consequently have important climate mitigation potential [21].
The cocoa sector in Côte d'Ivoire
Côte d'Ivoire comprises the main cocoa-producing region in the world. In 2016, the country provided one third (1.472 million tons) of global cocoa supplies on 2.851 million ha of land. Average cocoa yield in Côte d'Ivoire was thus 516 kg/ha of dry cocoa beans, which is slightly below the global average yield of 538 kg/ha for that year [5]. In Côte d'Ivoire, cocoa is exclusively produced by around 1,000,000 small-scale farmers, each cultivating on around 2-3 ha [22]. These smallholders operate in a difficult context. Between 2000 and 2011, Côte d'Ivoire was generally considered a failed state with frequent occurrences of violent conflicts, where cocoa tax revenues were often used to fuel the conflicts [23,24]. Moreover, the cocoa production in Côte d'Ivoire has often been linked to child slavery on plantations [23]. As in most cocoa-producing regions, fluctuating prices (between 1500 and 3500 USD per ton in the period 2011-2018, see [URL]x) affect Ivorian cocoa smallholders because in a situation with volatile prices, it is difficult to make informed choices on the "right" crop investments [25].
Cocoa pollination
Pollination intensity and fruit set largely determine cocoa yield [26]. If natural pollination is limiting cocoa yield, then enhancing pollinator population levels should result in increased fruit set and consequently yield.
Biology
Cocoa flowers are hermaphrodite. They are produced on the trees' trunks and branches (cauliflory). After 2-3 years, so-called flower cushions, i.e., thickened flower-producing leaf axils, are formed. Every cushion bears up to 50 flowers per flowering season. There are two flowering seasons per year, which thus yields 100 flowers per year. The pentamerous flower is about 15 mm in diameter. A petal consists of a pouch-which conceals the anthers-and a wide tip. The function of the latter tip is unknown, but it does not specifically attract pollinators [27]. A particular aspect of cocoa flowers is the outer whorl of purple staminodes around the style.
Right after anthesis, these staminodes align parallel to the style (Figure 1). Pollinators move around on the inner side of the staminodes, thereby rubbing their pollen grain-carrying bodies against the style. On older flowers, staminodes are somewhat withered and flexed away from the style, which obstructs pollen deposition on the style [28]. The ovary consists of 40-70 ovules with axile placentation [29]. At least 20 ovules need to be successfully fertilized for a pod to develop and mature. Maximum pollination is achieved when pollination intensity, i.e., the number of pollen grains deposited on the style, exceeds 115 [30]. Usually, a mature pod contains between 30 and 40 beans [31,32]. Flower morphological characteristics (size, color, and shape) can differ greatly among varieties. However, even the most noticeable differences (e.g., white vs. red sepals) have no effect on pollination [33].
Bud development and maturation
Flower bud development from meristem to receptive flower takes at least 20 days and can take up to 30 days [31,32,34]. In India, it was shown [32] that flower bud development is faster in months with higher mean temperatures (e.g. June with a mean air temperature of 28°C) compared to colder months (e.g. November with mean air temperature of 25°C). Prolonged dry (<125 mm per month) or cold (mean monthly air temperature < 23°C) periods inhibit flowering [35]. Flowering is optimal during rainy days with high relative humidity and moderate temperatures (100 mm per month, 70% RH, and 27°C). High solar radiation incidence is linked with increased flower abscission [32]. Pollen grains are only able to germinate on a receptive stigma [36,37]. The receptive period is at about 2-3 days after anthesis. Unsuccessful pollination leads to flower abscission. Reported flower abscission rates vary from 63% on the main trunk and 81% on the fan branches to over 90% for all flowers [27,32,35].
Anthesis starts at around 2-4 pm. The latter becomes evident through splitting of the five sepals [27]. The process of sepal splitting continues overnight and finishes at around 4-6 am. Complete anthesis (flower fully open) is quickly followed by pollen release from the anthers (also between 4 and 6 am). Higher air temperature, as well as low air humidity, facilitates Flowers are produced in clusters directly on the trunk and older branches (this is known as cauliflory) and are small, 1-2 cm in diameter, with pink calyx. The floral formula is ✶ K5 C5 A(5° + 5) G(5) [31]. While many of the world's flowers are pollinated by bees (hymenoptera) or butterflies/moths (Lepidoptera), cocoa flowers are pollinated by tiny flies, Forcipomyia midges in the family Ceratopogonidae [16,94]. The tree flowers profusely, but few flowers set particularly in the dry season. When the tree is under water stress, all flowers are dropped within about 5 days. Successful pollination requires the deposition of at least 35 suitable pollen grains on the receptive parts of the flower, and is dependent on the season [95].
Styles and stigmas mature later than anthers, and have maximum receptivity around 12 am-2 pm. Maximum stigma and style receptivity does not concur with maximum anther dehiscence, thus limiting the possibility of self-pollination. The period during which the stigma is receptive to pollen and consequently during which successful pollination is possible, only lasts one day. Non-fertilized flowers will abscise the next day. About 1-5% of all flowers develop into a pod [31,32].
Cherelle wilt
Even after cocoa flowers are successfully pollinated and led to fruit set, not all young fruits (cherelles) will grow to mature cocoa fruits. Up to 80% of cherelles will shrivel, turn black, and become rapidly colonized by pathogens, while the pod remains on the tree. This so-called cherelle wilt is a physiological mechanism whereby the fruits are naturally thinned to balance nutrient allocation in the tree. Cherelles can wilt up to day 100 after fruit set [38]. Poor soils and impeded photosynthesis result in increased cherelle wilting [39,40]. Leguminous shade trees, which supply nitrogen to the soil, can therefore lower cherelle wilt [7]. Wilting in an early stage saves energy that can be invested in the development of the remaining fruits [30,31]. Apart from resource limitation, inadequate pollination (insufficient pollen grains deposited on the stigma surface) and incompatible pollen may also cause cherelle wilting [41].
Pod maturation
There are 130-160 days between fertilization and pod harvest [32]. The cocoa fruit is an indehiscent drupe. During the first 40 days after fertilization, pod growth is slow. Afterward, growth accelerates. The first division of the zygote only takes place between day 40 and 50.
Pod and ovule growth decrease from day 85 onwards, when embryos start to develop. On day 140, the embryo has completed its development and pod ripening starts [38].
Self-incompatibility
Most cocoa trees are self-incompatible. Self-pollination on a self-incompatible variety will not result in successful fertilization; as such, cross-pollination is then the only way for successful fertilization [42]. Self-incompatible trees are mostly cross-compatible; i.e., they are able to successfully fertilize flowers on other trees, including trees of the same variety. Incompatibility takes place at the stage of gamete fusion: incompatible gametes are unable to fuse. The underlying mechanism is of a genetic nature [32,43]. Following unsuccessful fertilization due to incompatibility, the flower drops off after 2-3 days. Even within a single variety, not all trees are necessarily either self-incompatible or self-compatible. However, the proportion of selfincompatible trees of a certain variety is determined by the specific variety. Self-compatible varieties that are cross-incompatible can restrict bean yield. In commercial plantations, it is therefore recommended to always plant different varieties [31]. Self-compatible hybrids produce larger fruits with a higher dry bean yield [44].
Viable pollen is able to germinate (producing a pollen tube) when it reaches the stigma. Pollen viability lies between 80 and 90%, hence does not limit fertilization [32,45].
Flower phenology
The number of flowers per tree varies throughout the season and is a function of climatic factors, such as photoperiod and temperature regime [46], whereas it is also cultivar dependent [31]. Furthermore, it seems that fruit production in the previous year determines flower production in the following year. Years of high pod production alternate with years with a low level of flowering [47,48]. In most tropical countries, flowering occurs year-round.
Flowering peaks are often preceded by increased temperature and rainfall, and occur at the onset of the rainy season, after which flower numbers gradually decline [45]. In West Africa, the major rainy season commences in April and climaxes in June, a period that is characterized by intense flowering (flowers on branches and trunks) [6]. In the minor rainy season (September-November), flowering intensity is lower (flowers on branches only). Few flowers are observed during the dry season (December-March) [47]. When pods are developing and this sink for assimilates is increasing, new flower production diminishes [40].
Overview of cocoa pollinating species
Early studies have ruled out wind as a pollinating agent-pollen grains form chunks, due to their viscosity and become too heavy to travel on their own [49]. However, in South America, experiments have been conducted to increase pollination by artificially increasing air currents in the field with motorized knapsack sprayers, thus stimulating wind pollination. This technique, however, only proved to be effective (doubling of cocoa bean yield) on self-compatible varieties [50].
Cocoa is almost exclusively pollinated by insects. The most important pollinators are midges from the family Ceratopogonidae. In reference [26], the author claims based on a review of five papers that female specimens are the main pollinators, although in reference [28], four times more males than females were collected in cocoa flowers. Ceratopogonids are biting midges of 1-4 mm length [51]. Males also pollinate, but to a lesser extent. It is not clear why females visit cocoa flowers more frequently than males [28,52,53]. Females presumably visit cocoa flowers to feed on the protein-rich pollen grains, necessary for egg maturation.
Besides ceratopogonids, other small dipteran insects such as Cecidomyiidae (gall midges), Chironomidae (non-biting midges), Drosophilidae (fruit flies), Psychodidae (moth flies), and Sphaeroceridae (small dung flies) have been documented to visit cocoa flowers. Other insects, such as aphids, coccids and cicadellids (Hemiptera), thrips (Thysanoptera), and ants (Hymenoptera), also occasionally visit cocoa flowers. However, their contribution to pollination is most probably very low. Up to date, pollen grains have not been detected by microscopic observation on insects other than Forcipomyia spp. In some cases, observations suggest that cecidomyiids (in Cameroon) and drosophilids (in Ghana) may contribute to some extent to pollination [26].
Only Diptera, and particularly genus Forcipomyia (Fam. Ceratopogonidae), are morphologically able to pollinate cocoa. Forcipomyia holds the largest number of cocoa pollinators. Within that genus, the most frequently reported pollinators belong to the subgenera Euprojoannisia (before: Proforcipomyia and Euforcipomyia), Thyridomyia, and Forcipomyia [26].
It is well-documented that ceratopogonids breed in humid, decaying organic material such as cocoa leaf litter, decomposing cocoa pod husks, banana pseudostems, and bromeliads [4,54]. Besides being moist, these breeding substrates are cooler than the ambient environment and provide dark conditions which all benefit ceratopogonid breeding [31].
In the 1970s, cage experiments [28,52,53] were performed to characterize the pollination capacity of different ceratopogonid species. However, results of these experiments have little value as they were performed under unrealistic conditions (exposure of a high number of flowers to a single midge and use of small cages, both causing pollination levels that the same midges would not achieve in nature). The only valid method to determine whether a species is a pollinator is through field observation [26]. It has been shown that artificial circumstances bias lab experiment results considerably; for example, successful pollination by Tyora tessmanni was shown under lab conditions, but could not be confirmed under field conditions, where the putative pollinator was abundantly present [55].
Female ceratopogonids, in search for sugary nectar, start pollinating cocoa flowers early in the morning (5-8 am) and also actively visit flowers in the afternoon (4-6 pm) [3,52]. Ceratopogonids carry cocoa pollen grains on their thoracic hairs. Weather conditions affect their flower visiting activities: rain and clouds decrease their activity whereas sunny weather increases it [3]. Some trees receive more attention from pollinators than others, resulting in a greater fruit set in some trees as compared to others. The interest for particular trees shifts with time. Why this happens, is not clear. Female ceratopogonids commonly visit cocoa and other flowers everywhere in the world [4,61].
Ceratopogonid midge flights might cover long distances, but it is not known how far exactly [26]. Distance traveled during one foraging event, and consequently during which pollination is performed, can reach up to 50 m. However, midges mostly deposit pollen from a certain cocoa tree on flower stigmas of neighboring cocoa trees [31,62]. It has been shown that there are 5-7 times more Forcipomyia specimens above the cocoa canopy than below the canopy [26]. Since wind speed above the canopy is higher than below, it can be expected that wind could play an important role in horizontal cocoa pollinator distribution over the cocoa field.
Besides feeding on flower nectar, adult ceratopogonids also suck the blood of other insects and mammals. In general, pollinating activity is very limited in time during the lifetime of these pollinators.
Ceratopogonid pollinator populations can be abundant and exceed one million individuals per ha [3]. Moist environments favor ceratopogonid midge abundance. In fact, there is a positive correlation between soil moisture and ceratopogonid population levels [26].
Stable moist conditions are indispensable for successful development of eggs and larvae [63]. It is suggested that the West African harmattan (dry, hot wind from the north) results in withered breeding places, rendering them unsuitable for insect breeding [26]. Pollinator populations thus increase with each rainy period, to decrease again with the onset of a drier period [31].
Pollination gap in cocoa
The yield gap in cocoa (i.e., the difference between yield at optimal, experimentally determined growing conditions and the current cocoa farm yield) is caused by multiple factors including disease, pest and weed pressure as well as inadequate phytosanitary practices, lack of improved varieties, low soil fertility, etc. [64]. However, there is increasing evidence that the present yield gap is also linked with inadequate pollination. This so-called pollination gap was already observed in the late 1970s when it was found that during the dry season, the number of ceratopogonid pollinators, as well the relative number of pollinated flowers were lower than in the wet season [3,4,26]. Because rotten, moist organic material is an ideal breeding substrate for ceratopogonid midges, attempts have been made to increase reproduction opportunities for these midges by adding such organic material in cocoa plantations. In an experiment in Ghana, banana pseudostems, cocoa pod husks and leaf litter were added as pollinator breeding substrates next to cocoa trees. It was found that midge population increased to 500% of the control tree levels whereas fruit set in treated trees was four times higher than in control trees. Cherelle wilt also increased in treated trees but was lower than increased fruit set rates so that the final number of mature fruits was twice as high for all substrate-treated trees compared to the control trees [65]. A more direct proof of the pollination gap was found when cocoa trees in Sulewesi (Indonesia) were artificially pollinated. Optimum dry bean yield was achieved when 40% of flowers were hand-pollinated [14]. The latter treatment increased dry bean yield by 350 kg per ha as compared to a pollination intensity of 10%, which concurs with natural pollination intensities observed over the past 20 years [30,66]. In North Queensland (Australia), it was recently shown that adding cocoa pod husks as a pollinator breeding substrate considerably increased fruit set (110 times more cherelles) and yield (60 times more fresh fruit production). However, hand pollination in fields where breeding substrate had been added did not result in extra yield, indicating that breeding substrates had already increased pollination intensities to optimum levels [16].
Cocoa monoculture
Cocoa is a shade-tolerant tree. Traditionally, cocoa is grown in shaded, agroforestry systems where it is intercropped with forest trees that were spared when the forest was cleared for cocoa cultivation. However, it was shown that-provided soil nutrition levels are adequate-cocoa production with shade trees is lower when compared to full-sun production [31,[67][68][69]. As a consequence, agroforestry systems have globally been replaced by monoculture systems with low shade provision [70]. Over the past few decades, cocoa cultivation has intensified not only by removing shade trees but also by extensive application of fertilizers and pesticides. As a result, the insect assemblage of cocoa cultivation systems has changed considerably. When compared to agroforestry systems or natural forests, insect biodiversity has decreased in present-day cocoa plantations, often at the expense of predators, leading to increased pest outbreaks and pollinators [71][72][73][74][75].
Landscape degradation
In Côte d'Ivoire, the cocoa sector is largely responsible for landscape degradation [19]. Over the past few decades, cocoa was typically cultivated on freshly cleared land where its production rapidly expanded, after which the land was abandoned 10-15 years later due to declining yields. Since the 1970s, such continuous so-called boom-and-bust cycles, as well as cocoa expansion from the southeast to the southwest of Côte d'Ivoire, have led to massive deforestation in the country [9,22,76]. In the 1960s, total tropical primary forest cover amounted to around 8.14 million ha. In the 1980s, that area had dropped to 2.6 million ha, whereas in the 2000s, primary forest cover was just over 1.35 million ha, meaning that since its independence, Côte d'Ivoire has lost 80% of its forest cover [77].
Almost all cocoa plantations in Côte d'Ivoire have less than 50% of shade, meaning that the majority of trees are fully exposed to sunlight, leading to biodiversity loss and soil deterioration, often resulting in reduced addition of organic matter to the cocoa plantation soils [78]. It has been extensively shown that Forcipomyia spp., which are the predominant pollinating midges, require moist and decaying organic material to breed [15,28,52,79]. Also, the vicinity of natural forest and moist refuges promote diversity of Forcipomyia spp. and cocoa pollinators in general [52,80]. It is therefore fair to assume that in Côte d'Ivoire, massive landscape degradation has led to decreased breeding opportunities and consequently to lower population levels of cocoa pollinating midges.
Target pests and insecticide products used
The major cocoa pest problem in West Africa is caused by mirids (Order: Hemiptera, Fam. Miridae). Sahlbergella singularis and Distantiella theobromae suck the sap from cocoa pods and young shoots, causing commercial cocoa losses of up to 30% [81,82]. In West Africa in general, more than 75% (in some areas 100%) of cocoa farmers use chemicals to control mirid infestation [83]. Nowadays, most frequently used insecticides in cocoa cultivation are the pyrethroids bifenthrin, cypermethrin, deltamethrin and lambda-cyhalothrin, and the neonicotinoids acetamiprid, imidacloprid and thiacloprid [84]. In Côte d'Ivoire, almost all farmers who use insecticides, apply commercial products containing a systemic neonicotinoid insecticide, usually in combination with a contact pyrethroid insecticide two times per year (July-August and January-February). The pyrethroid would thereby kill the mirid adults as well as the nymphal instars, whereas the systemic neonicotinoid would ensure that mirids that hatch after insecticide applications are also killed (personal communication with local pesticide dealers) ( Table 1). However, the precise impact of these specific insecticides on cocoa pollinators in Côte d'Ivoire is unclear and should be further investigated.
Impact of pesticides on cocoa pollinators
Broad-spectrum insecticides (such as β-hexachlorocyclohexane and dichlorodiphenyltrichloroethane), which were historically widely applied in cocoa crop production, did not affect pollinator population levels [26]. It is suggested that breeding sites are protected from
Pollination in Plants 48
insecticide sprayings by leaves and other organic material. However, residual effects of insecticides might affect cocoa pollinators [3,54]. A study in West Africa on the effect of large-scale insecticide treatments (against mirids; Fam. Miridae) on both pollinator population levels and cocoa pod production showed that there is only a short-term negative impact of insecticide treatments on pollinator population levels [31]. Also in West Africa, it was shown that fogging instead of spraying insecticides is less harmful for pollinators, as fogging only negatively influences the population level for 2 days compared to 8 days with spraying [85]. Alternative approaches are to (i) reduce insecticide dosages during the period that pollinator population levels are low, and (ii) use narrow-spectrum insecticides.
Integrated pest management (IPM) options in the Ivorian cocoa sector
Despite the currently widely applied spraying programs, mirid infestation remains the most severe cocoa production limitation factor [81]. Although the precise impact of pyrethroids and neonicotinoids on cocoa pollinators in Côte d'Ivoire is unknown, it cannot be excluded that apart from these regular pests, pollinators are also affected by these products. Therefore, novel and more integrated pest management (IPM) approaches should be tested against mirids. The latter approaches might include: (i) further development and testing of mirid pheromones [86]; (ii) increasing shade levels by planting shade trees to avoid so-called "mirid pockets" (i.e., mirids particularly occurring in non-shade areas of the plantations) [81]; and (iii) enhancing ant populations as they are most probably natural mirid predators [87].
Enhancing cocoa pollinator environment
In cocoa plantations, pollinator population levels can be increased by augmenting the amount of natural pollinator breeding sites or by adding artificial breeding substrates. Since it is known that Forcipomyia spp. breed in moist and rotting organic material, introducing such material in the cocoa field will most likely enhance pollinator breeding and subsequently their population levels. Banana pseudostems are preferred as a pollinator breeding substrate over cocoa husks, because the latter are a possible source of black pod disease [88]. Intercropping with fruit trees will not only provide shade, but (provided that not all fruit is harvested), will also introduce rotting fruit in the plantations as potential pollinator breeding sites. As shown in Figure 2, in Côte d'Ivoire, we currently investigate, together with the cocoa farmers of the local cooperatives, the effect on pollination levels of squared pits (0.5 × 0.5 m and 0.3 m deep) that are spaced in 10 × 10 m squares and in which organic material such as fresh empty pod husks, cut banana pseudostems, and fruits from intercropped trees such as Citrus spp. will be deposited to enhance pollinator breeding.
Pollinator mass breeding and mass release
Mass breeding and subsequent mass release of Forcipomyia spp. at times when cocoa flowering peaks, might also have a significant effect on effective cocoa flower pollination. The idea is based on similar practices commonly applied in the horticultural sector where bumblebees (Bombus terrestris) are commercially bred and subsequently released in tomato (Solanum lycopersicon) greenhouses for tomato flower pollination [89]. As compared to the earlier used vibrating sticks to induce pollen release from tomato flowers, bumblebees increase tomato fruit set by 45%. Another example is the black soldier fly (Hermetia illucens) (Order: Diptera, Fam: Stratiomyidae) that is used to enhance composting of food waste and reduction of organic manure volumes, and which can be mass bred prior to release on organic material [90]. To our knowledge, no mass breeding attempts for Forcipomyia spp. have been undertaken up to date.
The hematophagous nature of Forcipomyia midges can be a constraint to their mass breeding success [91]. Laboratory experiments showed that F. townsvillensis eggs will not develop without complete blood meals [92]. Research is needed to test the most appropriate midge rearing conditions (temperature, humidity and feeding).
Forcipomyia spp. mostly pollinate flowers neighboring the ones where they have collected pollen [93]. We assume therefore that they do not swarm further than 10 m from their breeding sites. Under that assumption, mass release should be performed at least each 20 × 20 m in cocoa plantations (25 releases per ha). Given the wide diversity of Forcipomyia spp. that have been [53,96]. Indeed, early on, scientists figured out that most Theobroma cacao trees are not able to self-pollinate, but for years, they could not figure out what moved cocoa pollen between trees. It turned out that cocoa flowers are pollinated by midges not much bigger than tiny specks of airborne dust. Midge populations are greatest in the rainy season. Adult midges spend the day in shady spots such as between the buttress roots of large trees, in crevices in logs, in hollow stumps or in piles of husk debris.
They emerge at variable times of the day to swarm near their hiding locations, and disperse in the early morning and late afternoon. Most midges do not move further than about 6 m. The females lay batches of eggs on damp piles of plant debris, on moist decomposing wood, cocoa husks and other plant debris, in batches of 40-90 eggs. Eggs hatch after 2-3 days and the larvae pass through four instar stages before pupating at about 12 days; the pupal stage lasts 2-3 days. The adults survive for about a week and there are thought to be about 12 midge generations per year. Adult females require liquid plant food for survival and oviposition, although ovary maturation is independent of adult food intake or mating. In a joint project between Ghent University and Barry-Callebaut, and in collaboration with local cocoa smallholders and their cooperatives, we introduced squared pits of 50 × 50 cm and 30 cm deep, filled with organic material such as cut banana pseudostems, fresh empty pod husks and fruits from intercropped trees such as Citrus spp., at a density of 1 pit per 100 m 2 (spaced at 10 × 10 m as based on presumed midge flight radius of the midges) to enhance the establishment of the cocoa pollinating midge populations in the field (main picture is a photo by Guy Smagghe in cocoa plantation at Tiassalé, Côte d'Ivoire, 15/01/2018; inset photo is of a mating pair of Forcipomyia midges by Christophe Quintin, [URL] in Plants
identified as cocoa flower visitors and the fact that some are restricted to either Africa, Central America, or South America (only one cocoa flower pollinator, F. fuliginosa was observed in all regions), it can be assumed that specific pollinating midges are restricted to certain areas [4]. It is therefore recommended that Forcipomyia spp. mass breeding for use in a certain cocoa area would start with locally sampled Forcipomyia midges, as exotic midges might disturb local biotic equilibria. Obviously, as a precondition to adoption of commercial mass breeding of pollinating midges by resource-poor smallholders in Côte d'Ivoire, the technology should be cost-effective.
Conclusion
Since cocoa production essentially depends on insect pollination, any threat to pollinators will have a negative impact on cocoa production. There is evidence that currently, cocoa pollination is below the optimum level and that enhancing pollinator populations in cocoa fields could increase cocoa production [14][15][16]. It is clear that cocoa pollinators are threatened by the currently predominant cocoa production system, which consists of full-sun cultivation on often deforested land with degraded soils and chemical pest control. Pest control, shade tree planting, and landscape management all influence cocoa pollinator presence, making pollination management very complex.
Many research questions on cocoa pollination remain. They include: (i) quantification of the pollination gap (only in [14] attempts have been made, but just by comparing hand-pollination treatments with unpollinated controls); (ii) evolution of the pollination gap throughout the year (e.g., in West Africa, the gap might be narrower during the dry season when flowering is less abundant); (iii) the relation between pollination and cherelle wilt (can cherelle wilt be decreased by improving pollination efficiency?); (iv) success rates of artificial pollination (a difficult task requiring a lot of agility and experience); (v) influence of insecticide applications on pollinator and other insect population levels; (vi) role of landscape and cocoa cropping systems (agroforestry, intercropping, soil mulching) on pollinator species composition and abundance; (vii) pollen load and pollination efficiency of cocoa flower visitors other than Forcipomyia spp.; (viii) evaluation of pollinator roles in self-compatible as compared to selfincompatible cocoa trees; (ix) promotion of self-pollinating self-compatible trees; and (x) effectiveness of enhancing ant populations to improve cocoa pollination.
As final conclusion, we believe that the answers to these research questions will undoubtedly lead to decreasing the current cocoa yield gap, which is the only sustainable solution to increasing global cocoa supplies.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-04-02T13:14:38.406Z", "created": "2018-03-15T00:00:00.000", "id": "90501756", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://api.intechopen.com/chapter/pdf-download/59982.pdf", "pdf_hash": "3e3d9faee0b65d1db4d3d455c75fd242f0a8d9d7", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:38298", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "2a26185f168e98fe0b02132f876662a56af9d939", "year": 2018}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
The effect of ethyl methane sulphonate and diethyl sulphate on chilli ( Capsicum annuum L.) in M 1 generation
In recent years, the demand of chilli has tremendously increased due to its attractive market price and multifarious used in cooked and processed forms. At present people are much concerned about the fruit quality and yield. Therefore, attention is being paid for development of genotypes having high yield potential with desirable fruit quality characters in a short period of time. For this purpose, seeds of chilli were mutagenised with ethyl methane sulphonate (EMS) and diethyl sulphate (DES) to determine their mutagenic sensitivity in M 1 generation. The increasing concentration of EMS and DES decreased in morphological and yield characters. The spectrum of mutation and induced variability for various quantitative traits were observed in M 1 generation such as germination (%), plant height, primary and secondary branches per plant, days to first flowering, fruit length (cm), fruit girth (cm), total number of fruits per plant, number of seeds per fruit, seed weight per fruit (g), 100 seed weight (g) and pericarp: seed ratio showed variability in chilli with the effect of EMS and DES. The percentage of chromosomal abnormalities in different mitotic stages was significantly higher than that of the control in all the treatment concentrations.
INTRODUCTION
Different definitions of the term "mutation" and this may create the impression, that the term is somewhat woolly. Definitions range from "a sudden phenotypic change in a character of an individual, not due to crossing or segregation" up to "an alteration in the macromolecules in the DNA" (where it remains open, whether the alteration leads to a change in gene function or not). Included under the term "mutation" is also the augmentation of genetic material through nucleotide or gene copies, through additional individual chromosomes, as well as through the multiplication of whole genomes towards polyploidy. In order to speak more clearly about mutations and their potential for crop improvement, it would seen desirable to have different terms at least for (a) the phenotypic alteration and (b) the various underlying molecular and numerical changes. But in any case, a mutation has to be phenotypically expressed to be selectable all other mutations are only of scientific interest (Alexander Micke, 1999). Mutation methodology has been used to produce many cultivars with improved economic value and study of genetics and plant developmental phenomena (van den Bulk et al., 1990 andBertagne Sagnard et al., 1996). Induced mutations have great potentials and serve as a complimentary approach in genetic improvement of crops (Mahandjiev et al., 2001). It has been used to improve major crops such as wheat, rice, barley, cotton, peanut and cowpea, which are seed propagated. Various mutagenic agents are used to induce favourable mutations at high frequency that include ionizing radiation and chemical mutagens (Ahloowalia and Maluszynski, 2001). Chemical mutagens are the one cause of mutations in living organisms. Many of the chemicals have clastogenic effects on plants via reactive oxygen-derived radicals (Yuan and Zhang, 1993).
Chilli (Capsicum annuum L.) belongs to the family Solanaceae. The domestication of chilli first occurred in Central America, most likely in Mexico, with secondary centres in Guatemala and Bulgaria (Salvador, 2002). Chilli was introduced to Europe by Columbus in 15 th century and spread to rest of the globe. In 17 th century Portuguese introduced it into India. It is an indispensable spice essentially used in every Indian cusine due to its pungency, taste, colour and aroma. Chilli fruits are rich sources of vitamin C, A and E. In this pivotal vegetable plant Capsicum provides evidence to improve the cytogenetical, morphological and yield parameters with the effect of EMS and DES.
MATERIALS AND METHODS
The dried seeds of chilli var. Kovilpatti 1 were treated with different EMS (10, 20, 30, 40 and 50 mM) and DES (5, 10, 15, 20 and 25 mM) concentration. Seeds were presoaked in distilled water for 12hrs to allow uptake of chemical mutagens. After treatment, seeds were thoroughly washed in running tap water for 4h to leach out the residual of chemicals. Untreated seed stock was used as a control. The treated and control seeds were sown in sand beds and watered at least once a day. After 25-30 days, seedlings were shifted to new pots as one plant per pot. The M 1 generation (produced directly from mutagen treated seeds) was grown in the pot culture experiment at the Botanical Garden, Department of Botany, Annamalai University. The recommended package of practice for the crop was followed. The M 1 plants were harvested on a single plant basis. From each entry, 10 plants were randomly selected for recording observations on important yield attributing characters on plant height, primary and secondary branches per plant, days to first flowering, fruit length (cm), fruit girth (cm), total number of fruits per plant, number of seeds per fruit, seed weight per fruit (g), 100seed weight (g) and pericarp: seed ratio.
1. Cytogenetical analysis
For the cytogenetical analysis, root meristems of chilli (2n = 24) were used. The chilli root tips about 3 cm in length were excised, fixed in glacial acetic acid: alcohol (1:3) solution for 48hr. Then root tip squashes were made by using iron alum, haematoxylin squash technique (Marimuthu and Subramanian, 1960). Cell divisions and cytogenetical abnormalities were observed and photographed under a Nikon image capturing system. The various types of cells with normal and abnormal chromosomal behaviour at various stages were observed and counted.
RESULTS AND DISCUSSION
Cytological studies revealed that the use of the use of chemical mutagens stimulated the mitotic activity in the roots of chilli, since the mitotic index increased with the increase in the concentration. The mitotic index value increased up to a certain level concentration. However, EMS treatments induced insignificant of mitotic abnormalities compared to control roots (Table 1). The data showed a significant increase in the percentage of prophase cells 34.7 % with using the 15 mM DES. All the concentrations were capable of inducing various types of chromosomal abnormalities in almost all the stages of mitosis. Sticky chromosomes, precocious movements, bridges, micronucleus, laggards and anaphase with polar deviation were the most common anomalies recorded with the use of EMS and DES in metaphase and anaphase (figures not shown). The percentages of chromosomal abnormalities in different mitotic stages were significantly higher than that of the control and calculated on mitotic index, frequency of phases and percentage of abnormalities in mitosis.
These chromosomal aberrations may consider as indicators of clastogenic effects of their inducers (Badr, 1983). This may indicate increase the impairment of mitotic apparatus, which was not completely inhibited. It is clear from our results that chromosomal stickiness is 20 ILNS Volume 10 the most dominant abnormality produced in different concentrations. Stickiness is a common physiological phenomenon, which may be the result of an action by the chemicals on chromatin fibres (Badr et al., 1987). It has been attributed to an action on the protein of chromosomes (El-Sadek, 1972). Chromosomal bridges may be due to chromosomal stickiness and subsequent failure of free anaphase separation or may be attributed to unequal translocation of chromosome segments (Najjar and Soliman, 1980). The seeds are good explants for chemical mutagens to create mutations in a genome of a cell. These mutagens affect the germination process in seeds. The percent of germination in seeds depends on the nature of the mutagen and its treatment dose. After mutagenesis, seeds show the effects of mutagen as modified morphological traits from disturbed physiological processes. The effect of chemical mutagens measured by reduction of germination and growth of seedlings decreased with increase conc. of EMS and DES in chilli. Constantin et al. (1976) observed linear relationship between conc. and reduction survival of field growth of soybean. The effect of mutagens was measured quantitatively by reduction in germination survival (lethality) [Ramasamy, 1973]. Changes in specific activity of enzyme (Endo, 1967) and reduction in productivity of IAA (Miura et al., 1974) and were also causes for reduced growth in the M 1 generation.
Observations showed decrease in plant height, number of pods per plant, number of clusters per plant, fruit length and fruit girth with increasing concentration of EMS and DES than control. In the present study, the reduction of these parameters was prominent in EMS and DES, such as inhibitory effects of various mutagens were reported in several other crops (Reddy et al., 1992). Koteswara Rao et al., (1983) reported that irradiation significantly reduced some polygenic characters in lengthy of pods, number of pods and number of clusters International Letters of Natural Sciences Vol. 10 in M 1 generation. Days to first flowering increased with increasing conc. of EMS and DES. However, number of primary branches per plant, number of pods per plant, decreased mean performance value with increasing dosage. The mutagenic effect was found decreasing in quantitative characters in soybean (Pavadai and Dhanavel, 2004) and cluster bean (Velu et al., 2007). Mutagen treatment causes complex genetic and physiological damages. The first generation (M 1 ) developed from treated seeds, for example, suffers from growth inhibition, may be partly sterile, and may lose many plants before flowering and seed set (Ojiewo et al., 2006a).
CONCLUSION
The cytological studies revealed that EMS and DES induced more aberrations. Chromosomal stickiness was the most common anomaly observed in root tips treated with the chemical mutagens. Magnitude of induced variation was found to depend upon the mutagen used, character under study and the genotypic background of the mutant. These promising mutant lines need to be further utilized in next generations to derive distinct lines with improved agronomic traits.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-29T09:47:59.475Z", "created": "2014-02-15T00:00:00.000", "id": "53125892", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://www.scipress.com/ILNS.10.18.pdf", "pdf_hash": "500491644e86d642d9875f31c42064a2529e2d6c", "pdf_src": "ScienceParsePlus", "provenance": "peS2o-0075.json.gz:38044", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "a55f93805a29c8d9894f1dd70b0147d1280fa86b", "year": 2014}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Correlation of Leaf NPK and Leaf Pigments of Coleus atropurpureus L. Benth during Vegetative and Generative Phases
Coleus atropurpureus L. Benth is a annual plant that has a distinctive leaf aroma and bitter taste. C. atropurpureus leaves contain phenolic compounds and antioxidants that can capture free radicals; free radicals play an important role in preventing various human diseases. A study was conducted to determine the correlation between leaf position (1st to 4th) at the vegetative and generative phases with leaf pigments, N, P, K, and total fl avonoid concentrations. The results showed that leaf chlorophyll a, chlorophyll b, total chlorophyll, carotenoid, anthocyanin, nitrogen, and total fl avonoids were higher in the vegetative phase. Therefore, C. atropurpureus is better harvested in the vegetative phase, and the 2nd leaf position can be used as indicator for N, K, pigments and total fl avonoid content.
Introduction
Indonesians are well known to use medicinal plants as a treatment for health problem. One of the medicinal plants that has been widely used is jawer kotok (Coleus atropurpureus L. Benth). Coleus atropurpureus an annual herbaceous plant which can grow up to 100 cm tall (Wiart, 2006). C. atropurpureus grows upright and has branches with square rod shapes and jagged leaf edges (Figure 1). The length of the leaf stalk can reach 7.5 cm with an oval leaf shape 5-10 cm long. Flowers are purplish, white, or bluish on the terminal stalks with a shape like nails arranged 10-20 cm long. The colorful C. atropurpureus leaves make the plants to be used as ornamentals. The colors of the leaves diff er with diff erent types and cultivars. According to Osman (2013) Coleus blumei with purplish red and red to dark red leaves contains high phenolic levels, which indicates it is potential as medicinal plant. The genus of Coleus belongs to Lamiaceae or Labiatae family; many species from this family can be used in traditional medicine. C. atropurpureus leaves are usually used to overcome dermatitis, post partum, abdominal pain, coughing and muscles pains, particularly by people in West Java (Roosita et al., 2008). In addition, its uses to cure bronchitis, asthma, angina, digestive disorders, animal bites (Suva et al., 2016), for dengue fever and malaria drugs in Philippines (Gascon, 2011), for hemorrhoids, antioxidants and anti-tuberculosis (Ahmad and Massi, 2014) have been reported. C. atropurpureus contain saponins, fl avonoids, alkaloids, polyphenols, quercetin and essential oils (Moektiwardoyo et al., 2011). The compounds which have antioxidant properties can capture free radicals and play an important role in preventing various chronic diseases (Gross, 2004). One of antioxidants is fl avonoid, which has been reported to inhibit proliferation of SP-C1 tongue cancer cells (Achmad et al., 2014). This study aims to determine whether or not diff erent leaf positions contain diff erent levels of secondary metabolite. The correlation between the position of leaves (1 st to 4 th ) and the leaf nutrient content (NPK) and secondary metabolites during the vegetative and generative phases were also determined.
Plant Materials
Coleus atropurpureus leaves used are 5 MAP (months after planting) which planted in IPB University experimental station, Bogor, West Java, Indonesia. The fully open leaves from the 1 st to the 4 th position from the shoot tip were collected for analysis (fi gure2).
Leaf Nutrient and Pigment Analysis
Analysis of nutrient levels of N, P, and K was carried out in the Testing Laboratory of the Department of Agronomy and Horticulture, IPB University. Total N analysis used Kjeldahl method, P was determined using a UV-1800 UV-VIS spectrophotometer, and K analysis used AAS (Atomic Absorption Spectrophotometry).
Analysis of chlorophyll, carotenoid, anthocyanin and total fl avonoids was carried out in the Postharvest Laboratory of the Department of Agronomy and Horticulture, IPB University. Analysis of chlorophyll, carotenoid and anthocyanin used Sims and Gamon (2002) method with the following protocol: 200 mg of fresh leaves were weighed, mashed with 2 ml acetic solution, centrifuged (6000 rpm, 10 minutes), 1 ml supernatant was added 3 ml of acetic acid then measured with wavelengths of 470, 537, 647 and 663 nm. The total level of fl avonoids was analyzed using the method of Chang et al (2002): 10 mg quercetin was dissolved in 80% ethanol then diluted to 25, 50 and 100 μg/ml. The diluted 0.5 ml solution was mixed with 1.5 ml of 95% ethanol, 0.1 ml of 10% aluminum chloride, 0.1 ml of potassium acetate 1M and 2.8 ml of distilled water. This solution was incubated at room temperature for 30 minutes, absorbance was measured at 415 nm. Blanks were made by replacing the amount of aluminum chloride with distilled water. An extract in 0.5 ml of ethanol was reacted with aluminum chloride to determine the fl avonoid concentration. According to Aziz (2015) production of bioactive compounds can be carried out with the following method: bioactive compound production = leaf dry weight (g per plant) x concentration of leaf bioactive compounds (%).
Data Analysis
Data were analyzed using R for t-student test and SAS 9.4 for Pearson correlation test.
Leaf Pigment at The Vegetative and Generative Phase
The leaf clorophyll a and clorophyll b at the vegetative phases is in Figure 3, and the highest content are in the 3 rd leaf. However, chlorophyll a and chlorophyll b levels in each leaf position showed fl uctuating results. Leaves at positions 1 st , 2 nd , and 4 th had 21.61%, 23.80%, and 22.37% lower chlorophyll a than the 3 rd leaf, whereas leaves at the 1 st , 2 nd , and 4 th had 20.16%, 22.62%, and 20.98% lower chlorophyll b than the 3 rd leaf. Leaves at position 1 st , 2 nd , and 4 th have total chlorophyll levels which were 21.14%, 23.40%, and 21.92% lower than the 3 rd leaf. The carotenoid levels in the 1 st and 3 rd leaves were not signifi cantly diff erent from each other, but were signifi cantly higher than the 2 nd and 4 th leaves. The 2 nd leaf showed a markedly lower carotenoid level of 12.97% and the 4 th leaf showed a markedly lower carotenoid level of from the 1 st leaf position.
Intan Annisa Respita, Sandra Arifi n Aziz, Ani Kurniawati The level of anthocyanin in the 1 st and 2 nd leaf was not signifi cantly diff erent, but was signifi cantly higher than 3 rd and 4 th leaves. At the 4 th leaf position there was an increasing anthocyanin levels of 10.81%. The level of leaf pigment at the generative phase is described in Figure 4, and it shows fl uctuating results. Chlorophyll a and carotenoid levels showed the highest results at the 4 th leaf position. The 3 rd position leaves have anthocyanin levels that are signifi cantly diff erent from the 1 st , 3 rd and 4 th leaf.
Leaf NPK Levels at The Vegetative and Generative Phase
The levels of N and P at the vegetative and generative phases were not signifi cantly diff erent from those in the generative stage ( Figure 5) but N levels in the vegetative phase was 0.22% higher. The highest P at the vegetative and generative phase was found in the 1 st leaf. Nutrient K from the 1 st and 2 nd leaf at the generative phase is greater than the vegetative phase. N levels of the 2 nd and 3 rd leaves, as well as the 3 rd and the 4 th leaves were not signifi cantly diff erent. Leaf P decreases along the leaf position and this occurs in the vegetative and the generative phases. In the generative phase, the levels of P of the 1 st , 2 nd , 3 rd leaf was 0.08%, 0.03%, and 0.003% higher than the 4 th leaf. The 1 st and 2 nd leaves had higher potassium levels than the 3 rd and 4 th leaves, but the levels of the 3 rd and 4 th leaves were similar. In the vegetative phase, the levels of n decreased with from the 1 st to 4 th leaves, the levels of the 1 st , 2 nd , 3 rd were 0.82%, 0.30%, and 0.24% higher than in the 4 th leaf. Similar trend was noticed in the generative phase, where the levels of N of the 1 st , 2 nd , 3 rd leaves were 0.51%, 0.24%, and 0.07% higher the levels of N of the 4 th leaf. In addition, the nutrient P levels in the generative phase was higher 0.27% compared to the vegetative phase.
The fi rst leaf had the highest K level in the vegetative and generative phase, and K levels was 2.40% higher in the generative compared to the vegetative phase. K level of the 1 st , 2 nd , and 3 rd leaves were 0.27%, 0.13%, and 0.18% higher than K level of 4 th leaf. In the generative phase, there was a decreasing K levels from the position of the 1 st to 4 th leaves; K levels of the 1 st , 2 nd , 3 rd leaves were 1.48%, 0.90%, and 0.03% higher than the 4 th leaf.
Total Flavonoid Concentration and Total Flavonoid Content
The highest total fl avonoids concentration in the vegetative and generative phases were found at the 1 st leaf ( Figure 6). In the vegetative phase, total fl avonoids concentration decreased according to the position 1 st to 4 th leaves; the levels of the 1 st , 2 nd , and 3 rd leaves were 49.00%, 16.50%, and 7.76% higher than the 4 th leaf. Similar trend was noticed during the generative phase; total fl avonoids of the 1 st , 2 nd , and 3 rd leaves were at 25.58%, 5.92%, and 3.23% higher than the 4 th leaf.
There was a fl uctuating total fl avonoid content in the vegetative phase from 1 st to 4 th leaf but the 4 th leaf had highest total fl avonoid content in both vegetative and generative phase (Figure 7). The total fl avonoids content was obtained from multiplication of total fl avonoids concentration to leaf dry weight. The 3 rd and 4 th leaves were larger and heavier than other leaves. The total fl avonoids content of the 1 st , 2 nd , and 3 th leaves were 0.39%, 0.77% and 0.25% lower than the 4 th leaf in vegetative phase. Total fl avonoid content decreased from the 1 st to 4 th leaves at the generative phase, which was 0.41%, 0.14%, and 0.17% lower Table 1 showed the correlation between leaf NPK with leaf chlorophyll a, chlorophyll b, total chlorophyll, anthocyanin, carotenoids and total fl avonoid of the fi rst to fourth leaves at the vegetative phase. N levels were positively correlated with chlorophyll a, total chlorophyll, carotenoid levels, at the 1 st , 2 nd and 3 rd leaves positions, and positively correlated with anthocyanin and total fl avonoid levels and levels in the 1 st , 2 nd , and 4 th leaf positions. In the 2 nd , 3 rd , and 4 th leaf N has a positive correlation with chlorophyll b. In the 1 st , 2 nd , and 3 rd leaf P has a positive correlation with chlorophyll a, chlorophyll b, carotenoids, total chlorophyll. In the 1 st , 2 nd and 3 rd leaf K was positively correlated with chlorophyll a, chlorophyll b, carotenoids, and positively correlated with anthocyanin in the 2 nd and 4 th leaf. A positive correlation between K and total content of fl avonoids were recorded in all leaf positions. Table 2 showed the correlation at the generative phase. N was positively correlated with levels of chlorophyll a, chlorophyll b, carotenoid, and total chlorophyll of the 4 th leaf , but positively correlated with anthocyanin of the 1 st , 2 nd , and 3 rd leaf Leaf N also positively correlated with levels and the total content of fl avonoids in the 1 st and 3 rd leaf. Leaf P has a positive correlation with the levels of chlorophyll a, chlorophyll b, total chlorophyll in the position of the 2 nd and 3 rd leaves, while with the carotenoids in the position of the 3 rd leaf. P nutrient content was positively correlated with anthocyanin at the 4 th leaf position, while positively correlated with the levels and total fl avonoid content in the 1 st and 2 nd leaf positions. Nutrient content K was positively correlated with levels of chlorophyll a, chlorophyll b at the 2 nd and 3 rd leaf position, while positively correlated with carotenoids at the 1 st , 2 nd , and 3 rd leaf positions. K nutrient levels were positively correlated with anthocyanin in leaf position 2 nd and 4 th , while positively correlated with total chlorophyll at the 3 rd leaf position. Nutrient K has a positive correlation with the level and total content of fl avonoids in the 1 st leaf position.
Discussion
The presence of pigments in the form of chlorophyll a and chlorophyll b play a role in absorbing solar radiation during photosynthesis. Photochemical processes can act to release electrons, so that light energy is converted into chemical energy. This level of chlorophyll can aff ect photosynthesis (Richardson et al., 2002). Photosynthesis rates are directly proportional to the concentration of photosynthetic pigments. Therefore, the position of leaf that close to apex will have a low pigment concentration. Marschner (2012) reported that in mature leaves ~ 15% of the volume of all cells is occupied by chloroplasts, cytoplasm and cell walls while the remainder is by vacuoles (85%). Therefore, in mature leaves the levels of bioactive compounds had higher levels of anthocyanin, chlorophyll and fl avonoids.
In croton concentration of chlorophyll a, chlorophyll b and total chlorophyll were found to be higher in older leaves, and the carotenoid level in the generative phase is lower than the vegetative phase (Gogahu et al., 2016). Research by Tjhia et al., (2018) reported that the levels of carotenoid in Vernonia amygdalina Del at the generative phase was higher than that the vegetative phase. Chlorophyll activity plays a role in the process of organogenesis which can aff ect the generative phase (Simova et al., 2001). All pigment levels in the vegetative phase are higher than the generative phase. Similarly for leaf weights. The level of anthocyanin in the generative phase decreases because in the early phase of this development anthocyanins are required to carry out photoprotection. Anthocyanin is needed because at this stage, chlorophyll cannot develop properly to absorb excessive sunlight. Decreased anthocyanin levels in the generative phase indicate that anthocyanin function might have been replaced by the presence of carotenoids (Hughes et al., 2007).
Higher nitrogen levels aff ect chlorophyll levels in each phase because nitrogen is one of the important component of chlorophyll (Marschner, 2012). Photosynthesis has a positive relationship with the growth process of all parts of the plant (Diem et al., . In cotton, increasing photosynthesis rate is aff ected by increasing CO 2 uptake, while CO 2 uptake is aff ected by ion concentration including K + , NO 3 -, PO 4 - (Longstreth et al., 1980). With a decrease in the concentration of nitrogen and phosphorus, plants experience stress and will respond with an increase in anthocyanin. Anthocyanin functions as an antioxidant which will free radicals when stress occurs (Scott, 1999).
Plants have young leaf parts that act as sinks and adult leaves act as sources (Marschner, 2012). The position of the leaf can indicate the direction of the nutrient translocation path and the water for the plant sink towards the source. Usually, the young leaves have higher nutrient levels and become strong sinks. Nitrogen concentration in both vegetative and generative phases decreased from the 1 st to 4 th leaf position, and potassium concentration decreased in the generative phase but the level is still high.
A decrease in nitrogen in the tissue causes a decrease in protein and chlorophyll content. Munawar (2011) reported that nitrogen plays an important role for plants. Nitrogen is involved in the synthesis and transfer of energy, plant growth, improves leaf quality, seed and fruit production, and plays a role in the preparation of amino acids, proteins, chlorophyll, nucleic acids and co-enzymes.
There is a relationship between the availability of nutrients and the accumulation of fl avonols; reduction of nitrogen will increase fl avonol levels. But in the availability of high nitrogen, phosphate reduction can facilitate the formation of fl avonols. Nitrogen defi ciency in tomato plants produces accumulation of fl avonol in adult leaves. Conversely, when phosphorus defi ciency causes accumulation of fl avonol at the beginning of fruit ripening (Stewart et al., 2001). Karimuna et al. (2015) reported the total fl avonoids of Murraya paniculata leaves were negatively correlated with potassium at diff erent leaf ages and positions. Potassium concentrations can be categorized very high (3.59-4.10%), phosphorus was high (0.28-0.29%) or very high (0.33-0.35%).
Conclusion
Coleus atropurpureus leaf chlorophyll a, chlorophyll b, carotenoids, anthocyanin, total chlorophyll, nitrogen, total levels of fl avonoids and total fl avonoid content at the vegetative phase were higher than those at generative phase. The indicator leaf at the vegetative phase which have positive correlations with pigment and total fl avonoids are the second leaf, whereas all leaf positions at the generative phase do not have correlations with the leaf NPK, leaf pigments, and leaf total fl avonoids.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-11-28T12:48:36.898Z", "created": "2019-10-24T00:00:00.000", "id": "213165703", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBYSA", "oa_status": "GOLD", "oa_url": "https://doi.org/10.29244/jtcs.6.03.174-181", "pdf_hash": "b09731bbb422f843dd0bee23b9a178156eea11eb", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:35080", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "2f34900c6f54823c0b225e39b654897de720938c", "year": 2019}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
IMPACT OF GENETICALLY IMPROVED FISH SPECIES AND TECHNOLOGY ON SELECTED HATCHERY AND FISH PRODUCTION IN BANGLADESH
The study was carried out in IAPP commanding areas from July to September 2015. A total of 8 hatchery and 240 farmers were selected for this study from Rangpur and Barisal region. About 153% Tilapia production increased which was from 34 to 86 lakh, which was 148% in Rangpur district. Thai koi production was increased about 320% in Rangpur and it was 152% in Barisal. It was observed that, per hatchery Tilapia profit was Tk. 17.35 lakh and Tk. 17.18 lakh in Rangpur and Barisal, respectively. While, total profit was 3.9 times more for Thai koi in Rangpur and it was about 1.7 times more in Barisal after IAPP-BFRI project implementation. Impact of improved germplasm on grow out system was estimated. Finding shows that before IAPP-BFRI project the average harvesting weight of tilapia fish was 122g but after using IAPP-BFRI germplasm, it increased to 194g in Rangpur district. In case of Thai Koi, the harvesting weight gain was 26% in Rangpur district and it was statistically significant at 1% level. Survey results also show that per acre profit was only Tk.86671 for Tilapia farming before IAPP whereas it was increased to Tk. 234853 after IAPP-BFRI intervention. At the same time, profit from Thai Koi was increased about 189% after IAPPBFRI activities. Similarly, profit was increased about 86% in case of Pangus farming and this positive impact was statistically significant at 1% level. Therefore, it may conclude that, farmers can significantly increase Tilapia, Thai Koi and Pangus production as well as can maximize profit using IAPP technology.
Introduction
Bangladesh is endowed with unique aquatic resources for fish and fisheries development. The inland water resources of this country are considered to be one of the richest in the world both in terms of area and its potential for fisheries development. Fisheries sector is contributing in food security through proving safe and quality animal protein; almost 60% animal protein comes from fish. It contributes 3.69% to our national GDP and around one fourth (23.12%) to the agricultural GDP (FRSS, 2016). It provides full time employment to about 1.20 million people and generates part time employment for 11 million people. Around 10% of total export earnings come from fisheries. It was argued that rural aquaculture plays a significant role in Bangladesh in view of its contribution towards national food and nutritional security, rural job opportunities and income (Mazid and Sinha, 2000). Rural aquaculture thus needs much more attention now than ever before. Although Bangladesh is rich in vast water resources but per hectare yield from ponds only 3.0 tones per year. This can further be augmented. The yield of inland fisheries is therefore could be increased substantially through adopting appropriate aquaculture technologies and judicious as well as efficient uses of existing resources. However, sustainable aquaculture technologies as well as quality fish fry are required in this country for increasing production of fisheries.
Integrated Agricultural Productivity Project (IAPP) is donor funded project. This project is implemented by Ministry of Agriculture and Ministry of Fisheries and Livestock, Bangladesh. Under this project eight agencies (DAE, DLS, DOF, BADC, BARI, BRRI, BFRI and SCA) are working together to promote productivity of crops, livestock and fish. During 2011-16, with the financial assistance GAFSP, BFRI has been working for developing improved generation of GIF tilapia, Thai Koi and Pangus. During the timeframe, BFRI have successfully developed 04 generation of GIF tilapia, 03 generation of Thai Koi and 02 generation of Pangus through selective breeding. These improved generations were distributed to eight selected private hatcheries of Rangpur and Barisal region for mass seed production. These selected hatcheries are being produced millions of quality fry from BFRI supplied improved generation and sold out among IAPP fisheries group member. The IAPP group members were growing table sized fish for economic welfare. The present study was undertaken to analyze the potential impact of quality fry production in hatcheries and culturing the improved generations on fish production and profitability of fish farming.
Materials and Methods
The investigation was carried out in IAPP areas from July to September 2015. Integrated Agricultural Productivity Project (IAPP) worked in 54 upazillas in Rangpur, Kurigram, Lalmonirhat, Nilphamari, Barisal, Jalokathi, Patuakhali and Borguna. The survey schedule was prepared to record the required information on various aspects of fish farming activities. A total of 8 hatchery and 240 farmers were selected from Rangpur and Barisal region for this study. After the schedule was finalized, the selected farmers were interviewed individually by personal interviews when they had little work on the farms. Before the actual interview was made, the aim and purpose of the study were explained to each hatchery operator and farmers.
Analytical technique
Data, thus collected, were analyzed in accordance with the objectives of the study. In this study, the tabular technique of analysis was used. The tabular technique of analysis included classification of the data in the form of tables. It is generally used to find out the crude association of difference between two sets of variables. This technique is based on arithmetic mean, percentage, ratio, etc.
Results
Table 1 shows the average fry production and its profitability status in Rangpur and Barisal district. Fry of Tilapia production has reached to about 87 lakh per hatchery per year after IAPP implementation, which was only 35 lakh before the project implying that Tilapia fry production has increased about 148% in Rangpur district. Same success was observed in Barisal. About 153% Tilapia production has increased which is from 34 lakh to 86 lakh. On the other hand, a significant change has occurred in case of Thai koi in Rangpur district. Thai koi production has increased about 320% in Rangpur and it was 152% in Barisal. Cost of production has increased with fry production but profitability has increased significantly for both places and species. It is found that per hatchery Tilapia profit was Tk. 17.35 lakh and Tk.17.18 lakh in Rangpur and Barisal, respectively. On the other hand, total profit was 3.9 times more for Thai koi in Rangpur and it was about 1.7 times more in Barisal after IAPP implementation. Therefore, it can be concluded that, a significant success has achieved in case of fry production and profit in hatchery. It was found that after IAPP implementation, fertilization rate for Tilapia fingerling has increased about 13 and 16 percent in Rangpur and Barisal district (Table 2). It was 11 percent and 5 percent for Thai Koi in Rangpur and Barisal respectively. On the contrary, hatching rate has increased for Tilapia from 84 percent to 95 percent in Rangpur and from 82 percent to 90 percent in Barisal. Survival rate shows the significant improvement which was increased about 8 to 9 percent for Tilapia and for Thai Koi it was increased about 9 percent after IAPP implementation. Table 4 shows the survival rate of fry and average weight gain of fry after treatment during primary nursing stage. It was found that after IAPP survival rate of Tilapia increased about 16 percent and 8 percent in Rangpur and Barisal, respectively during nursing period. On the other hand, survival rate was very low (55%) before the IAPP project for Thai Koi in Rangpur but it reached to 67 percent after commencement of the project. In Barisal survival rate was good (82%) and it also increased to 88 percent after project implementation. In Rangpur, about 180 percent weight has gained for Tilapia after project implementation and it was 140 percent in Barisal.
On the other hand, weight has gained about 88 percent for Thai Koi in both places. Table 5 shows how many people were trained through this project and how much area has increased due to project implementation. It was found that after project implementation, average hatchery staff increased about 60 percent and 80 percent in Rangpur and Barisal district, respectively. Through this project, a total of 6 hatchery staffs were trained in Rangpur and it was 8 staff in Barisal. In case of physical structure, after project implementation, hatchery area was increased from 8 acre to 13 acre in Rangpur district but there no change was observed in Barisal. On contrary, breeding pond area has increased about 66 percent and 50 percent in Rangpur and Barisal, respectively. On the other hand, nursery area has increased from 5 acre to 7 acre in Rangpur and it has increased from 4 acre to 7 acre in Barisal.
Grow out Results
Impact of IAPP was estimated in various steps of production such as harvesting weight gain, production gain, survival rate increased, production cost increased or decreased and profit earn. Tilapia and Pangus species was used to assess the impact of IAPP technology in Barisal district.
Result shows that before IAPP, on an average per tilapia weight was 127 g at harvesting time but it grew up to 170 gm after IAPP. It implies that about 34% harvesting was gain from per tilapia after using IAPP technology and this gain was statistically significant at 1% level. On the other hand, about 40% harvesting weight was gain (from 741 g to 1035 g) from per fish after using IAPP technology in case of pangus species and this gain was significant at 1% level.
To assess the impact of IAPP on productivity, per acre fish production was estimated for Tilapia, Thai Koi and Pangus in Rangpur and Barisal district (Table 7). Result reveals that per acre Tilapia fish productivity increased 1.99 times after using IAPP technology and it has reached to 4799 kg acre -1 . On the other hand, a remarkable change has been achieved in Thai Koi. Before IAPP, farmers were able to produce about 2002 kg Thai Koi in one acre of pond area but after using IAPP technology, farmers produced 3655 kg Thai Koi within the same area of pond implies that 82% productivity can be increased through using IAPP technology in case of Thai Koi. This productivity gain was statistically significant at 1% level for both species of fish. On the other hand, farmers were able to increase production about 28% in case of Tilapia after using IAPP technology in Barisal district. The productivity increased from 2489 kg acre -1 to 3195 kg acre -1 , which was significant at 1% level. A notable achievement was found in case of Pangus in Barisal. Result reveals that on an average Pangus productivity has increased about 43% after using IAPP technology. Before IAPP technology, farmers produced only 7810 kg Pangus in one acre of pond area but it has increased to 11176 kg after IAPP project.
Before, IAPP project, the survival rate was 82% and 77% for Tilapia and Thai Koi respectively in Rangpur district but after IAPP project, it has increased to 91% and 89%, respectively (Table 8). On an average, survival rate was increased 10% and 15% for Tilapia and Thai Koi, which was statistically significant at 1% level. On average, 25% survival rate has increased for Tilapia which was 72% before IAPP but after IAPP is increased to 90%. On the other hand, Pangus survival rate has increased about 20% (from 74% to 89%) and it was statistically significant at 1% level. This increase of survival rate was lead to increase the productivity and profitability of the farm.
In this section, we only show the comparative scenario of production cost for before and after IAPP. Finding shows that, production cost was Tk. 56 for producing per kg of Tilapia fish before IAPP but after project, it increased to Tk. 66 implies that production cost has increased about 17% for Tilapia in Rangpur region. Alternatively, cost for per kg Thai Koi production has increased about 15% (from Tk. 70 to Tk. 81) and these increases was statistically significant (Table 9). On the other hand, the percentage of increase of profit is comparatively lower in Barisal than Rangpur region. In case of Tilapia, per acre profit increased about 1.65 times which was statistically significant at 1% level. Similarly, profit increased about 86% after using of IAPP technology in case of Pangus farming and this positive impact was statistically significant at 1% level. Therefore, it may conclude that, farmers can significantly increase Tilapia, Thai Koi and Pangus production as well as can maximize profit using IAPP technology.
Discussion
Before involvement with IAPP, most of the farmers stocked fry from local fry traders. But after involvement with IAPP, farmers stocked fry in their ponds from selected IAPP assisted hatchery to ensure good quality fish fry for increased fish production. After engaged with IAPP, most of the farmers of followed BFRI evolved technologies to get more fish production. Prior to IAPP, the farmers' average fish production of Tilapia, Koi and Pangus were 2,451, 2,002 and 7,810 kg acre -1 , respectively. After involvement of IAPP intervention, the farmers' fish productions were 3,997; 3,655 and 11,176 kg acre -1 , respectively. Islam and Haque (2010) stated that the average yield was 1,144 kg acre -1 in Northwest Fisheries Extension Project, which was remarkable less than the present study.
According to FRSS (2016), the national average pond fish production in Bangladesh was 1,558 kg acre -1 . From the above, fish production was being increased because of the involvement with IAPP and farmers used genetically improved fry and BFRI evolved culture technology for their culture management.
Fish farming has a positive impact on aquaculture production but numerous types of constraints affect potentiality of fish farming in the Rangpur and Barisal region. Very poor retention capacity of soil, sudden tidal flow, flash flood, sudden cyclone, lack of loan facilities, low quality and scarcity of fish seeds in proper time etc. Rahman (2003) stated that the major constraints of carp farming were lack of money and production cost. Khan et al. (1998) identified that the lack of knowledge about fish culture was one of the most important problems. Hossain et al. (1992) observed that the largest problems faced by the fish farmers are multiple ownerships.
Table 1 .
Average fry production status.
Table 2 .
Average survival status during incubation.
Table 3 .
Average survival status of fry during hormone treatment.
Table 4 .
Average survival status during primary nursing stage.
Table 6 .
Average harvesting weight (g) of Rangpur and Barisal region.
Table 8 .
Average survival rate of Rangpur and Barisal region.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-15T05:52:57.318Z", "created": "2017-07-25T00:00:00.000", "id": "55642110", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.banglajol.info/index.php/IJARIT/article/download/33323/22442", "pdf_hash": "5a3145e0e440998f30ed3f29ac199c7dd5ab7174", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:30493", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "5a3145e0e440998f30ed3f29ac199c7dd5ab7174", "year": 2017}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
IN-VITRO STUDIES OF SOME SELECTED BOTANICALS AND BAU-BIOFUNGICIDE ON MYCELIAL GROWTH AND CONIDIAL GERMINATION OF Cercospora arachidicola AND Cercosporidium personatum
Efforts have been made to assess some plant extracts namely, Lycopersicon esculentum, Tagetus patula, Achras sapota, Azadirachta indica, Datura metel, Cymbopogon citrates, Polyalthia longifolia, Allium sativum and Allium cepa in vitro for the management of leaf spot (tikka) disease of groundnut cultivar Dhaka-1 caused by Cercospora arachidicola and Cercosporidium personatum. Results indicated that all the tested plant extracts and BAUBiofungicide suppressed the growth of mycelium and inhibition of conidial germination of C. arachidicola and C. personatum. Among the treatments, the leaf extracts of L. esculentum showed the most effective followed by leaf extract of D. metel, A. indica and BAU-Biofungicide in case of mycelial growth and conidial germination. Other plant extracts also had inhibitory effects. In case of conidial germination and germination inhibition, the least effective plant extract was C. citrates. Leaf extract of A. sapota was the least effective in case of mycelial growth.
Introduction
Groundnut (Arachis hypogaea L.) is an important annual legume crop belonging to the family Fabaceae growing in many tropical and subtropical countries of the world (Wudiri and Fatoba, 1992). It is a multipurpose and highly nutritious crop containing oil, food and its foliage or haulm provides a valuable fodder for livestock. More than 40 fungal diseases attack groundnut (Jackson and Bell, 1969), but in Bangladesh, the crop is subjected to attack by twenty one diseases (Talukder, 1974;Ahmed and Hossain, 1985). Among the diseases, early leaf spot caused by Cercospora arachidicola and late leaf spot caused by Cercosporidium personatum are the most devastating and economically important foliar fungal disease and major yield reducing factor of groundnut worldwide (Backman and Crawford, 1984;Khaleque, 1985). The early and late leaf spots of groundnut although caused by two fungal species, C. arachidicola and C. personatum, they are commonly referred together as Tikka disease. This disease results in early defoliation thereby affecting the pod formation (Worthington and Smith, 1973). Loss in pod yield due to the diseases was recorded as 70% in groundnut (Subrahmanyam et al., 1980). The yield loss was calculated in the groundnut variety Dhaka-1 due to early and late leaf spot (Tikka) by over 30-48% in Bangladesh (Hossain et al., 2005). In addition to direct yield loss, they hamper seed quality by reducing seed size and seed weight (Souta, 1912;Arthur, 1929) and oil content (Gupta et al., 1988). Severe attack of the disease resulting in heavy defoliations of groundnut leaves (Harrison, 1969). The disease can be controlled by developing resistant varieties, seed treatment with non-conventional chemicals (Maiti et al., 2005), spraying fungicides, influence by sowing times (Naidu and Vasanthi, 1995), by using indigenous medicinal plants and biological control means (Kishore and Pande, 2005). The most acceptable method for controlling this disease is cultivation of resistant variety but there was no absolute resistant variety in the world (Wynne et al., 1991). Increasing concerns about environment hazards caused by excessive usage of chemical fungicide, necessitates the development of more economical and ecofriendly alternative components of disease management. Biological control represents a less cost, environmental friendly natural and ecological approach for controlling diseases that reduces chemical inputs and their effect on environment. The main purpose of the study was to evaluate selected nine plant extracts for their in vitro antifungal activity against C. arachidicola and C. personatum.
Materials and Methods
Laboratory experiment was carried out at Disease Resistant Laboratory, Department of Plant Pathology, Bangladesh Agricultural University, Mymensingh following Completely Randomized Design (CRD) with four replications for bioassay of nine botanical extracts, BAU-Biofungicide (Trichoderma based preparation) and Bavistin (Carbendazim) against mycelial growth of C. arachidicola and C. personatum. The treatments were also tested in the present experiment following cavity slide technique to find out their effect on conidial germination of C. arachidicola and C. personatum.
Preparation of plant extracts
Plant extracts were prepared from fresh leaves of tomato (Lycopersicon esculentum), marigold (Tagetus patula), sapota (Achras sapota), neem (Azadirachta indica), datura (black) (Datura lemongrass (Cymbopogon citrates) and debdaru (Polyalthia longifolia) and bulb of garlic (Allium sativum) and onion (Allium cepa). The collected fresh leaf samples of different plants were washed in running tap water to make free from dust. The plant materials were cut into small pieces and extracts were prepared by grinding in a mortar and pestle followed by crashing plant parts in an electric blender with sterilized distilled water at different doses. The mixtures were filtered through linoleum cloth. The extracts were kept in conical flasks separately before use. Concentration (weight/volume) of tomato leaf, datura leaf (black) and sapota leaf was 20%, neem leaf, marigold leaf, garlic clove and onion bulb was 25%. Lemongrass, BAU-Biofungicide and Bavistin were use tested at 1:2, 2.5% and 0.1% concentration, respectively.
Bioassay of botanicals, BAU-Biofungicide and Bavistin on C. arachidicola and C. personatum
Carrot leaf extract agar medium was prepared and poured into 90 mm Petri plates at 20 ml/plate. After solidification, three 5 mm discs of the medium were scooped from three places maintaining equal distance of 4 cm from the centre using a sterilized disc cutter. One millimeter of each of plant extracts and suspensions of BAU-Biofungicide and Bavistin was put into each hole and the plates were stored overnight in refrigerator for diffusion of the test materials into the medium surrounding the hole. Next day, the plates were inoculated at the center with 6 mm blocks of 15 days old culture of C. arachidicola and C. personatum. Three plates (replications) were maintained for each material. Control plates did not receive any material. To prevent contamination, the plates were covered with the Para film and the plates were incubated at 24 ± 1 O C.
The isolates of C. arachidicola and C. personatum were grown on carrot leaf extract in Petri plates for 12 days at room temperature (24 ± 1 O C). The culture plates were kept under NUV light for 3 days for maximum production of conidia. Conidia were collected from the plates by scraping with a sterilized glass slide and conidial suspension was prepared in sterilized distilled water. The suspension was filtered through twoply cheese cloth to remove mycelial fragments and lumps of agar. The concentration of conidia in suspension was adjusted to 2x10 4 per milliliter using a hemocytometer (Krishna et al., 2001).
Observation was made regularly to record the mycelial growth. After inoculation periods of 10, 13 and 15 days the linear growth of mycelial of C. arachidicola and C. personatum was measured (McKeen et al., 1986;Nene and Thaplial, 1993;Islam et al., 2001) and percent inhibition of growth was calculated using the following formula as suggested by Sundar et al. (1995): Where, X = Mean mycelial growth (radial) of pathogen in control plate Y = Mean mycelial growth (radial) of pathogen in treatment
Growth inhibition of C. arachidicola and C. personatum.
To assay the antifungal activity of the botanical extracts, Bavistin and BAU-Biofungicide, 50µl of conidial suspension and equal volume of the suspension of test materials transferred to the well of each cavity slide and mixed thoroughly. Three single cavity slides (replications) were used for each treatment. All slides were kept in a humid chamber prepared by lining 90 mm diameter Petri dishes with wet tissue paper and incubated in the dark at 24± 1 O C. The slides were directly observed under light microscope for conidial germination at 24, 48 and 72 hours after incubation. Immediately after inoculation, a drop of lacto phenol-cotton blue was added to each well to prevent further germination of conidia. Fifty conidia in a well were observed under a compound microscope and number of germinated and non-germinated conidia was counted. Percentage inhibition of conidial germination in each treatment was calculated from the formula: Percentage inhibition = (number of conidia germination in control-number of conidia germination in treatment/ number of conidia germination in control ×100 (Krishna et al., 2001).
Effect of nine different selected botanicals, Bavistin and BAU-Biofungicide on mycelial growth of C. arachidicola and C. personatum
The efficacy of different nine selected botanicals, Bavistin and BAU-Biofungicide were evaluated against mean mycelial growth and growth inhibition of C. arachidicola and C. personatum at 10, 13, 15 days after inoculation (DAI). The results are presented in Table 1.1993), Srivastava and Lal (1997).
Effect of nine different selected botanicals, Bavistin and BAU-Biofungicide on conidial germination of C. arachidicola and C. personatum
The comparative effects of different nine selected botanicals, Bavistin and BAU-Biofungicide on conidial germination of C. arachidicola and C. personatum are presented in Table 2. All the botanicals, Bavistin and BAU-Biofungicide significantly reduce germination of conidia of the pathogens in the well of cavity slides compared to control (only water) at 24, 48 and 72 hours after incubation (Table 2).
At 24 hours of incubation, conidial germination ranged 0.0 to 68.0%, where the lowest conidial germination was observed in leaf extract of Lycopersicon esculentum, which was followed by leaf extract of Datura metel, Bavistin and leaf extract of Azadirachta indica. The highest conidial germination was recorded in untreated control. The extracts of onion bulb, Tagetus patula, Achras sapota and Cymbopogon citrates showed statistically similar effect on conidial germination after 24 hours of incubation. Among the treatments, leaf extract of Lycopersicon esculentum showed maximum (100%) inhibition of conidial germination. All the botanical extracts and BAU-Biofungicide inhibited conidial germination over 80% except leaf extract of Tagetus patula, Cymbopogon citrates and bulb of Allium cepa. BAU-Biofungicide, Bavistin, leaf extract of Datura metel and Azadirachta indica showed statistically similar effect on the inhibition of conidial germination (Table 2). In case of inhibition of conidial germination, the highest per cent of inhibition was observed in leaf extract of Lycopersicon esculentum, whereas the lowest was found in plain water (control).
At 72 hours of incubation, conidial germination was reduced significantly over control under all the treatments with plant extracts, Bavistin and BAU-Biofungicide. Conidial germination ranged 8.0 to 100.0%, while the lowest conidial germination was recorded in leaf extract of Lycopersicon esculentum followed by Bavistin, leaf extracts of Datura metel, BAU-Biofungicide, leaf extracts of Azadirachta indica and Polyalthia longifolia. The highest (100.0%)conidial germination was observed in plain water (control). The conidial germination inhibition ranged 0.0 to 92.0%, where maximum and minimum conidial germination inhibition was recorded from the wells containing leaf extract of Lycopersicon esculentum and water (control), respectively (Table 2). The findings of the present research work are in consonance with the findings of Natarajan et al. (2005), Kishore and Pande (2005), Abdulrahman and Alkhail (2005), Aage et al. (2003).
Among the treatments, the most effective was the leaf extracts of Lycopersicon esculentum followed by Datura metel, BAU-Biofungicide, leaf extract of Azadirachta indica in case of mycelial growth and conidial germination. Other plant extracts also had inhibitory effects but not as much as the leaf extracts of Lycopersicon esculentum, Datura metel, BAU-Biofungicide, leaf extract of Azadirachta indica. Leaf extract of Achras sapota was the least effective against mycelial growth of C. arachidicola and C. personatum. In case of conidial germination and germination inhibition the least effective plant extract was Cymbopogon citrates. The potentials of these plant extracts for pathogen control have not been fully realized largely because the experiment was performed in vitro. However, their effectiveness in field condition could be of a potential advantage as it will help to determine the in vivo inhibitory effect of the botanicals and BAU-Biofungicide.
Table 1 .
In-vitro evaluation of nine different selected botanicals, Bavistin and BAU-Biofungicide on mycelial growth of C. arachidicola and C. personatum
Table 2 .
Effect of nine different selected botanicals, Bavistin and BAU-Biofungicide on conidial germination of C. arachidicola and C. personatum following cavity slide method
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-03-16T13:12:48.242Z", "created": "2014-02-02T00:00:00.000", "id": "7873275", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.banglajol.info/index.php/IJARIT/article/download/17842/12503", "pdf_hash": "d963f417c51e550f01a2b464ba73be410d8fc9e1", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:39841", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "d963f417c51e550f01a2b464ba73be410d8fc9e1", "year": 2014}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
In Vitro Evaluation of the Antimethanogenic Potency and Effects on Fermentation of Individual and Combinations of Marine Macroalgae
Contribution of ruminants to total greenhouse gas emissions in Australia is approximately 10% and likely to increase with demand for livestock products, thus an efficient method of mitigation must be implemented. The red marine macroalgae Asparagopsis taxiformis reduces enteric methane production by up to 99% in vitro. Other macroalgae with less potent antimethanogenic properties may complement inclusion of Asparagopsis in livestock feeds. Adoption of environmental based changes in livestock systems must provide benefits to producers if change in management is to be adopted. This study used 72 h in vitro fermentations with rumen inoculum to characterize and rank seven species of macroalgae at low inclusion that previously demonstrated some degree of antimethanogenesis at higher inclusion concentration. The seven were assessed at 5% inclusion (OM basis) and in combination with Asparagopsis to evaluate beneficial effects on fermentation. When tested individually, improvements in volatile fatty acids were generally observed, however, minimal effect on gas production and no clear justification for a ranking order were demonstrated. When tested in combination with Asparagopsis, the effects on fermentation were dominated by presence of Asparagopsis at 2% and no further benefits demonstrated. Therefore, Asparagopsis remains the only macroalga inducing near elimination of methane in vitro and benefit of combinations with other macroalgae evaluated in this study was not demonstrated. However, combination with high protein macroalgae is proposed to provide productivity enhancement during seasonal lows in grass quality and thus reduce methane emissions intensity providing a stronger conduit for environmental responsibility while increasing productivity.
Introduction
A large proportion of methane (CH 4 ) emitted into the atmosphere derives from agriculture, and specifically ruminant enteric fermentation which contributes approximately 28% of global anthropogenic CH 4 emissions [1].
Ruminants rely on a complex rumen microbial consortium of bacteria, protozoa, archaea, fungi, and bacteriophages [2] associated with fermentation of feed. Fermentation of fibrous and nitrogenous feedstuff results in the production of volatile fatty acids (VFA) and microbial protein, used by the animal for growth, metabolism, and productivity [3]. During fermentation, carbon dioxide (CO 2 ) and hydrogen (H 2 ) are utilized by methanogenic archaea in the reductive methanogenesis pathway, which reduces partial H 2 pressure, but also results in CH 4 being emitted into the atmosphere [3]. Low CH 4 producing ruminants tend to be more productive and increasing productivity of animals could also lessen CH 4 emissions [4]. As a result, many strategies are being evaluated to mitigate enteric methanogenesis, including vaccines, bacteriocins and probiotics, bacteriophage therapy, genetic selection, feeding management and feed additives, and plant secondary metabolites [4].
Feed management and additives such as high-quality forages, grains, ionophores, fats, yeasts, enzymes, microbes, plant extracts and algae have the potential for CH 4 abatement [5]. Algae products can improve ruminant health and productivity [6] [7], increase feed quality [8], and inhibit methanogenesis [9]. However, reducing enteric methanogenisis is challenging, and to be adopted as a methodology it needs to be sustainable, practical, economically viable, and improve animal productivity [4]. Recently, the marine macroalgae Asparagopsis taxiformis ([Delile] Trevisan de Saint-Léon 1845; Asparagopsis) has demonstrated effective inhibition of CH 4 production in vitro at a dose of 2% of substrate organic matter (OM). In addition, Asparagopsis maintains the apparent in vitro degradability of OM (IVD-OM), and increases propionate but with a concomitant decrease of acetate [10] [11]. These rumen fermentation parameters are important indicators of fermentation efficiency [12]. Some other marine macroalgae that also reduce CH 4 , but to a lesser extent than Asparagopsis, have demonstrated better anaerobic fermentation in vitro compared to equal supplementation (17%) with cotton seed meal by increasing the concentration of total VFA (TVFA) when included with a low-quality grass substrate [13]. Therefore, it may be possible to improve rumen fermentation using combinations of macroalgae at practical dose concentrations when paired with Asparagopsis by increasing VFA production and improving the VFA profile.
It was hypothesised that macroalgae demonstrating CH 4 abatement in vitro at high dose concentrations would maintain antimethanogenesis at variable potency at low dose concentrations, and that when combined with the highly potent Asparagopsis they would enhance rumen fermentation. The aim of Experiment 1 was to rank seven select macroalgae for their potential as antimethanogenic feed additives using dose concentrations practical for livestock feeding, and establish their effects on fermentation parameters. The aim of Experiment 2 was to evaluate combining the seven macroalgae with Asparagopsis to determine the effect of the combinations on methanogenesis and potential benefits to VFA production.
Selection and Preparation of Macroalgae and Rhodes Grass Substrates
The macroalgae species were selected to represent the major groups of marine macroalgae (red, brown, green) based on their demonstrated ability to decrease enteric CH 4 andimprove fermentation [13] and are listed in gross energy quantification were determined as previously described by [13]. Crude protein of the macroalgae was based on total nitrogen (wt %) content using a nitrogen factor of 5 for the macroalgae [15] and 6.25 for the Rhodes grass.
Research Animals and Preparation of Rumen Fluid Inoculum
Rumen fluid inoculum (RF) was collected from four fistulated Brahman steers (Bos indicus; LW 460 ± 20 kg) fitted with 10 cm Bar Diamond (Parma, OH, USA) rumen cannulas. The steers were maintained at the College of Public Health, Medical and Veterinary Sciences of JCU according to current guidelines [16] and approved by the CSIRO animal ethics committee (A5/2011). The steers were maintained on Rhodes grass hay ad libitum for 6 months before the collection of RF, which was extracted 2 h after morning feeding by sampling from four quadrants of the rumen and hand squeezing to completely fill pre-warmed 1 L stainless steel thermos flasks.
Inoculation of in Vitro Fermentations
The RF was pooled and immediately processed by filtration through a 0.5-mm sieve and combined with incubation buffer (GVB) [17] at a ratio of 1:4 (RF:GVB
Experimental Design
Experiment 1 was conducted to rank macroalgae according to their antimethanogenic potency and effects on rumen fermentation parameters. This was accomplished using in vitro fermentation batch culture (Ankom) to determine effects of macroalgae inclusion on total gas (TGP), CH 4 , and VFA production, and IVD-OM. A series of four incubation periods were completed and consisted of 1.0 g of Rhodes grass substrate as control and appropriate quantities of each macroalga to achieve a dose concentration of 5% of the substrate OM according to the biomass composition described in Table 1.
Fermentations were also completed as controls (no macroalgae), a positive-control (Pcontrol) of Asparagospis at 2%, and RFB blanks. The inclusion of the macroalgae at 5% was determined as an arbitrary feasible level of feeding for livestock and based on results previously described for inclusion approaching 20% [13]. The inclusion of Asparagopsis at 2% was set based on previously determined optimum dose for this sea- weed in rumen fermentations [10] [11]. The seven macroalgae and Asparagopsis were randomly assigned to the four incubation periods (n = 3), and controls and blanks were included in all periods (n = 4). Within each incubation period, there was duplication of each macroalga, and controls and blanks at each sampling time point (12,24,48, and 72 h).
Experiment 2 was conducted to determine if the seven macroalga when combined with Asparagopsis studied in Experiment 1 would benefit in vitro rumen fermentation.
This was accomplished using the same fermentation techniques as Experiment 1. A series of five incubation periods were completed and consisted of 1.0 g of Rhodes grass substrate as control and appropriate quantities of each macroalga to achieve a dose concentration of 5% (OM basis) in combination with Asparagopsis at 2% according to the biomass composition described in Table 1. Fermentations were also completed as controls (no macroalgae), a P-control of Asparagospis at 2%, and RFB blanks. The seven combinations and Asparagospis were randomly assigned to the five periods (n = 3), and controls and blanks were included in all periods (n = 5). Within each incubation period, there was duplication of each macroalgae combination, and controls and blanks at each sampling time point (12,24,48, and 72 h).
Total Gas and Methane Production
The fermentation methods used in this study were similar to [18] but using the Ankom RF1 system and gas analysis as described by [10]. Pressure accumulation in the incubation bottles was measured continuously and recorded every 20 min. The maximum fermentation pressure inside the incubation bottle was set to 3 psi which, when exceeded, caused venting for 250 ms and the pressure change accounted in the cumulative pressure recording. Gas pressure was measured every 60 s and cumulative pressure was recorded at 20 min intervals. The TGP was expressed as mL of gas produced per gram of substrate OM (mL•g −1 OM) by application of the natural gas law to the accumulation of the recorded gas pressure while accounting for individual bottle volume.
In vitro CH 4 production was quantified in time series at multiple time points of 12, 24, 48 and 72 h during the fermentation. The mL of CH 4 g −1 of substrate OM was estimated using concentrations in headspace at the time series points while assuming homogeneity of headspace gas. The headspace samples were collected into 10 mL Labco Exetainer vials (Lampeter, GBR) and quantified using gas chromatography (GC) according to parameters described by [10]. Concentrations of CH 4 in headspace gas were converted to mL•g −1 substrate OM based on TGP at the relative time series points and assuming headspace homogeneity at system venting [19].
Apparent in Vitro Digestible Organic Matter and Volatile Fatty Acids
The IVD-OM and VFA production was determined as described by [10] and quantified to coincide with CH 4 determinations at each time series point. Each fermentation was chilled to cease bacterial activity and the in vitro fluid (IVF) was then vacuum filtered through a Duran No. 1 porosity glass fritted crucible containing a 0.5 cm layer of sand filtration aid. The pH was measured on the filtrate. Crucibles and wet residue were oven-dried to constant weight at 105˚C for DM determination. Residue OM was determined as loss on ignition in a muffle furnace at 550˚C for 8 h [14].
The TVFA accumulated in the IVF were quantified after termination of fermentation. Preparation of IVF prior to GC analysis consisted of addition 4 mL of IVF to 1 mL of 20% metaphosphoric acid containing 11 mM of 4-methylvaleric acid (Sigma-Aldrich; Castle Hill NSW, AUS) providing 2.2 mM internal standard. The samples were mixed and stored at −20˚C until a 1.5 mL subsample was centrifuged for 15 min at 13,500 rpm and 4˚C (Labnet Prism R; Edison NJ, USA). The supernatant was filtered through 0.2 µm PTFE syringe tip filters (Agilent; Santa Clara, CA, USA) and quantified by GC according to parameters described by [10].
Statistical Analysis
Two-factor repeated-measures permutational analysis of variance (PERMANOVA) was used to test for significant differences in TGP, CH 4 production, and IVD-OM over time. A one-factor PERMANOVA was used to test for significant differences in the production of VFA between treatments (fixed factor) using Primer 6 (version 6.1.13; [20] statistical software and PERMANOVA+ (version 1.0.3)[21]. For PERMANOVA, Bray-Curtis similarity matrices were produced using the untransformed raw data and dummy variables (0.0001) were used to account for zero values. The P-values were calculated from 999 (TGP) and 9,999 (CH 4 , IVD-OM, and VFAs) random permutations.
Pair-wise a posteriori comparison was used to determine significant groupings, where applicable. For PERMANOVA, differences were considered significant if P < 0.05. The TGP data were also fitted with generalized additive models (GAM) using cubic regression spline smoothers to predict the relationship and examine differences between TGP over time. The generalized additive models were produced using the mgcv package within the R language (version 3.0.1)[22]. Goodness-of-fit of the individual smoothers was quantified using the hydroGOF package within the R language and was assessed from the proportion of variance in the data that was accounted for by the model (r 2 ) [23].
Experiment 1: Ranking
Based on predictions of the GAM, which had a high goodness-of-fit (r 2 > 0.97), there was a significant difference (P < 0.001) in TGP in the fermentations over 72 h (Figure 1). In addition, the TGP reached its plateau approaching 48 h in all fermentations and the Asparagopsis induced a lower TGP rate compared with all other treatments (P < 0.001). There was no significant difference in cumulative TGP between the seven macroalgae with inclusion at 5% (OM basis) and the Rhodes grass control after 72 h of fermentation. The TGP for all macroalgae other than Asparagopsis ranged between 171 and 176 mL•g −1 OM, representing an insignificant 2.4% -4.8% reduction in TGP com- pared to the Rhodes grass control (180 mL•g −1 OM). However, the Asparagopsis at 2% had significantly less TGP (127 mL•g −1 OM) representing a 26% -29% reduction compared to all other macroalgae and controls.
The CH 4 production as affected by inclusion of the seven macroalgae other than Asparagopsis was not altered significantly compared to the Rhodes grass control during the course of the 72 h incubations (Figure 1). The rate of CH 4 production as defined by time series headspace sampling was variable in the same manner as for TGP. The only significant reduction in CH 4 was induced by Asparagospis in the same way as previously demonstrated [10] and no detectable CH 4 was produced. The CH 4 production from the other fermentations ranged from a low of 14 mL•g −1 OM to a high of 16 mL•g −1 OM for U. ohnoi and the Rhodes grass control, respectively.
A typical pattern of IVD-OM increasing with time was demonstrated with or without macroalgae (Figure 1), however, the rate of digestion of the substrate varied between treatments in the first half of the fermentation period (P = 0.004). Asparagospis induced the earliest onset of substrate degradation in the same fashion as demonstrated by [10] for Asparagopsis at dose concentrations ≤ 5% of substrate OM. The C. patentiramea was the slowest to achieve maximum rate of IVD-OM. Approaching 48 h, IVD-OM in all the fermentations coalesced, and after 72 h had similar IVD-OM ranging from a low of 70% to a high of 72% for C. patentiramea and the Asparagopsis, respectively. In the absence of macroalgae the IVD-OM for the Rhodes grass control was 72%.
In Experiment 1 the series of in vitro fermentations of the seven macroalgae or Asparagopsis P-control did not induce significant differences between any of the treat- ments or Rhodes grass control for TVFA (P = 0.093) or propionate (P = 0.098). However, fermentations including the Asparagopsis were significantly lower for acetate (P = 0.002) and higher for butyrate (P = 0.004) compared to the other fermentations (Table 2). A significant increase in propionate and butyrate with 2% Asparagopsis inclusion has been previously reported [10], however this feature was slightly dampened in the present study for propionate but confirmed the previously observed butyrate concentrations. Compared to the Rhodes grass control, after 72 h of fermentation the macroalgae induced marginal reductions in TVFA on a molar concentration basis from a low of 9% to a high of 23% for D. bartayresii and C. patentiramea, respectively, and Asparagopsis reduced TVFA by 22%. The production of acetate was marginally reduced by a low of 10% to a high of 24% by inclusion of D. bartayresii and C. patentiramea, respectively, and a significant 44% with Asparagopsis, respectively. Conversely, propionate was marginally decreased by a low of 5% to a high of 24% with inclusion of D. bartayresii and C. patentiramea, respectively, and increased by 9% with Asparagopsis. The production of butyrate was marginally decreased by a low of 11% to a high of 37% for D. bartayresii and C. patentiramea, respectively, and significantly increased by 76% with Asparagopsis.
There were not adequate differences in TGP, CH 4 , or VFA production to justify well defined ranking for the seven macroalgae evaluated. Clearly, none of the seven macro-Table 2. The effect of inclusion of seven macroalgae species on accumulation of volatile fatty acids after 72 h of in vitro fermentation with rumen fluid. The Rhodes grass control substrate was equal in all fermentations, the Asparagopsis P-control was included at 2% of substrate OM, and the other seven macroalgae included at 5%.
Experiment 2: Combinations
Similar to the results of Experiment 1, based on predictions of the GAM, which had a high goodness-of-fit (r 2 > 0.98), there was a significant difference (P < 0.001) in TGP in the fermentations over 72 h (Figure 2). In addition, the TGP reached its plateau approaching 48 h in all fermentations, however, in contrast with Experiment 1, the combination treatments were equivalent to the Asparagopsis P-control at 2% of substrate OM in the reduction of TGP, but all were significantly different from the Rhodes grass control (P < 0.001) after 72 h of fermentation. The TGP for the combinations ranged between 124 and 136 mL•g −1 OM representing a significant 28% -34% reduction in TGP compared to the Rhodes grass control (188 mL•g −1 OM). The Asparagopsis induced a TGP of 134 mL•g −1 OM and therefore a 28% reduction, similar to the combinations demonstrating an overwhelming effect of Asparagopsis when combined with other macroalgae.
The CH 4 production as affected by inclusion of the seven macroalgae combined with Asparagopsis was not different compared to the Asparagopsis alone during the course of the 72 h incubations (Figure 2). In the absence of macroalgae the fermentation initiated production of CH 4 immediately after inoculation with RFB, however, the inclusion of the macroalgae combinations completely inhibited CH 4 until 48 h when trace amounts were measured for the S. flavicans combination (0.2 mL•g −1 OM) representing 98% inhibition. After 48 h of fermentation S. flavicans and D. bartayresii combined with Asparagopsis accumulated small amounts of CH 4 such that at 72 h of fermentation these two combinations had reduced CH 4 by 95% (0.8 mL•g −1 OM) compared to the Rhodes grass control (16.0 mL•g −1 OM). The Asparagopsis alone reduced CH 4 by 99% (0.2 mL•g −1 OM) after 48 h and by 96% (0.6 mL•g −1 OM) at 72 h.
In the same way as Experiment 1 the typical pattern of IVD-OM increasing with time was demonstrated with or without macroalgae combinations (Figure 2). However, in Experiment 2 IVD-OM induced by the macroalgae combined with Asparagopsis was not as variable throughout the 72 h in vitro fermentations. Other than the Rhodes grass control, all fermentations contained 2% Asparagopsis which negated effects of the other macroalgae in the combinations. All the fermentations coalesced from onset of fermentation and after 72 h of fermentation the IVD-OM ranged from a low of 68% to a high of 72% for the D. bartayresii combined with Asparagopsis and Rhodes grass control, respectively.
In Experiment 2, in vitro fermentations with inclusion of the seven macroalgae combinations or Asparagopsis induced significant reductions in TVFA (P = 0.002) and acetate (P < 0.001), and significant increases in propionate (P < 0.001) and butyrate (P < 0.001) compared with the Rhodes grass control (Table 3). However, there was no significant difference between the Asparagopsis alone and the combinations, therefore improvements in VFA profiles due to combining Asparagopsis with any of the seven candidate macroalgae was not demonstrated. Compared with the Rhodes grass control, the macroalgae combinations induced reduction in TVFA on a molar concentration Table 3. The effect of inclusion of seven macroalgae species combined with Asparagopsis on accumulation of volatile fatty acids after 72 h of in vitro fermentation with rumen fluid. The Rhodes grass control substrate was equal in all fermentations, the Asparagopsis P-control was included at 2% of substrate OM, and the seven macroalgae combinations resulted in inclusions of Asparagopsis at 2% and each of the seven macroalgae at 5%. basis from a low of 7% to a high of 12% and 14% by inclusion of the P. australis and C. taxifolia combined with Asparagopsis and Asparagopsis alone, respectively. The production of acetate was reduced by a low of 24% to a high of 29% and 31% by inclusion of S. flavicans and C. taxifolia combined with Asparagopsis, and Asparagopsis alone, respectively. Conversely, propionate was increased by a low of 21% to a high of 26% and 21% by inclusion of the P. australis and C. taxifolia combined with Asparagopsis, and Asparagopsis alone, respectively. The production of butyrate was increased by a low of 75% to a high of 91% and 72% for the U. ohnoi and C. taxifolia combined with Asparagopsis, and Asparagopsis alone, respectively.
Discussion
This report represents the only current study of these seven macroalgae species (Table 1) at low inclusion in vitro. The purpose of this study was to rank them for beneficial effects in vitro prior to selection of one candidate for evaluation in vivo. However, there was no clear ranking order based on CH 4 mitigation and improvements in in vitro rumen fermentation. In combining the seven macroalgae with Asparagopsis there was no significant difference between the combinations and in all cases the Asparagopsis overwhelmed the effects on fermentation which eclipsed potential benefits of combining these macroalgae.
These macroalgae have been evaluated in rumen fermentations in vitro at high dose approaching 20% of substrate OM and the effects on TGP was variable and mostly significant reductions compared to controls were reported [13] [24]. In those studies the reductions in TGP were concomitant with reductions in CH 4 and in some cases detrimental effect on IVD-OM. Fermentations with the individual (pure) macroalgae as inclusions with Rhodes grass in Experiment 1 of the present study also demonstrated variable TGP. However, lack of significant reduction can be attributed to the much lower inclusion concentration of 5% in the fermentations. It was hypothesized that the effect demonstrated at high dose would remain at 5%, however only Asparagopsis at 2% maintained its potency at low dose which was a direct result of methanogenesis inhibition. The halogenated bioactive compounds inherent in Asparagopsis spp.have potent antimethanogenic properties [25] that are not inherent in the other macroalgae evaluated in the present study. Their mode of action at high dose may be attributed to organic acids, tannins, phlorotannins, polyphenoloics, aminoglycans and other compounds that have antibacterial or inhibitory effect on rumen microbial metabolism, but have minimal effect at a low dose.
The reduction in CH 4 induced by these macroalgae has been reproduced at high inclusion concentrations in vitro [13] [24], unfortunately in the same way as TGP, at 5% inclusion, the effect was minimal. However, the Asparagopsis at 2% was consistent with previous studies specific to that macroalga demonstrating nearly complete elimination of CH 4 production in vitro [10] [11]. Without significant differences in TGP and CH 4 the ranking order was not clear and only numerical differences could be applied in the ranking. Thus, currently the Asparagopsis spp.are the only macroalgae demonstrating CH 4 abatement ability exceeding 70% reduction at low dose and typically abatement > 99% is demonstrated.
Early in the fermentations the IVD-OM was variable with some of the macroalgae inducing a lag in the onset of fermentation. Previous research has shown a lag in TGP with addition at high dose [24] and this feature appears to be present at low dose, however both TGP and IVD-OM coalesced approaching 48 h of fermentation, and except for TGP induced by Asparagopsis, all fermentations were equal after 72 h. The Asparagopsis did not have different effect compared to the other macroalgae on IVD-OM demonstrating the importance of dose concentrations of macroalgae on rumen fermentation in vitro and presumably in ruminant animals. This indicates the importance of low dietary concentrations to maintain rumen efficiency and the seven macroalgae used in Experiment 1 demonstrated little benefit on IVD-OM or CH 4 emissions at low dose. An important feature to note is that none of the macroalgae had a negative effect on IVD-OM at the dose concentration studied.
The effect of inclusion of antimethanogenic compounds on production of VFA during in vitro fermentations has been variable in most studies, however, a trend concomitant with significant CH 4 reduction is in favour of increased propionate [26]. This has also been reported in previous studies using macroalgae which sometimes demonstrate a decrease in acetate in favour increased propionate and is more prominent with increasing dose. Macroalgae species that have a moderate or weak antimethanogenic capacity may not induce changes in the VFA profile [11] [27]. In the present study it was apparent that the seven macroalgae are weak antimethanogenic agents in vitro and as such they have minimal effects on VFA production. The TVFA and acetate results at 5% inclusion indicate a variable and small but not significant reduction with no apparent change in propionate or butyrate. These same macroalgae induced an increase in propionate with decreasing acetate at high dose, however, at high dose some species decreased TVFA as a result of detriment to IVD-OM [13]. It is important to maintain or improve IVD-OM to maximize fermentation efficiency which reiterates the requirement for appropriate dose concentrations of any dietary inclusion. Notably, Experiment 1 demonstrated that six of the seven macroalgae (excluding C. patentiramea) induced production of marginally more TVFA and significantly more acetate than Asparagopsis and so demonstrated their potential to enhance fermentation efficiency when included in combination with Asparagopsis.
It was hypothesized that combining macroalgae with Asparagopsis in vitro would improve fermentation because these were identified in Experiment 1 as weakly antimethanogenic and resulted in some increase in VFA compared with Asparagopsis alone. However, at 5% in combination with Asparagopsis at 2% there was no evidence demonstrating improvements at a level to provide incentive for a follow-up in vivo study.
There was some variable but small decreases in TGP for some of the combinations compared with Asparagopsis alone, however all were significantly reduced compared to fermentations without macroalgae. The universal decrease in TGP was attributed to the Asparagopsis proportion of the combinations and a direct result of the near elimination of CH 4 production. However, a minimal concentration of CH 4 was detected in the final measurement (72 h) suggesting a gradual loss in antimethanogenic ability over time at the dose concentration used in this study. This phenomenon was described with Asparagopsis at doses ≤2% of substrate OM [10]. In the present study this occurred with S. flavicans and D. bartayresii combined with Asparagopsis, however, this does not indicate an effect specific to these combinations as a minimal level of CH 4 was detected after 72 h with the Asparagopsis alone. Total depletion of CH 4 is not expected in vivo where rumen fermentation is much more robust than in batch cultures. Also, feed residence time in the rumen is typically less than 72 h [28].
The loss of energy as CH 4 has potential to be reclaimed as productivity, however this can't be demonstrated in vitro and in vivo studies are necessary to determine the extent of productivity gains. When demonstrated in vivo productivity gains would dramatically increase the value of Asparagopsis and macroalgae combinations for livestock production systems. Adoption of any CH 4 mitigation strategy requires more than environmental benefits. The value of carbon abatement may eventually provide revenue incentive for producers to adopt macroalgae feed additives based on CH 4 abatement.
However, improvements in productivity enhance the environmental value of macroalgae.
The IVD-OM was not negatively affected, however, the hypothesis of improvements to fermentation was not demonstrated by the combinations compared to Asparagopsis.
All the fermentations, with and without macroalgae were stable and IVD-OM was not different. Improvements in IVD-OM would provide a conduit for improved utilization of feed and offset the cost of supplementation with macroalgae. Further investigation into macroalgae on improved feed energy utilization, productivity, and feed quality is necessary, particularly relative to periods of poor grass quality for grazing livestock [29].
In light of the VFA results of previous work [13] and Experiment 1 it was hypothesized that when combined with Asparagopsis the other macroalgae would have improved VFA production compared to the Asparagopis alone. This effect unfortunately was not demonstrated and the combinations adopted similar profiles as Asparagopsis alone and the variability between treatments observed in Experiment 1 was muted in Experiment 2, thus again demonstrating the dominant effect of Asparagopsis in vitro. It is typical with Asparagopsis that decrease in CH 4 is concomitant with decrease in acetate and increase in propionate. It is common for antimethanogenic feed additives to have this effect in vitro [10] and in vivo [30] [31] and is believed to be due to reductive propionate production being more favourable than acetogenesis in the presence of excess hydrogen [26].
Although the present study did not support the use of macroalgae combinations to decrease CH 4 production in vitro the utility of combinations of macroalgae to enhance ruminant animal productivity and reduce CH 4 emissions is worthy of further exploration. Supplementation of high protein macroalgae such as the freshwater green Oe-dogonium sp. is feasible up to 25% of intake [11] which could increase the proportion of rumen bypass protein thus benefiting productivity [32]. Alternative sources of protein can also reduce CH 4 emissions intensity by improving productivity of grass fed beef during those periods of decreasing diet quality. Supplementation with macroalgae can therefore directly reduce methanogenesis and reduce emission intensity by improving the product to emissions ratio.
Conclusion
There was not enough difference induced in rumen fermentation efficiency or CH 4 production in vitro to support a conclusive ranking order between the seven individual macroalgae at the 5% inclusion concentration of this study. A numerical difference indicates U. ohnoi and C. patentiramea were the most and least antimethanogenic, respectively. When macroalgae were combined with Asparagopsis, a known potent antimethanogenic agent in vitro, there was not an adequate effect to justify proceeding to in vivo evaluation or recommendation for use of the combinations in livestock feed.
However, high protein macroalgae supplemented at higher dietary concentrations may provide greater benefit when combined with Asparagopsis by contributing to reduced CH 4 emissions through further improved productivity at times of low feed quality thus reducing emissions intensity per product output.
Figure 1 .
Figure1. The effect of inclusion of seven different macroalgae on gas production and substrate digestibility over 72 h of in vitro fermentation with rumen fluid. From top down: Total gas production (TGP); CH 4 production; and apparent in vitro digestibility (IVD-OM). The Rhodes grass control substrate was equal in all fermentations, the Asparagopsis control was included at a concentration of 2% of substrate OM, and the other seven macroalgae included at 5%. No ±SE is presented for TGP because SE was smaller than the symbols.
Figure 2 .
Figure 2. The effect of inclusion of seven different macroalgae combined with Asparagopsis on gas production and substrate digestibility over 72 h of in vitro fermentation with rumen fluid. From top down: Total gas production (TGP); CH 4 production; and apparent in vitro digestibility (IVD-OM). The Rhodes grass control substrate was equal in all fermentations, the Asparagopsis control was included at a concentration of 2% of substrate OM, and the seven macroalgae combinations resulted in an inclusion of Asparagopsis at 2% and each of the seven macroalgae 5%. No ±SE is presented for TGP because SE was smaller than the symbols.
Table 1 .
. The biomass was sourced from either large scale culture at the Centre for Macroalgal Resources and Biotechnology Compositional parameters of the macroalgae and Rhodes grass hay.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-04-21T01:56:45.709Z", "created": "2016-09-29T00:00:00.000", "id": "5250294", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scirp.org/journal/PaperDownload.aspx?paperID=71304", "pdf_hash": "916c3d80e80bf545cc3adf4cad78bc4f23ff84e0", "pdf_src": "ScienceParseMerged", "provenance": "peS2o-0075.json.gz:38280", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "916c3d80e80bf545cc3adf4cad78bc4f23ff84e0", "year": 2016}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effects of salicylic acid elicitor against aphids on wheat and detection of infestation using infrared thermal imaging technique in Ismailia , Egypt
Wheat (Triticum sativum L.) is one of the most important cereal crops in Egypt. Insect pests, such as aphids, are major threats in terms of yield reduction. Induced resistance in wheat using salicylic acid as a foliar application was tested on the farm of the Faculty of Agriculture, Suez Canal University during 2012/2013 and 2013/2014 seasons. Three wheat cultivars, Gemeza 9, Sakha 93 and Giza 168, were sprayed three times with two concentrations of salicylic acid (SA), 200 mg/l and 100 mg/l, after early detection of aphid infestation by infrared thermal imaging. The infrared thermal imaging technique is based on significant differences in surface temperature between infested and healthy leaves. Imaging data are digital, and a computer program can be used to detect infestation rapidly. The results showed that aphid infestation raised the temperature of infested leaves, compared to healthy leaves. The range temperature difference between maximum and minimum temperatures (At) was 1.1 oC in healthy leaves and 3.9 oC in infected leaves. The results of SA application showed significant differences in the mean number of aphids and in reduction of infestation among treatments and cultivars. The higher of the two SA rates (200 mg/l) gave higher efficacy in the three cultivars than the lower rate (100 mg/l) over the five weeks of trial. The highest efficacy against aphids was reached one week after application (86.28% for Giza, 85.89% for Gemesa and 70.54% for Sakha). Moreover, SA treatment enhanced the wheat yield of all three cultivars, compared with control plants. The three cultivars (Giza, Gemesa and Sakha) produced higher yields than the control when sprayed with 200 mg/l SA. Their grain yield was 2,491.5, 2,455.0, and 2,327.25 kg/feddan (1 fed = 0.42 ha), respectively. In conclusion, infrared thermal imaging can be employed in identification of infected leaves. Also, the application of SA on wheat induced plant resistance to aphids.
INTROdUCTION
Wheat (Triticum sativum L.) is one of the oldest and most important cereal crops in Egypt. Although wheat production per unit area has significantly increased in Egypt over the past years, wheat production supplies only 40% of its annual domestic demand. Wheat occupies about 32.6 percent of the total winter land area and is mostly used to make bread, a very important component of the Egyptian diet (Elhamid, 2014).
Several insect pests infest wheat and cause enormous damage during two important growth stages (heading and flowering) (Freier et al., 2007). Major threats regarding yield reduction of wheat come from insect pests. Wheat aphids are devastating insect pests of wheat (Steffey & Grey, 2012). Management of wheat aphids is difficult due to their rapid reproduction and extremely short life cycle. Currently, the use of conventional insecticides is the main approach in controlling aphids (Dogimont et al., 2010). These insecticides have negative impact on the environment, especially on beneficial organisms,. Moreover, plant treatment with insecticides leads to resistance of aphids and other insect pests, which also contributes to management problems (Dogimont et al., 2010).
Induced plant resistance to insect pests has been documented in several crops (Hussein et al., 2014;Mahmoud, 2013). Several investigators have proposed the use of elicitors of plant resistance as a means of controlling insect pests in agriculture (Thaler et al., 1999;Boughton et al., 2006). This new control approach is gaining in topicality because insect pests and diseases are serious constraints in efforts to increase productivity per feddan. Salicylic acid (SA) is a plant phenolic widely distributed throughout the plant kingdom. It is a hormone-like substance which plays an important role in the regulation of many aspects of plant growth and development (Raskin, 1992). However, it is especially famous for its ability to induce systemic acquired resistance in plants (Ryals et al., 1996).
Near infrared reflectance has been proposed for detecting grain insects in wheat (Dowell et al., 1998) and has been applied to tree fruit pests (Hansen et al., 2008), but this technology has not been applied to wheat leaves infested by aphids. Thermal imaging is a technique which converts the radiation emitted by an object into temperature data without establishing contact with the object. It has been successfully used in civil engineering, manufacturing industries, electrical engineering and medicine, but has been rarely applied in agriculture, e.g. for detection of bruises in fruits and vegetables (Varith et al., 2003), detection of foreign substances in food (Meinlschmidt & Maergner, 2003), and detection of insect infestation of stored grain (Manickavasagan et al., 2008).
The objective of the current study was to investigate the efficacy of the elicitor SA in controlling wheat aphids under field conditions, and its impact on wheat yield. Also, the infrared thermal imaging technique was assessed as a new approach in entomological research for early detection of wheat infestation by aphids.
Experimental design
The experiment was carried out at the Experimental farm, Faculty of Agriculture, Suez Canal University, Ismailia, Egypt during the growing seasons of wheat (2012/2013 and 2013/2014). Three wheat cultivars (Gemesa 9, Sakha 93 and Giza 168) were selected for this experiment. The cultivars were sawn on December 9, 2012 and 2013. Two rates of salicylic acid and an untreated control were used. The experimental plots were designed as a randomized complete block. Each treatment was replicated 4 times in each block.
Thermal imaging camera
An infrared thermal camera (Fluke Ti32), which is sensitive to 7.5-13 µm wavelengths at 320 X 240 pixels, and a visual mode of camera were used. Thermal calibration ranged from -10.0°C to 600.0°C with 9Hz image speed. A software converts temperatures to images based on a palette. Five temperature features (average, maximum and minimum temperatures of wheat leaves, temperature range difference between maximum and minimum temperatures (At), and standard deviation) were extracted from each thermal image using a Matlab algorithm.
Wheat thermal imaging
Infrared thermal imaging was performed several times on wheat plants before the experiment to adjust surface temperature. Temperature measurements were taken on 5 leaves of each wheat cultivar at different heights. For comparison, infested wheat leaves were also measured.
Treatments with salicylic acid
Plots with the three wheat cultivars were sprayed with salicylic acid (SA) three times after early detection of aphid infestation by the infrared thermal imaging camera. SA was applied at two rates (200 mg/l and 100 mg/l). SA was dissolved in a few drops of ethanol and then dispersed in water to give the required rates. After early detection of aphid infestation, wheat cultivars were sprayed with the two rates of SA at weekly intervals three times using a hydraulic sprayer. Control plots were sprayed only with distilled water and each treatment was replicated 4 times.
Wheat aphid counts
For population counts of wheat aphids, twenty-five plants were randomly selected from each replicate. Aphid counting was done one day before spraying. After SA spraying, aphid population counts were recorded at intervals of one, two, three, four and five weeks. Moreover, in view of these pre-treatment differences, corrected percentage efficacies were calculated according to the following modification of Abbott's formula, as described by Henderson and Tilton (1955): where T b is infestation in treated plot prior to application; T a is infestation in treated plot after application; C b is infestation in control plot prior to application; C a is infestation in control plot after application.
Wheat yield
The yields of SA-treated wheat plants were compared with the control. At the end of the experiment random samples were taken to measure the following characters: kernel number, weight of 1000 kernals and grain yield.
Statistical analysis
Data obtained in the present study was subjected to an analysis of variance (ANOVA) with the honestly significant difference value calculated as Tukey's statistic at P ≤ 0.05.(SAS Institute 2004).
RESULTS
Reflectance values of infrared thermal images indicated that the leaves infected by aphids reflected higher in the visible region of the electromagnetic spectrum, rather than in the infrared region, compared to healthy leaves (Figure1). The average surface temperature of 20 healthy wheat leaves at 70 pixels was 28.6°C ± 0.23 whereas the average surface temperature of wheat leaves infested by aphids was 30.1°C ± 1.04. The maximum surface temperature of infected and healthy leaves was 32.2ºC ± 1.07 and 29.3°C ± 0.35, respectively. In addition, the minimum temperature of infected and healthy leaves was 28.8°C ± 0.91 and 28.2°C ± 0.2, respectively. The background temperature (30.1°C) and emissivity (0.97) were the same for both the infected and healthy leaves of wheat. The (At) range was higher in infected than in healthy leaves. It was 3.9 and 1.1, respectively (Table 1). Thermal images of the healthy and infected leaves are shown in Figure 2.
RESULTS
The average maximum and minimum surface temperatures of healthy leaves were 29.3°C ± 0.35, and 28.2°C ± 0.21, respectively. The average maximum and minimum surface temperature of infested leaves were 32.7°C ± 1.07, and 28.8°C ± 0.91, respectively. Our classification of healthy and infested leaves using thermal imaging was based on that variation in leaf temperatures. The results in Table 2 show the mean number of aphids on three wheat cultivars. Their population decreased after SA application and then increased again in the following counts, particularly when sprayed with the high rate of 200 mg/l. Aphid counts were consistently lower on the plots sprayed with SA than on the control plots. There was a significant difference in the mean number of aphids on SA-treated and control plots. After a week, it was 8 for Gemesa (F= 227.239; p≤ 0.0000), 10 for Giza (187.948;p≤ 0.0000) and 24 for Sakha (F= 104.564; p≤ 0.0000). After two weeks, it was 23 for Gemesa (F= 110.873; p≤ 0.0000), 27 Giza (F= 121.510; p≤ 0.0000) and 36 for Sakha (F= 42.879; p≤ 0.0000). After three weeks, it was 41 for Gemesa (F= 60.931; p≤ 0.0000), 58 for Giza (F= 54.071; p≤ 0.0000) and 78 for Sakha (F= 11.815; p≤ 0.0030). After four weeks, it was 83 for Gemesa (F= 94.659; p≤ 0.0000), 79 for Giza (F= 24.930; p≤ 0.0002) and 100 for Sakha (F= 2.242; p≤ 0.1621). After five weeks, it was 141 for Gemesa (F= 14.333; p≤ 0.0016), 186 for Giza (F= 25.723; p≤ 0.0002) and 211 for Sakha (F= 0.607; p≤ 0.5657). Data in Table 3 show that the higher rate of SA (200 mg/l) gave higher efficacy in the three cultivars than the lower rate (100 mg/l) over the five-week period. The highest efficacy against aphids was found one week after application (86.28% for Giza 168, 85.89% for Gemesa 9 and 70.54% for Sakha 93, respectively). The efficacy of the two SA rates decreased gradually and reached 33.61% and 5.09% for Giza 168, 14.39% and 0.8% for Gemesa 9, and 12.89% and 0.0% for Sakha 93, respectively. SA achieved a satisfactory reduction in aphid population within the five-week period, especially on the cultivars Gemesa 9 and Giza 168. A comparison of yields of the three wheat cultivars (Gemesa 9, Sakha 93 and Giza 168) revealed significant differences in kernal numbers/ear, weight of 1000 kernals (g), and grain yield (kg/feddan) between the untreated control and the SA treatments. There were significant differences among the cultivars regarding kernal numbers/ear (F=19.0027;p≤0.0009 for Gemesa 9, F=19.529;p≤0.0000 for Sakha 93 and F=19.529; p≤0.0000 for Giza 168), weight of 1000 kernals (g) (F=7.142;p≤0.0032 for Gemesa 9, F=32.255;p≤0.0000 for Sakha 93 and F=32.255; p≤0.0000 for Giza 168), and grain yield (kg/fed.) (F=13.143;p≤0.0021 for Gemesa 9, F=16.002;p≤0.0011 for Sakha 93 and F=22.228; p≤0.0003 for Giza 168).
The three cultivars (Giza 168, Gemesa 9 and Sakha 93) sprayed with 200 mg/l SA had higher yields than the control, i.e. 2,491.5, 2,455.0,and 2,327.25 kg/fed, respectively. Control yield was 2,189.0,2,157.0 and 2,227.5. Also, yield data revealed that the lower rate of salicylic treatments enhanced yield relative to control. A significant difference was detected between treated and control cultivars (Table 4).
dISCUSSION
Early detection of wheat infested by aphids using infrared thermal imaging can help us prevent increase in aphid numbers and their management at the beginning of infestation. It can improve the yield and quality of crop, and reduce the use of pesticides.
Aphid numbers were lower in the treated than in untreated plots, which may be attributed to salicylic acid application that deterred aphids from wheat and increased foraging by parasitoids and predators attacking herbivorous insects of wheat crop. Plants treated with SA produce a volatile of methyl salicylate, which normally repels polyphagous herbivores and may attract specialist herbivores and their natural enemies which use a volatile as a host location cue. This study was similar to that reported by Pickett and Poppy (2001), who indicated that methyl salicylate prevented aphids from colonizing plants and populations of natural enemies of herbivorous pests. Salicylic acid acted as a signal in some induced responses to pathogens, as well as insect pests (Pickett & Poppy, 2001). Over 40 insect species from five separate orders have been identified as having olfactory receptors for the methylated form, methyl salicylate (Chamberlain et al., 2000;Pickett et al., 2003). The cereal aphids R. padi, S. avenae and Metopolophium dirhodum have a specific olfactory neuron on the sixth antennal segment to detect methyl salicylate (Pickett et al., 2006).
Salicylic acid caused a reduction in aphid population within five weeks after application. Pettersson et al. (1994) reported that cereal crops treated with a slowrelease formulation of methyl salicylate had been avoided by many insects. Thus, in spring field trials, methyl salicylate applied to wheat significantly reduced (by 30-40%) the overall number of aphids colonizing the crop.
The experiments showed that salicylic acid affected the yield of the three wheat cultivars. The significant increase in wheat yield may have been due to the SA treatments that reduced wheat aphid damage. This result is consistent with Ibrahim et al. (2014), who reported that SA applied to the wheat cultivar Sakha 93 had increased the yield of crop and its components.
CONCLUSION
The infrared thermal imaging technique can be used to identify aphid-infected leaves of wheat. Also, the results indicate that SA application after such early detection could be helpful in aphid management strategies to increase wheat yield and reduce the use of chemical insecticides.
Figure 1 .
Figure 1. Thermal image of infected and healthy leaves
Figure 2 .
Figure 2. Variation of temperature between healthy and infected leaves at 70 pixels.
Table 1 .
Mean temperature values (± SD) of healthy and infested wheat leaves
Table 2 .
Mean number of wheat aphids on three wheat cultivars
Table 3 .
Efficacy of salicylic acid in reducing aphid infestation on three wheat cultivars
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-29T21:03:14.937Z", "created": "2015-01-01T00:00:00.000", "id": "59452704", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBYSA", "oa_status": "GOLD", "oa_url": "http://www.doiserbia.nb.rs/ft.aspx?id=1820-39491502091M", "pdf_hash": "df1b12e127f5e475d2485c50122b9e0e46953444", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:32281", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "df1b12e127f5e475d2485c50122b9e0e46953444", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Animal performance and sward characteristics of Mombaça guineagrass pastures subjected to two grazing frequencies Desempeño animal y características de pasturas del pasto guinea cv. Mombaça sometidas a dos frecuencias de pastoreo
The aim of this work was to compare grazing management practices of Mombaça guineagrass (Megathyrsus maximus syn. Panicum maximum cv. Mombaça) based on the sward incident light interception (LI) concept. We tested, when the regrowth period in rotationally stocked Mombaça guineagrass ended, if LI (90 or 95%) affected forage accumulation, sward characteristics and animal performance. Both treatments had a common post-grazing canopy height of 50 cm and were replicated 4 times in a randomized complete block design. Pastures were sampled preand post-grazing to determine forage mass, morphological composition and forage accumulation rate (FAR). Nutritive value (NV) was estimated in pre-grazing samples. Stocking rate was adjusted twice a week, and animals were weighed every 28 days. Pre-grazing conditions of 90 and 95% LI were reached at pasture heights of approximately 80 and 90 cm, respectively. FAR, sward structure and NV were similar for pastures grazed at 90 and 95% LI. Consequently, stocking rate, average daily gain and liveweight gain/ha were similar for both LI treatments. Data suggest that Mombaça guineagrass can be grazed at pre-grazing heights of 80–90 cm (90–95% LI) without compromising pasture structure and animal performance provided moderate defoliation severity is employed. Further testing of this grazing strategy over longer periods should be carried out with this species as well as other tropical grasses.
Introduction
Some studies with tropical grasses under intermittent stocking have shown that the point at which the canopy intercepts 95% of photosynthetically active radiation (PAR) approximates an ideal time period to interrupt regrowth. After this point, forage accumulation and nutritive value decrease (Carnevalli et al. 2006; Barbosa et al. 2007;Zanini et al. 2012) as proportions of stem and dead material in pregrazing forage mass increase (Silva et al. 2009).
However, insistence that an interruption of the rest period must occur precisely when the canopy intercepts 95% of the PAR can be restrictive and impractical for producers. Flexibility of management greatly facilitates the planning of livestock systems because it is common for more than one paddock to reach the ideal grazing condition at the same time during periods of vigorous forage growth (Zanine et al. 2011). On the other hand, when weather conditions are unfavorable for plant growth, the time required to achieve the target of 95% LI can be very long (Carnevalli et al. 2006; Barbosa et al. 2007;Giacomini et al. 2009), hindering the rotation of animals in paddocks available.
Using mathematical models, Parsons et al. (1988) demonstrated that, regardless of variation in management, there was a range in level of interception of PAR by the canopy in which forage production remained relatively stable. In this context, Barbosa et al. (2007) and Zanine et al. (2011) found no difference in the accumulation of leaf blades of guineagrass (Megathyrsus maximus) cv. Tanzania when the canopy LI was 90 or 95%. This suggests there could be some flexibility in the definition of pre-grazing targets, i.e. instead of a specific point there could be a range of possible values.
The end point of grazing events is also important. Maximization of short-term forage intake rate was achieved when the reduction in pasture height during grazing did not exceed 40% of the initial height (Fonseca et al. 2012;Mezzalira et al. 2014). This indicates that, regardless of the pre-grazing goals, an important condition for the maintenance of high livestock production is the use of relatively lenient defoliation levels.
Against this background, we aimed to evaluate forage accumulation and nutritive value, canopy characteristics and animal production in Mombaça guineagrass (Megathyrsus maximus syn. Panicum maximum cv. Mombaça) pastures subjected to 2 grazing frequencies, defined by 90 and 95% LI by the canopy, in conjunction with a common postgrazing canopy height of 50 cm.
Materials and Methods
The experiment was conducted during a single growing season from September 2012 to May 2013 at the National Beef Cattle Research Center in Campo Grande, MS, Brazil (20º25' S, 54º51' W; 530 masl). The climate, according to the Köppen classification, is rainy tropical savanna, corresponding to the Aw subtype, characterized by a seasonal distribution of rainfall with a well-defined dry period during the colder months. Average annual rainfall is about 1,500 mm, of which 80% falls during the 7-month wet period (October-April). The historical average minimum and maximum temperatures in the coldest month were 15.3 and 27.3 °C, respectively, and during the summer 18.2 and 31.2 °C. Weather data during the experimental period were collected from a meteorological station located 2 km from the research site ( Figure 1).
Average temperature and monthly precipitation were used to calculate the water balance ( Figure 2). The soil water storage capacity was determined to be 75 mm.
Based on these analyses, commencing in October 2012 well-established pastures (planted in 2009) were fertilized with 39 kg P, 75 kg K and 200 kg N/ha, divided equally among 4 application times, namely: October, December, January and February. Nitrogen was applied as ammonium sulfate in October and the remaining applications were as urea.
The experimental area was 12.0 ha, divided into 8 pastures measuring 1.5 ha, and these pastures were subdivided into 6 paddocks of 0.25 ha each. A 6.0 ha reserve pasture was used for holding extra animals when they were not grazing experimental pastures. The experimental design was a randomized complete block with 2 treatments and 4 replications. The grazing method used was rotational stocking with a variable stocking rate. The treatments comprised 2 grazing frequencies, characterized by pre-grazing conditions in which the canopy intercepted 90 and 95% of PAR at interruption of pasture growth, i.e. introduction of grazing animals. Stock were removed from each paddock of both treatments when grazing height had been reduced to 50 cm.
When each pasture reached the predetermined level of light interception it was grazed by 6 Senepol × Caracu (50:50) tester steers (approximately 11 months of age and with an average weight of 224 ± 16 kg initially). The testers were assigned randomly to experimental units at the beginning of the experimental period; the differences in allocation weights across treatments were not significant at the beginning of the growing season. The tester animals grazed the same pasture (1.5 ha divided into 6 paddocks) for the Deficiency Excess entire experimental period. Fifty-two regulator steers, similar to the tester steers in weight, age, background and breeding, were kept in the reserve pasture and used whenever the stocking rate needed to be increased. The animals were treated with a broad-spectrum anthelmintic at the beginning of the experiment and with pour-on ectocide during the experiment as needed for the control of ticks and horn flies. Animal health management was performed as recommended by the National Beef Cattle Research Center. All animals received water and a mineral mixture ad libitum.
Sward LI was monitored in 2 paddocks of each pasture, using a canopy analyzer apparatus (AccuPAR Linear PAR/LAI ceptometer, Model PAR-80; DECAGON Devices) at 20 random points per paddock, with one reading being taken above the canopy and one at ground level at each point. The measurements were performed weekly. When LI reached 85%, LI was monitored daily until the target was reached. Concurrently with the LI measurements, canopy height was monitored, using a 1 m ruler graduated in centimeters, at 40 random points per paddock. The readings of sward non-extended leaf height were taken from ground level to the 'leaf horizon' on the top of the sward as a reference, even during periods when plants were reproductive and produced taller flowering stems. Average heights corresponding to 90 and 95% LI were used as target heights for the other 4 paddocks from each pasture. Postgrazing heights were measured as soon as the animals left each paddock, as described above.
Forage mass, morphological composition and total forage and leaf accumulation rates were measured in a single paddock per pasture for each grazing cycle. Preand post-grazing forage mass were estimated by cutting 9 randomly selected samples (1 m 2 each) at ground level in each paddock using a manual mower. The samples were divided into 2 subsamples: 1 subsample was weighed and oven-dried at 65 °C until constant weight, and the other subsample was separated into green blades (leaf blades), green stems (stem and sheath) and dead material, and these fractions were dried at 55 °C until constant weight.
Forage accumulation rate was calculated as the difference between the current pre-grazing and the previous post-grazing forage mass, considering only the green portion (leaves and stems), divided by the number of days between samplings. For leaf accumulation rate, we used the same procedure, considering only the leaf portion in the samples. The total herbage accumulated from the entire experimental period, i.e. grazing season, was the sum of forage accumulation values across all grazing cycles.
In a second paddock of each pasture, 3 stratified samples were collected. A 1 m 2 frame was placed in areas that were representative of the average sward condition (based on visual assessment of height and herbage mass). At each location, the canopy was sampled using scissors in 4 vertical strata: >80, 60-80, 40-60 and 0-40 cm, commencing from top to basal layers. Samples from each stratum were weighed and handled as described above to estimate forage mass and its morphological components. Leaf samples were dried, ground and analyzed for crude protein (CP), neutral detergent fiber (NDF) and acid detergent lignin (ADL) concentrations, as well as in vitro organic matter digestibility (IVOMD), using nearinfrared spectroscopy (NIRS).
Steers were weighed at 28-d intervals following a 16hour fasting period to minimize gut-fill effects on liveweight measurements, i.e. fasted from both water and feed. The average daily gain was calculated as the increase in live weight of the testers divided by the number of days between weighings.
The stocking rate per cycle was calculated as the sum of the animal days (tester and regulator steers) spent in each of 6 paddocks (0.25 ha) divided by the total number of grazing days of a complete cycle, and divided by the pasture area (1.5 ha). It was expressed in animal units (AU = 450 kg live weight) per hectare. Liveweight gain/ha was calculated as the product of average daily gain and the number of steers/ha. The data were grouped by season as follows: spring (15 October-20 December), summer (21 December-20 March) and autumn (21 March-16 May). The experimental unit for both vegetation and animal data was the pasture. The data were subjected to an analysis of variance using the Mixed Procedure in SAS (Statistical Analysis Systems, version 9.4). The choice of the covariance matrix was made using the Akaike Information Criterion (AIC) (Wolfinger 1993), and analysis was performed considering sward light interception levels and season of the year and their interactions as fixed effects and blocks as a random effect (Littell et al. 2000). The season effect (spring, summer and autumn) means were compared using a Tukey test at a 5% significance level. For the stratified herbage samples, the same model was applied, but the effect of the stratum was added and considered fixed. Average daily gain data were analyzed via multivariate analysis with repeated measures according to Littell et al. (2000). Furthermore, we performed analyses of the relationships between the means of pre-grazing sward height and the means of interception of incident light by the canopy for each experimental unit for the entire experimental period
Results
There were no significant (P>0.05) interactions between LI and season for all variables associated with pasture characteristics. However, pastures grazed at 95% LI had longer rest and grazing periods, greater pre-grazing sward heights, forage mass, green stem (GSP) and dead material (DMP) percentages, plus fewer grazing cycles with lower green leaf percentages (GLP) and leaf:stem ratios (LSR) than those managed at 90% LI (Table 1). On the other hand, LI had no significant effect on forage accumulation rates (FAR; P = 0.248) and leaf accumulation rates (P = 0.085). The means and standard errors were: 86.7 ± 4.3 kg DM/ha/d and 59.6 ± 2.2 kg DM/ha/d, respectively.
With regard to seasonal effects ( Table 2), lengths of rest periods followed the order summer<spring<autumn, while the reverse order (autumn<spring<summer) was observed for forage and leaf accumulation rates. Grazing periods were longer in autumn than during summer, with those in spring being intermediate. During autumn, pastures had greater stem percentages and lesser forage and leaf accumulations per cycle, leaf percentages and leaf:stem ratios than in spring and summer. However, pregrazing forage mass (P = 0.725) and dead material percentages did not differ (P = 0.6738) between seasons.
There was no effect of LI (P>0.05) on forage dry mass in each layer and the distribution of the various morphological components in the vertical canopy profile.
However, a stratum effect was observed for those variables. Forage dry mass and percentages of green stem and dead material decreased, but green leaf percentage increased from the basal to upper strata of the canopy (Table 3). Furthermore, no interactions were observed for LI by stratum (P>0.05), season by stratum (P>0.05) or LI by season by stratum (P>0.05). Post-grazing residues were maintained close to the target height of 50 cm throughout. Means ± SD were: 47.1 ± 1.3 and 49.7 ± 1.5 cm for pastures grazed at 90 and 95% LI, respectively.
On the other hand, when variables associated with the nutritive value of green leaf were evaluated in the vertical canopy profile, percentages of CP and IVDOM increased and concentrations of NDF and ADL decreased from the basal to the top strata (Table 5). No interactions were observed between LI and stratum, LI and season and season and stratum (P>0.05) for the variables associated with nutritional value of leaves.
There was no interaction between LI and season for stocking rate (SR; P = 0.578) or for average daily gain (ADG; P = 0.671). Moreover, there was no effect of light interception on SR, ADG or liveweight gain/ha (Table 6).
With regard to seasons, ADG was least in autumn and SR and liveweight gain/ha were greater in summer than in spring and autumn (Table 7).
Discussion
Pre-grazing canopy heights for the pastures managed at light interceptions (LI) of 90 and 95% remained relatively stable during the experimental period (Table 1). There was a positive correlation (P = 0.0001; r 2 = 0.86) between LI and sward height, which highlights the potential use of canopy height as a field guide for monitoring grazing management of this cultivar. This result supports Silva and Nascimento Júnior (2007), who suggested that canopy height could be used as a reliable criterion on which to base the optimal time to interrupt pasture regrowth. Regardless of the LI target used to define the time to regraze pasture, forage accumulation resumed quickly after defoliation because a lenient grazing strategy was adopted (post-grazing target of 50 cm), which led to 42 and 44% decreases in the pre-grazing heights for pastures managed at 90 and 95% LI, respectively. According to Parsons et al. (1988), the rate of photosynthesis is reduced less by defoliation and the maximum rate of photosynthesis is restored sooner in more leniently defoliated swards. Total forage accumulations were similar for pastures managed at 90 or 95% LI. This was in agreement with the results of Barbosa et al. (2007) and Zanine et al. (2011), who found that leaf accumulation was similar in Tanzania guineagrass pastures managed at 90 and 95% LI, and those of Sbrissia et al. (2013), who observed similar forage accumulation values in kikuyu grass (Cenchrus clandestinus syn. Pennisetum clandestinum) pastures managed at 15 and 25 cm (25 cm corresponding with 95% LI).
However, pre-grazing green stem and dead material percentages were greater in pastures managed at 95% LI (Table 1), indicating that stem elongation may have started even before the pasture reached 95% LI. Santos et al. (2016) observed up to a 7-fold increase in stem elongation rate in annual ryegrass when the pastures exceeded a height of 17 cm, a condition in which there was still no restriction by high light interception. This supports the hypothesis that stem elongation can be initiated with a LI of the PAR lower than 95%.
In this context, Barbosa et al. (2012) observed that the forage mass of Tanzania guineagrass pasture grazed at 90% LI was composed of younger tillers than that in pastures grazed at 95 and 100% LI. These authors also observed that younger tillers had higher leaf appearance and leaf elongation rates, and consequently a greater leaf length and number of live leaves than mature and/or older tillers.
By contrast, fluctuations in weather conditions (Figures 1 and 2) and the dates of nitrogen application (1/3 in spring and 2/3 in summer) affected forage and leaf accumulation rates throughout the experiment (Table 2). This, in turn, influenced the variation in rest periods (Table 2; Figure 3) and stocking rates (Table 7; Figure 3) of the pastures, throughout the experiment. It is highlighted that weather conditions were similar to the historical 30-year average rainfall. Considering that post-grazing target height was the same for both treatments and forage accumulation rates were similar for these treatments, light interception levels determined the lengths of the resting periods (Table 1; Figure 3). Pastures grazed at 90% LI required less time to reach the pre-grazing target, resulting in an additional 1.4 grazing cycles for these pastures than for pastures managed at 95% LI (Table 1).
The changes in lengths of the grazing periods throughout the study (Tables 1 and 3) could be explained by the variation in forage accumulation rates (Table 3), stocking rate adjustments (Figure 3) to maintain the pregrazing treatment targets and the need for animals to remain in their current paddocks until the next paddocks to be grazed reached the pre-grazing LI target.
The greater pre-grazing forage mass values for pastures managed at 95% LI (Table 1) did not result in a higher stocking rate in these pastures (Table 6). This can be explained by the need to use fewer animals because the grazing period was longer (Table 1) as a longer resting period was required for these pastures to reach 95% LI (Figure 3).
Despite the greater green stem and dead material percentages in pastures managed at 95% LI, when considering vertical distribution in the canopy profile, we found that about 95% of the green stems and dead material were located in the 0-40 cm stratum (Table 3). This stratum is below the post-grazing target (approximately 50 cm), so theoretically the animals did not have to explore this stratum. This finding supports the results of Zanini et al. (2012), wherein approximately 90% of all stem mass is located in the lower half of the canopy, regardless of the grass species or the targeted pregrazing height.
Considering only the theoretical grazing horizon (that part of the canopy above 40 cm), green leaf and green stem percentages were 92.3 and 3.9%, respectively (Table 4), resulting in a leaf:stem ratio of 24:1. This indicates that, regardless of the pre-grazing LI target, the canopy structure above 40 cm did not limit the selection and prehension of leaves, and consequently, forage intake by the animals.
Even with the strict control of pre-and post-grazing targets, the morphological composition of the forage varied between seasons. The decrease in leaf percentage and increase in stem percentage during the autumn (Table 2) can be partly explained by the onset of flowering of the Mombaça guineagrass in mid-April. In this period, 6.5% of the forage mass was inflorescences, regardless of the pre-grazing height targets. It is known that, after the inflorescence emerges, the appearance of leaves ceases and stem elongation increases; this was confirmed by the lowest leaf percentage and the highest stem percentage in the pre-grazing forage being recorded in this period of the year (Table 2). This greater growth of stems may explain the high stem percentage in the stubble in autumn (Table 4). On the other hand, regardless of the management strategy used, dead material percentage was higher in spring than in summer and autumn (Table 4). The increased presence of dead material is common in early spring when pastures begin to recover from the dry season (Barbosa et al. 2007;Difante et al. 2009).
The similarity in nutritional value of the leaves and stems in the pastures managed using these 2 grazing strategies could be explained by their very close stage of growth, since the major changes in nutritive value occurring in pasture plants are those that accompany maturation (Van Soest 1994).
The similarity in animal performance in pastures grazed at 90 and 95% LI (Table 6) can be explained by the similarities in the canopy structures (Table 3), percentages of the stratum removed and nutritional value of the forage, indicating that the animals accessed similar pasture conditions. In this context, when analyzing the nutritional value of the leaves in the strata over 40 cm (Table 5) and considering the stem percentages above 40 cm (Table 3) and their nutritional values, the average crude protein concentration and in vitro digestibility of organic matter were 11.5 and 58.6%, respectively, for the forage theoretically available to the animals. The estimated average daily gains of the animals as a function of the amount of protein and energy (NRC 1996) revealed that the daily gain possible from the nutritive value of this grass was 810 g, a value close to those observed in the spring and summer (Table 7).
However, average daily gain in autumn was much lower (Table 7). Since there was no change in pasture nutritive value between seasons, the variation in pasture structure (Table 2) was the probable cause of the decrease in forage intake, and consequently, weight gain of the animals in autumn. According to Benvenutti et al. (2008), in pastures in the reproductive stage stems act as a physical barrier by interfering with the process of bite formation, thus affecting bite dimensions and selectivity, and consequently daily nutrient intake. Recent studies have shown that maximum short-term forage intake rates could be maintained until forage in the upper 40% of the optimal pre-grazing canopy height had been consumed (Fonseca et al. 2012;Mezzalira et al. 2014). In this study, similar (P = 0.258) defoliation severity (in percentage of the height removed) was found for both treatments. The averages and standard errors for extent of reduction in canopy height during grazing were 43.8 ± 0.3 and 42.4 ± 0.3% for the pastures managed at 95 and 90% LI, respectively. Therefore, these results suggest that relatively moderate defoliation levels are more important than pre-grazing goals per se (provided the maximum height limit does not exceed the critical PAR) when the objective is to maximize animal performance.
Similarly, because there was no change in forage accumulation or stocking rate, the similar levels of liveweight gain/ha with the two LIs indicate that Mombaça guineagrass pastures can be managed using either of these management strategies. Thus, instead of basing decisions on a specific LI, some flexibility exists in the pre-grazing target used, without impairment of the productive performance of the animals (Table 6).
Our data indicate that Mombaça guineagrass pastures can be grazed under a rotational system using pre-grazing heights of 80-90 cm (90-95% LI) without compromising the performance of either the pasture or the animals provided a moderate defoliation severity is employed, i.e. approximately 45% of the optimal pre-grazing height of pasture is consumed before animals are removed. This hypothesis should be tested further with this pasture and other erect grass species plus prostrate species.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-02-06T09:11:22.610Z", "created": "2020-01-31T00:00:00.000", "id": "213345530", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.tropicalgrasslands.info/index.php/tgft/article/download/582/377", "pdf_hash": "6a8c472a4932786874a9a2fd9f5c1a471f81d9fb", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:35475", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "c0ff45ddea70cca1b74d4c9e9adc40871df60572", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Effect of Sulphur on Yield and Biochemical Constituents in Groundnut (Arachis hypogaea L.) grown on Vertic Ustropept of Tamil Nadu
A pot culture experiment was conducted with groundnut (CO 7) in a sulphur deficient (7.19 mg kg-1) Inceptisol (Vertic Ustropept) at the Radioisotope (Tracer) Laboratory, Department of Soil Science and Agricultural Chemistry, Tamil Nadu Agricultural University, Coimbatore during October 2016 to January 2017. The treatment structure comprised five levels of sulphur (0, 20, 40, 60, and 80 kg ha-1) along with the recommended fertilizer dose. The variation in yield and the changes in starch, sugars, cysteine, methionine, protein, and oil content with kernel development as influenced by the sulphur application were studied. The yield attributes viz., number of pods pot-1, pod and kernel yield pot-1 and shelling percentage, were remarkably influenced due to the application of sulphur up to 60 kg ha-1, which was comparable with S @ 40 kg ha-1 and had an adverse effect with S @ 80 kg ha-1. In all the stages of sampling (30 DAS, 15 DOP (75 DAS), 30 DOP (90 DAS) and at harvest stage), starch, total sugars, reducing and non-reducing sugars of groundnut were found to decrease with increasing S levels with control recording the highest value and S @ 60 kg ha-1 recording the lowest value. During crop growth, protein content and sulphur-containing amino acids viz., cysteine, and methionine showed an increasing trend up to 60 kg S ha-1 application and recorded comparable values with S @ 40 kg ha-1. Similarly, oil content in the kernel steadily increased with stages of kernel development. The highest oil content at all stages of kernel development was recorded at 60 kg S ha-1.
INTRODUCTION
Groundnut (Arachis hypogaea L.), the king of oilseed crops, is the third most important oilseed crop of the world cultivated widely in 96 countries (Upadhaya et al., 2003). Though the share of groundnut to the total oilseed production in India has been falling since 1950, from 70 per cent to the present level of 33 per cent, groundnut is still the major oilseed crop in India which accounts for about 27 per cent of the global area and contributes 19 per cent to world groundnut production (Rai et al., 2016). Tamil Nadu ranks fourth in terms of groundnut area (4.419 lakh ha) and third in production (9.737 lakh tonnes, Singh, 2014). Sulphur, the fourth major plant nutrient after nitrogen, phosphorus, and potassium, is indispensable for the appropriate plant growth and development (Anjum et al., 2012). Sulphur is inevitable for oilseed crops as it is involved in the synthesis of essential amino acids and is a vital component of coenzymes involved in oil synthesis (Chaudhary, 2009). Sulphur has been reported to influence the productivity of oilseed crops and total oil content considerably (Jankowski et al., 2008;Egesel et al., 2009). It is rightly called as the master nutrient of all oilseed crops as each unit of fertilizer sulphur generates 3-5 units of edible oil (Ramdevputra et al., 2010). Their sulphur requirement for proper growth and yield is more than that of many other crops (Fahmina et al., 2013).
Areas of sulphur deficiency are becoming widespread throughout the world due to the use of high-analysis fertilizers with low S returns with farmyard manure, high yielding varieties, and intensive agriculture, declining use of sulphur -containing fungicides, and reduced atmospheric inputs caused by stringent emission regulations (Tandon, 1995;CeCeotti, 1996;Randhawa and Arora, 2000;Nader and Nadia, 2011). As the intensity of cropping is gradually increasing, the response of oilseeds to sulphur is also increasing 107 | 1-3 | 75 (Ghosh et al., 2002) and the variable response of groundnut to sulphur has been reported by many workers (Kumar et al., 2008;Ramdevputra et al., 2010;Giri et al., 2011). Hence, this investigation was attempted to study the importance of sulphur in realizing yield and quality of groundnut crop and to study the role of sulphur in influencing the biochemical constituents of groundnut.
Experimental description
A pot culture experiment was conducted at the Radioisotope Laboratory, Department of Soil Science and Agricultural Chemistry, Tamil Nadu Agricultural University, Coimbatore during October 2016 and January 2017. The experimental soil was sandy loam in texture, non-calcareous, and taxonomically classified as an Inceptisol (Vertic Ustropept) and was slightly alkaline in soil reaction (pH 7.63), non-saline (EC 0.18 dS m -1 ) with CEC of 13.09 cmol (p+) kg -1 . The soil was medium in available nitrogen (300 kg ha -1 ), phosphorus (15.5 kg ha -1 ), and high in available potassium (526 kg ha -1 ) with an organic carbon content of 4.70 g kg -1 . The available sulphur status of the soil was deficient (7.19 mg kg -1 ). The soil was sufficient with respect to available micronutrients like Zn,Fe,Mn,and Cu 2.57,21.20,8.80, and 2.15 mg kg -1, respectively. The soil was non-calcareous in nature with the calcium carbonate content of 4.30 per cent.
The groundnut variety CO 7 obtained from Department of Oilseeds, Centre for Plant Breeding and Genetics, TNAU was used in this study. Nutrients were applied to the pots on a soil weight basis. All the pots received uniform application of nitrogen @ 25 kg ha -1 , phosphorus @ 50 kg ha -1, and potassium @75 kg ha -1, which were applied through urea, diammonium phosphate, and muriate of potash (Crop Production Guide, 2012). Sulphur was applied basally @ 0, 20, 40, 60 and 80 kg ha -1 in the form of elemental sulphur along with Thiobacillus at 21 days before sowing. The crop was harvested at 105 days after sowing, and the yield was recorded.
Collection of samples
Plant samples were collected from the pots at vegetative (30 DAS), 15 days of podding (DOP) (75 DAS), 30 DOP (90 DAS), and at harvest stage. At 15 DOP, 30 DOP, and at maturity (45 DOP), kernels were separated from the plant for biochemical analysis (total sugars, reducing sugars, starch, cysteine, methionine, and protein) which was carried out in the fresh samples. The plants were uprooted at vegetative (30 DAS), 15 DOP, 30 DOP, and at harvest and separated into shoot and pod samples. The pod samples after using for biochemical analysis were oven-dried, and the yield was recorded.
Total sugars and starch was estimated by the Anthrone method (Hodge and Hoferiter, 1962), and reducing sugars by Nelson-Somogyi Method (Somogyi, 1952). Non reducing sugar content of the sample was computed by taking the difference between total sugars and reducing sugars. Protein content was estimated by Lowry's method (Lowry et al., 1951). Cysteine and methionine were estimated by spectrophotometric method (Gaitonde, 1967;Horn et al., 1946). The oil content in the samples was estimated by the method of A. O. A. C. (1960).
Statistical analysis
The experimental data were statistically analyzed as suggest ed by Gomez and Gomez (1984). For significant results, the critical difference was worked out at 5 per cent level.
Effect of sulphur levels on yield parameters and yield of groundnut
Pod yield A significant variation in pod yield was recorded due to the application of sulphur (Table 1). The highest pod yield of 23.42 g pot -1 was recorded by S@ 60 kg ha -1 and was comparable with S@ 40 kg ha -1 (21.95 g pot -1 ). The increase in pod yield was 54.69 and 44.98 per cent over control for the addition of S@60 and 40 kg ha -1, respectively. Yield enhancement in groundnut with the addition of sulphur has been reported by many workers (Giri and was comparable with S@40 kg ha -1 (0.228%). Non reducing sug trend as that of reducing sugars and showed a reducing trend from 4.82 An increase in the total, reducing and non reducing sugars with sulfur fertil stage of kernel development (15 DOP). Fazli et al., (2010) reported a s reducing and non reducing sugar content of seeds at an early stage of reducing, non reducing, and total sugars in sulfur applied treatments was ob 45 DOP (105 DAS). This is in line with the findings of Sukhija et al., (1987) i Cysteine and methionine content Sulfur application had a profound influence on cysteine and methionine con registered the highest cysteine content of 0.60 (15 DOP), 0.66 (30 DOP), an was on par with sulfur @ 40 kg ha -1 at all stages of kernel development (Ta ha -1 recorded the highest methionine content of 0.85, 1.40 and 1.92 per ce (90 DAS) and at 45 DOP (maturity) respectively (Fig. 2). Though at 15 DOP ( significantly different from each other, S@ 60 kg ha -1 was comparable with methionine content of 1.34 per cent at 30 DOP (90 DAS) and 1.86 per cent Sulphur nutritional deficiency has previously been reported to have a str concentration (Macnicol, 1983). In the absence of sulfur, the content of su affected, and the work of Vinod Kumar et al.,(1989) lends support to thi methionine content with sulfur application corroborate with the findings of found that sulfur fertilization had increased the sulfur-containing amino a Also, Jarvan et al., (2008) reported an increase in cysteine and methionin sulfur fertilization. The increased sulphur content in kernel was significan 0.840**) and methionine (r = 0.997**). This supports the findings of the p and was comparable with S@40 kg ha -1 (0.228%). Non reducing sugar content followed the same trend as that of reducing sugars and showed a reducing trend from 4.82 per cent to 3.49 per cent.
An increase in the total, reducing and non reducing sugars with sulfur fertilization was observed in the early stage of kernel development (15 DOP). Fazli et al., (2010) reported a significant increase in the total, reducing and non reducing sugar content of seeds at an early stage of development. The decrease in reducing, non reducing, and total sugars in sulfur applied treatments was observed at 30 DOP (75 DAS) and 45 DOP (105 DAS). This is in line with the findings of Sukhija et al., (1987) in groundnut. Cysteine and methionine content Sulfur application had a profound influence on cysteine and methionine content (Fig.2). Sulfur @ 60 kg ha -1 registered the highest cysteine content of 0.60 (15 DOP), 0.66 (30 DOP), and 1.59 per cent at maturity and was on par with sulfur @ 40 kg ha -1 at all stages of kernel development ( Dutta et al., 2015;Saha et al., 2015); Response to the increasing level of sulphur might be ascribed to an adequate supply of nutrients resulted in high production of photosynthates and their translocation to sink (Tomer et al., 1997;Patel et al., 2009). The differential impact of treatments was quite clear on the kernel yield of groundnut and the highest kernel yield of 16.90 g pot -1 (Table 1) was recorded by the addition of S at 60 kg ha -1 .
Enhancing the dose of sulphur beyond 60 kg ha -1 did not produce any significant advantage rather than a decrease in seed yield. The higher dose of sulphur fertilizer beyond 80 kg ha -1 decreased the seed yield considerably, and the negative response to higher sulphur might be due to the imbalance and toxic effect caused by increasing sulphur level. This result showed that pod yield of groundnut was increased with the increase in sulphur application up to a certain limit, and similar results were reported in groundnut (Dutta et al., 2015) and sunflower (Nasreen and Huq, 2002).
Shelling percentage
Imposed sulphur treatments had a significant influence on shelling percentage, which varied between 65.12 to 72.18 per cent (Table 1). The addition of S@ 60 kg ha -1 recorded significantly higher shelling percentage (72.18) and was found to be on par with 40 and 80 kg S ha -1 .
The increase in shelling percentage due to 60 kg S ha -1 was 10.84 per cent over control, and a similar increase in shelling percentage with sulphur application was reported by Singh and Singh (2016).
Starch
In the early stages of kernel development, sulphur application had a profound influence on starch content. Sulphur @ 60 kg ha -1 recorded significantly higher starch content (22.83 %). With development, the mean starch content decreased from 12.64 per cent at 30 DOP (90 DAS) to 8.45 per cent at maturity (Figure 1). Sulphur applied @ 60 kg ha -1 recorded the lowest starch content at 30 and 45 DOP (at maturity). Starch may serve as a temporary reservoir of energy to be made available during the period of maximum oil synthesis (15 to 30 DOP) by its conversion to glucose and consequently providing various precursors for fatty acid synthesis. A decrease in starch content with the advancement in the kernel development stage was also reported by Sukhija et al., (1987) in groundnut.
Application of sulphur affected adversely the starch content, which might be due to increased conversion of starch to oil in the grain of oil crops with increasing addition of sulphur. Starch is broken to produce glucose -1 phosphate, which ultimately enters the glycolytic pathway, resulting in the cent (no sulfur) to 26.66 per cent in S@ 60 kg ha -1 .
Accumulation of soluble protein in developing seeds during the e indicative of the synthesis of enzymes and membrane protein accumulation of oil. Improvement in protein content is of paramou the building block of the living system. Sulfur is a constituen methionine, cysteine, and cystine. It also helps in the conversion o protein. Sulfur application resulted in increased synthesis of me increased protein content, which is in accordance with the findin structure is essential for protein formation, and sulfur provides increasing the protein content. These results are in support of the The results of the present study confirm the observations of Fa sulfur increased the soluble protein content during seed developm the protein content of groundnut kernel by 36.90 per cent in S@6 40 kg ha -1 .
Oil content
Oil content in the kernel steadily increased with stages of kern 20.46 per cent recorded at 15 DOP increased to 37.15 per cent matured kernel. The treatment which received S@ 60 kg ha -1 rec stages of kernel development (Fig. 3). The maximum value of oil content in the kernel was observed wit content due to sulfur fertilization might be the outcome of better favourable environment created by sulfur application. As sulfur is availability of sulfur might have favourably influenced the s responsible for higher oil content. Sulfur is also known to be invo primary fatty acids, several enzymes catalyzing metabolic proce lipids.
According to Kumar and Yadav (2007), the increase in oil conte might be due to the involvement of sulfur in the electron tran between kernel sulfur content with oil content (r =0.959 **) draw in oil content with sulfur application has earlier been reported by (1994); Jena (2006); Noman et al., (2015)).
Protein content
Sulfur application had a significant influence on protein content (Fig. 3), and it showed an increase from control to S @ 60 kg ha -1 . Sulphur @ 60 kg ha -1 recorded comparable values with S@ 40 kg ha -1 at 30 DOP (90 DAS) and maturity. Protein content in the matured kernel ranged between 24.32 per cent (no sulfur) to 26.66 per cent in S@ 60 kg ha -1 .
Accumulation of soluble protein in developing seeds during the early stages of seed development is indicative of the synthesis of enzymes and membrane proteins required for the synthesis and accumulation of oil. Improvement in protein content is of paramount importance as it is considered as the building block of the living system. Sulfur is a constituent of the essential amino acid viz., methionine, cysteine, and cystine. It also helps in the conversion of these amino acids into high-quality protein. Sulfur application resulted in increased synthesis of methionine, cysteine, and resulted in increased protein content, which is in accordance with the findings of Tathe (2008). An appropriate structure is essential for protein formation, and sulfur provides disulfide chains and thus helps in increasing the protein content. These results are in support of the findings of Babhulkar et al., (2000).
The results of the present study confirm the observations of Fazli et al., (2010) that the supply of sulfur increased the soluble protein content during seed development. Application of sulfur increased the protein content of groundnut kernel by 36.90 per cent in S@60 kg ha -1 and 32.6 per cent in S @ 40 kg ha -1 .
Oil content
Oil content in the kernel steadily increased with stages of kernel development. The oil content of 20.46 per cent recorded at 15 DOP increased to 37.15 per cent at 30 DOP to 49.18 per cent in the matured kernel. The treatment which received S@ 60 kg ha -1 recorded the highest oil content at all stages of kernel development (Fig. 3). The maximum value of oil content in the kernel was observed with S@ 60 kg ha -1 . The increase in oil content due to sulfur fertilization might be the outcome of better availability of nutrients owing to the favourable environment created by sulfur application. As sulfur is an integral part of oil, the increased availability of sulfur might have favourably influenced the synthesis of essential metabolites responsible for higher oil content. Sulfur is also known to be involved in the increased conversion of primary fatty acids, several enzymes catalyzing metabolic process which promotes biosynthesis of lipids.
According to Kumar and Yadav (2007), the increase in oil content with an increase in sulfur dose might be due to the involvement of sulfur in the electron transport chain. The strong correlation between kernel sulfur content with oil content (r =0.959 **) draws support to the finding. An increase in oil content with sulfur application has earlier been reported by many workers (Mishra and Agarwal (1994); Jena (2006); Noman et al., (2015)).
Volume xxx | Issue xxxxx | 4
(90 DAS) and at 45 DOP (maturity) respectively (Fig. 2). Though at 15 DOP (75 DAS) all the treatments were significantly different from each other, S@ 60 kg ha -1 was comparable with S@40 kg ha -1, which recorded a methionine content of 1.34 per cent at 30 DOP (90 DAS) and 1.86 per cent at maturity.
Sulphur nutritional deficiency has previously been reported to have a strong negative effect on cysteine concentration (Macnicol, 1983). In the absence of sulfur, the content of sulfur-containing amino acids was affected, and the work of Vinod Kumar et al.,(1989) lends support to this. The increase in cysteine and methionine content with sulfur application corroborate with the findings of Dwivedi and Bapat (1998), who found that sulfur fertilization had increased the sulfur-containing amino acids in rapeseed and sunflower. Also, Jarvan et al., (2008) reported an increase in cysteine and methionine content in wheat grain due to sulfur fertilization. The increased sulphur content in kernel was significantly correlated with cysteine (r = 0.840**) and methionine (r = 0.997**). This supports the findings of the present study. Volume xxx | Issue xxxxx | 4 tarch and total sugars at different stages of kernel development in fluence on cysteine and methionine content (Fig.2). Sulfur @ 60 kg ha -1 nt of 0.60 (15 DOP), 0.66 (30 DOP), and 1.59 per cent at maturity and at all stages of kernel development (Table 5). Sulphur applied @ 60 kg content of 0.85, 1.40 and 1.92 per cent at 15 DOP (75 DAS), 30 DOP espectively (Fig. 2). Though at 15 DOP (75 DAS) all the treatments were r, S@ 60 kg ha -1 was comparable with S@40 kg ha -1, which recorded a at 30 DOP (90 DAS) and 1.86 per cent at maturity.
reviously been reported to have a strong negative effect on cysteine he absence of sulfur, the content of sulfur-containing amino acids was mar et al., (1989) lends support to this. The increase in cysteine and ication corroborate with the findings of Dwivedi and Bapat (1998), who creased the sulfur-containing amino acids in rapeseed and sunflower. an increase in cysteine and methionine content in wheat grain due to lphur content in kernel was significantly correlated with cysteine (r = **). This supports the findings of the present study.
cysteine and methionine at different stages of kernel development Volume xxx | Issue xxxxx | 4 tarch and total sugars at different stages of kernel development in fluence on cysteine and methionine content (Fig.2). Sulfur @ 60 kg ha -1 nt of 0.60 (15 DOP), 0.66 (30 DOP), and 1.59 per cent at maturity and at all stages of kernel development (Table 5). Sulphur applied @ 60 kg content of 0.85, 1.40 and 1.92 per cent at 15 DOP (75 DAS), 30 DOP espectively (Fig. 2). Though at 15 DOP (75 DAS) all the treatments were r, S@ 60 kg ha -1 was comparable with S@40 kg ha -1, which recorded a at 30 DOP (90 DAS) and 1.86 per cent at maturity.
reviously been reported to have a strong negative effect on cysteine he absence of sulfur, the content of sulfur-containing amino acids was mar et al., (1989) lends support to this. The increase in cysteine and ication corroborate with the findings of Dwivedi and Bapat (1998), who creased the sulfur-containing amino acids in rapeseed and sunflower. an increase in cysteine and methionine content in wheat grain due to lphur content in kernel was significantly correlated with cysteine (r = **). This supports the findings of the present study.
cysteine and methionine at different stages of kernel development formation of acetyl Co A. Sulphur being an essential component of enzymes helps in bringing about a higher turnover of starch to oil and protein leaving behind less starch in the grain. This confirms the findings of Yadav and Singh (1970).
Total, reducing and non reducing sugars
Total sugars varied from 4.94 per cent (control) to 5.92 per cent (S@ 60 kg ha -1 ) at 15 DOP (Table 2 & Figure 1). The variation was between 3.97 (S@ 60 kg ha -1 ) to 4.58 (control) per cent at 30 DOP and from 3.36 to 4.31 per cent at maturity. At all stages of sampling, S@80 kg ha -1 and S@ 40 kg ha -1 recorded comparable values. Sulphur applied treatments recorded lower reducing sugar content at 15 DOP, 30 DOP, and 45 DOP than control. At maturity, the lowest value was recorded by S@ 60 kg ha -1 (0.201 %) and was comparable with S@40 kg ha -1 (0.228%). Non reducing sugar content followed the same trend as that of reducing sugars and showed a reducing trend from 4.82 per cent to 3.49 per cent. An increase in the total, reducing and non reducing sugars with sulphur fertilization was observed in the early stage of kernel development (15 DOP). Fazli et al., (2010) reported a significant increase in the total, reducing and non reducing sugar content of seeds at an early stage of development. The decrease in reducing, non reducing, and total sugars in sulphur applied treatments was observed at 30 DOP (75 DAS) and 45 DOP (105 DAS). This is in line with the findings of Sukhija et al., (1987) in groundnut.
Cysteine and methionine content
Sulphur application had a profound influence on cysteine and methionine content (Figure 2). Sulphur @ 60 kg ha -1 registered the highest cysteine content of 0.60 (15 DOP), 0.66 (30 DOP), and 1.59 per cent at maturity and was on par with sulphur @ 40 kg ha -1 at all stages of kernel development (Table 5). Sulphur applied @ 60 kg ha -1 recorded the highest methionine content of 0.85, 1.40 and 1.92 per cent at 15 DOP (75 DAS), 30 DOP (90 DAS) and at 45 DOP (maturity) respectively (Figure 2). Though at 15 DOP (75 DAS) all the treatments were significantly different from each other, S@ 60 kg ha -1 was comparable with S@40 kg ha -1, which recorded a methionine content of 1.34 per cent at 30 DOP (90 DAS) and 1.86 per cent at maturity.
Sulphur nutritional deficiency has previously been reported to have a strong negative effect on cysteine concentration (Macnicol, 1983). In the absence of sulphur, the content of sulphurcontaining amino acids was affected, and the work of Vinod Kumar et al.,(1989) lends support to this. The increase in cysteine and methionine content with sulphur application corroborate with the findings of Dwivedi and Bapat (1998), who found that sulphur fertilization had increased the sulphur-containing amino acids in rapeseed and sunflower. Also, Jarvan et al., (2008) reported an increase in cysteine and 107 | 1-3 | 78 methionine content in wheat grain due to sulphur fertilization. The increased sulphur content in kernel was significantly correlated with cysteine (r = 0.840**) and methionine (r = 0.997**). This supports the findings of the present study.
Protein content
Sulphur application had a significant influence on protein content (Figure 3), and it showed an increase from control to S @ 60 kg ha -1 . Sulphur @ 60 kg ha -1 recorded comparable values with S@ 40 kg ha -1 at 30 DOP (90 DAS) and maturity. Protein content in the matured kernel ranged between 24.32 per cent (no sulphur) to 26.66 per cent in S@ 60 kg ha -1 .
Accumulation of soluble protein in developing seeds during the early stages of seed development is indicative of the synthesis of enzymes and membrane proteins required for the synthesis and accumulation of oil. Improvement in protein content is of paramount importance as it is considered as the building block of the living system. Sulphur is a constituent of the essential amino acid viz., methionine, cysteine, and cystine. It also helps in the conversion of these amino acids into high-quality protein. Sulphur application resulted in increased synthesis of methionine, cysteine, and resulted in increased protein content, which is in accordance with the findings of Tathe (2008). An appropriate structure is essential for protein formation, and sulphur provides disulfide chains and thus helps in increasing the protein content. These results are in support of the findings of Babhulkar et al., (2000).
The results of the present study confirm the observations of Fazli et al., (2010) that the supply of sulphur increased the soluble protein content during seed development. Application of sulphur increased the protein content of groundnut kernel by 36.90 per cent in S@60 kg ha -1 and 32.6 per cent in S @ 40 kg ha -1 .
Oil content
Oil content in the kernel steadily increased with stages of kernel development. The oil content of 20.46 per cent recorded at 15 DOP increased to 37.15 per cent at 30 DOP to 49.18 per cent in the matured kernel. The treatment which received S@ 60 kg ha -1 recorded the highest oil content at all stages of kernel development (Figure 3).
The maximum value of oil content in the kernel was observed with S@ 60 kg ha -1 . The increase in oil content due to sulphur fertilization might be the outcome of better availability of nutrients owing to the favourable environment created by sulphur application. As sulphur is an integral part of oil, the increased availability of sulphur might have favourably influenced the synthesis of essential metabolites responsible for higher oil content.
Sulphur is also known to be involved in the increased conversion of primary fatty acids, several enzymes catalyzing metabolic process which promotes biosynthesis of lipids.
According to Kumar and Yadav (2007), the increase in oil content with an increase in sulphur dose might be due to the involvement of sulphur in the electron transport chain. The strong correlation between kernel sulphur content with oil content (r =0.959 **) draws support to the finding. An increase in oil content with sulphur application has earlier been reported by many workers (Mishra and Agarwal (1994); Jena (2006);Noman et al., (2015)).
CONCLUSION
The study has brought out the response of groundnut (CO 7) to graded levels of sulphur on the yield variation and changes in starch, sugars, cysteine, methionine, protein and oil content with kernel development. Sulphur application at 60 kg ha -1 remained on par with S @ 40 kg ha -1 in all of the growth and biochemical parameters, which emphasizes that sulphur fertilization at 40 kg ha -1 would be adequate for improving the yield and quality of groundnut.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-03-05T10:35:50.071Z", "created": "2020-03-01T00:00:00.000", "id": "216269551", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "http://masujournal.org/107/190335.pdf", "pdf_hash": "deedeaaf5783ca6419fdb14d9cc59e8f6a085306", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:32902", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "d54b16afd663fd5c248b3c46a2c18cdfc0471365", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Global trends in antimicrobial resistance on organic and conventional farms
The important hypothesis that organic livestock management reduces the prevalence of antimicrobial resistance is either fiercely supported or bitterly contested. Yet, empirical evidence supporting this view remains fragmentary, in part because relationships between antimicrobial use and drug resistance vary dramatically across contexts, hosts, pathogens, and country-specific regulations. Here, we synthesize global policies and definitions of ‘organic’ and ask if organic farming results in notable reductions in the prevalence of antimicrobial resistance when directly examined alongside conventional analogs. We synthesized the results of 72 studies, spanning 22 countries and five pathogens. Our results highlight substantial variations in country-specific policies on drug use and definitions of ‘organic’ that hinder broad-scale and generalizable patterns. Overall, conventional farms had slightly higher levels of antimicrobial resistance (28%) relative to organic counterparts (18%), although we found significant context-dependent variation in this pattern. Notably, environmental samples from organic and conventional farms often exhibited high levels of resistance to medically important drugs, underscoring the need for more stringent and consistent policies to control antimicrobial contaminants in the soil (particularly on organic farms, where the application of conventional manure could faciliate the spread antimicrobial resistance). Taken together, these results emphasize the challenges inherent in understanding links between drug use and drug resistance, the critical need for global standards governing organic policies, and greater investment in viable alternatives for managing disease in livestock.
in country-specific regulations undermine efforts to develop universal standards, exhaust consumer confidence, and weaken economic efficiency. These critical gaps in knowledge represent a major impediment to policy and management decisions and hamper investment in the research and development needed to generate viable and scalable non-pharmacological alternatives for managing disease in livestock.
As a first step in addressing these concerns, we begin by synthesizing global differences in policies and definitions of 'organic.' Broadly speaking, conventional livestock production focuses on technologies for increased productivity, such as high-yielding breeds, modern feeding techniques and veterinary health products, as well as (synthetic) fertilizers and pesticides. In contrast, organic practices focus on reducing antimicrobial use in livestock by integrating cultural, biological, nutritional, and mechanical methods to ensure environmentally safe and residue-free foods, along with improved animal welfare standards [16][17][18][19] . In general, organic production strives to provide animals with a more spacious and enriched environment, access to an outdoor range, and limited group sizes, all of which ostensibly improve animal health and reduce the need for medications, including antimicrobials. However, a central obstacle to identifying the more specific broad-scale and generalizable differences between 'organic' and 'conventional' farming practices resides in the lack of a global or even regional consensus about standard practices.
After synthesizing current differences between organic and conventional farming practices (Supplementary Table S1), we conduct a literature review to examine where and when organic farming results in notable reductions in the prevalence of antimicrobial resistance over the range of contexts in which it has been examined alongside conventional analogs.
Results
We identified 1,833 unique studies published between 2000 and 2022, as well as six grey literature studies (e.g., WHO website, with reports on point prevalence of antimicrobial resistance). After the references were screened, 1,744 studies and all six grey literature publications were removed (Fig. 1a). After assessing the remaining 89 references for eligibility, 17 studies were excluded, leaving 72 studies that met the inclusion criteria: 46% (n = 33). These studies spanned North America (46%, n = 33), Europe (36%, n = 26), Asia (14%, n = 10), Oceania (3%, n = 2) and South America (1%, n = 1) (Fig. 1b). All the surveys covered antimicrobial classifications based on the World Health Organization, and classifications including those considered 'critically important' and 'highly important' for human medicine. For all drug acronyms, see Supplementary Table S2. We first discuss overall global patterns, which largely arise due to the pronounced differences in region-specific regulations for the use of antimicrobials in livestock and variation in 'organic' classification and certification standards (Supplementary Table S1). Then, we highlight key differences across host-region-and pathogen-specific contexts.
Host-specific patterns
Across all geographic regions, resistance patterns were highly variable among different host classes, with overall resistance higher in isolates from conventional farms. Conventional farms reported a higher AMR prevalence in isolates from cattle, chicken, pigs, and turkey. For instance, in isolates from cattle, the prevalence of AMR was 14.5% on conventional farms and 9% on organic farms (Fig. 2c and d). For chicken isolates, AMR prevalence was higher on conventional farms (22%) compared to organic farms (13.5%). Similar trends were reported for other hosts (Fig. 2c and d). On conventional pig farms 24.5% of isolates were resistant whereas on organic farms, only 15% of isolates were resistant. On conventional turkey farms, the prevalence of AMR was 46% as compared to 22.5% on organic farms. For environmental samples, the median prevalence of AMR in environmental isolates was slightly higher on organic farms, 16% relative to conventional farms, 11.5% (Fig. 2c and d). These patterns, however, were also highly variable across geographical regions.
AMR patterns across broad geographic regions
Examining region-specific variation in the prevalence of antimicrobial resistance may help identify areas where specific management practices warrant more attention (i.e., 'hot spots'). For example, in parts of the US, the prevalence of AMR was 64% (n = 135 isolates) and 35% (n = 135 isolates) in environmental isolates from conventional and organic farms, respectively. Countries with low antimicrobial usage in food production animals, like Sweden and New Zealand, reported low AMR prevalence from environmental samples. For instance, Sweden reported 5% (n = 725 isolates) AMR prevalence on both organic and conventional farms, while New Zealand reported that on conventional farms, only 5% of isolates were resistant (n = 814 isolates) and 3.8% were resistant (n = 814 isolates) on organic farms (Fig. 3).
Patterns for other hosts were also highly variable across broad geographic scales. However, for chickens, the prevalence of AMR was relatively similar across organic and conventional farms (Fig. 3), though patterns from organic farms were particularly notable. In Georgia, USA, the prevalence of AMR in isolates from chickens was marginally higher on organic farms (59%, n = 60 isolates) relative to conventional farms (40%, n = 60 isolates). In California, USA, the prevalence of AMR in chicken isolates was notably high on both farm types: 78% (n = 132 isolates) on conventional farms and 75% (n = 132 isolates) on organic farms. These results were primarily driven by drug-resistant Campylobacter spp, discussed in more detail below.www.nature.com/scientificreports/
Region-specific patterns in foodborne pathogens
Studies included in our review covered patterns of AMR prevalence in five pathogens sampled from a total of 61,299 isolates: Escherichia coli, Salmonella spp, Campylobacter spp, Enterococcus, and Staphylococcus aureus. These isolates were sampled from organic farms and conventional farms in Asia (n = 1,164 isolates), Europe (n = 11,759 isolates), North America (n = 44,979 isolates), Oceania (n = 3,085 isolates), and South America (n = 312 isolates).
European Union and the United Kingdom
Across Europe, and on both conventional and organic farms, we found high levels of resistance to quinolones, like norfloxacin and ofloxacin. These drugs are considered 'critically important antimicrobials' and were restricted in 2009 by the EU. For example, the prevalence of resistance in Campylobacter was 91% (CI: 82-100%, n = 43 isolates) on both conventional and organic farms (Fig. 5c). Notably, however, ampicillin-resistant Campylobacter were slightly more prevalent on organic farms (79%, CI: 67-91%, n = 43 isolates) relative to conventional farms Temporal trends in the prevalence of antimicrobial resistance from different hosts (including environment samples) on organic and conventional farms. Note: 12 studies were excluded from this analysis because they did not report the year that the data were collected. Sixty surveys with complete sampling dates were included in the analysis. Globally, the prevalence of antimicrobial resistance was slightly lower on (a) organic farms (18%, n = 56) relative to conventional farms (28%, n = 53). However, antimicrobial resistance appears to be increasing on both (c) organic and (d) conventional farms. From 2001 to 2020, the prevalence of antimicrobial resistance in isolates from organic farms (n = 29,417) increased from 10% (CI: 8-13%) to 24% (CI: 20-28%), while the prevalence of drug resistance in isolates from conventional farms (n = 31,882) increased from 18% (CI: 15-23%) to 37% (CI: 32-43%). Examining host-specific patterns indicates that the prevalence of antimicrobial resistance from cattle, chicken, pig, and turkey isolates was higher on conventional farms as compared to organic farms. However, antimicrobial resistance was slightly higher in environmental samples collected from organic farms compared to conventional farms. Data represent the median ± the first and third quartile ranges.www.nature.com/scientificreports/(66%, CI: 56-76%, n = 41 isolates). For Enterococcus, 100% of isolates (n = 36 isolates) from conventional farms were cefoxitin-resistant (Fig. 7a). In contrast, resistance to erythromycin was relatively low on both conventional and organic farms (6%, CI: 1-8%, n = 284 isolates, Fig. 7a). We also found that the prevalence of rifampicinresistant Enterococcus was higher on organic farms (19%, CI: 15-22%, n = 134 isolates) as compared to conventional farms (5%, CI: 3-7%, n = 134 isolates, Fig. 6b).
Figure 3.
Global patterns of antimicrobial resistance in isolates collected from organic and conventional farms. Studies spanned four hosts (cattle, pig, chicken, and turkey) and environmental samples collected from conventional and organic farms throughout North America, Europe, Asia, Oceania, and South America. Pie charts show the prevalence of antimicrobial resistance on conventional farms (outer pie chart, n = 66) relative to their organic counterparts (inner pie chart, n = 69). Geographic regions with a single pie (i.e., outer pie only) represent areas lacking data from organic farms. The map (Fig. 3) was generated with QGIS version 3.24.0-Tisler59 .
No data for organic farms in this region were available.
Discussion
Our findings suggest that overall, antimicrobial resistance (AMR) was slightly lower in organic livestock production systems relative to their conventional counterparts, while also revealing significant context-dependent variation in this pattern. Specifically, the prevalence of AMR was 18% on organic farms and 28% on conventional farms. However, the substantial region-and country-specific variations in regulations and policies governing organic farming obfuscate broad-scale and generalizable patterns on how organic farming practices affect AMR.
Countries have taken markedly different approaches to the guidelines and regulatory agencies that govern the use of antimicrobials in both conventional and organic livestock production (Supplementary Table S1). Further, in practice, these regulations are often overlooked or met with strong resistance at the industry-level. For example, the European Union (EU) banned the use of antimicrobials for growth promotion (APGs) in livestock production systems in 2006 20 and the United States followed suit in 2017 21,22 . These policy changes, however, resulted in a 'repackaging' of both labeling and marketing practices for these products, characterizing them as 'prophylactic therapeutics' instead of 'growth promoters' 22,23 . Moreover, the US and countries across Europe reported an increasein the use of antimicrobials for prophylactic purposes after the ban of AGPs 6,24 . Thus, these S2.
well-intentioned policy changes backfired due, in part, to loopholes that recharacterized the technical nature of the antimicrobialsa legal but questionable work-around that remains unaddressed. More stringent regulations on drug use in livestock matter not just for conventional farming but also for organic farming, which seeks to limit but not always eliminate antimicrobial usage in production animals (Supplementary Table S1). In the US, use of antimicrobials is prohibited for organic livestock, while the European regulation for organic dairy herds allows a maximum of three treatments with antimicrobials per cow per year [25][26][27] . Denmark, the United Kingdom, and Norway adopted their own regulations, imposing more stringent prohibitions on the use of antimicrobials for growth promotion and requiring supervision by a veterinarian for the use of a limited number of antimicrobials 28,29 . Outside the EU and US, however, conventions governing policies on organic production can be more variable within a country and can become even more challenging to standardize on a national level (Supplementary Table S1). For instance, in Canada, regulations can vary both within and across provinces for products that are distributed and sold solely within those regions 30 .
These differences underscore the need for a more comprehensive and global governance framework to review the science underpinning policies and regulations for both conventional and organic livestock production systems. Such advances are critical to improving investment in research and development to provide viable, nonpharmacological alternatives for disease management in livestock and to move toward global standards, policies, and regulations for livestock production systems. These directives are essential to understanding how, when, or where specific practices of 'organic' livestock production reduce the prevalence of AMR. Investment in research and development in these areas is crucial to identifying practical and scalable solutions for farmers who depend on antimicrobials to prevent outbreaks and maintain herd health.
The high levels of drug resistance found in environmental samples further emphasize the need to regulate antimicrobial use and contamination across livestock production systems (Fig. 2c and d). The use of manures S2.
from various sources on organic crops differs depending on the specific regulatory framework (e.g., US regulations allow use of manures on organic crops from conventional Confined Animal Feeding Operations (CAFOs), as opposed to EU regulations, which limit the use of manures from "industrial" animal operations). Thus, environmental contamination of antimicrobials may be more prevalent on organic land certified under regulatory frameworks that allow regular use of these industrial manures sourced from conventional farms that routinely use antimicrobials. For example, the use of conventional poultry manure is commonplace in organic grain production in the US, including the production of feed for organic poultry operations. While more detailed studies are needed to address the mechanistic underpinnings of these results, at least three key factors could play an important role. First, transition times between a conventional livestock production system that switches to organic management practices and previous land use patterns can substantially impact the levels and diversity of environmental contaminants. The studies included in our review likely differ in the timing of the transition from conventional to organic farming, especially given country-and region-specific variations in regulations on transition time. For instance, the US National Organic Program standards allow for a three-year transition period from conventional to organic management in livestock systems 31,32 , whereas the UK allows a two-year transition period 33 . We were unable to find specific regulations on transition timeframes for other countries.
Transition times matter because animal manure can increase drug residues in the environment, altering the selective pressures that drive antimicrobial resistance. Additionally, contaminated soil can function as reservoir of plasmids (mobile genetic elements) that can transfer resistance within and across species 34,35 . Recent advances in molecular and gene sequencing technologies have increased our awareness that plasmids can transfer among bacteria as well as to other species (e.g., cows to humans), largely through horizontal gene transfer among microbes in the microbiome, leading to rapid transfer of multi-drug resistance in various hosts. This potentiality may increase the risk of antimicrobial resistance 'spill-over' from soil to livestock, wildlife, and humans [36][37][38][39] . Future studies focused on quantitative risk analyses are needed to help identify approaches to mitigate environmental S2.
Vol contamination of antimicrobials on both organic and conventional farms. Examining how transition times and soil management practices affect the prevalence of pathogens and AMR is an important first step. Environmental contamination of antimicrobials may also contribute to the notably high prevalence of AMR on organic poultry farms (Fig. 3). Across some parts of the US, the prevalence of AMR in poultry was slightly higher on organic farms compared to conventional farms. A similar trend was also reported on organic farms in the UK and Portugal, but we are unable to compare the results to conventional farms in these two countries due to lack of data. While organic practices and regulations are, again, highly variable across different regions, access to outdoor grazing may expose chickens to contaminants in the soil, including antimicrobials and microbes or insects, which can serve as reservoirs for drug resistance 40 . Exposure to antimicrobial contaminants in soils may increase the prevalence of AMR in insect reservoirs and thus in poultry.
Our results highlight that drug resistance in Campylobacter spp.may warrant particular attention in organic poultry systems. For example, 60-80% of global Campylobacter cases originate from poultry products, and 400-500 million cases are reported globally every year 41 . Annually in the US, approximately 310,000 cases of Campylobacter are potentially untreatable due to resistance to azithromycin and ciprofloxacin, two important anti-Campylobacter antibiotics 42 . For Campylobacter, the prevalence of resistance to quinolones was notably high on organic farms in Europe. Yet, quinolones are considered critically important antimicrobials and have been restricted for use in livestock and humans in the EU since 2018 43 . We also found high levels of resistance to drugs considered critically and highly important to human medicine in Asia, Europe, and North America and across a wide gradient of organic and conventional management practices (Fig. 4b, 5a, 5b, 5c, and 5d). Given the public health concerns related to Campylobacter, our results join others in calling for greater regulations in these components of livestock management 44,45 .
Our study had several limitations. First, the variation in organic livestock management practices and regulations at national and regional levels strongly limits the generalizability of our results. Second, despite a comprehensive literature search with broad search terms, our study yielded a relatively limited number of studies (n = 72) and very few studies from low and middle-income countries (e.g., Africa [n = 0], Oceania [n = 2], and S2. South America [n = 1]). Third, our search criteria only included studies written in English and Portuguese. This language limitation may have caused us to miss studies written in other languages. While we recognize and regret this common limitation, additional language searchers are beyond the scope of this current study. In addition, AMR point prevalence surveys use various methodologies for susceptibility testing and thus results are relative though not quantitative per se. For example, the studies included here that report the prevalence of AMR to spectinomycin and lincomycin used binary metrics (excluding intermediate resistance), which may over-or under-estimate the prevalence of resistance. Given these limitations, our results should be interpreted with caution as they capture only a small snapshot of the true state of organic farming practices and global patterns of AMR. Indeed, our review and synthesis serve, in large part, to highlight these discrepancies and the paucity of data required to understand links between AMR use in livestock and broad patterns of AMR. The important hypothesis that organic practices for livestock production reduce the prevalence of antimicrobial resistance is often taken at face value. Yet, as we show here, data to address this hypothesis are largely lacking (as evidenced by the small sample size produced from our literature search). Moreover, rigorous and large sample sizes are especially needed to test this hypothesis because relationships between antimicrobial use and drug resistance vary dramatically across contexts, differing between hosts, pathogens, and country-specific regulations (Figs.1-6). The similarities in patterns of AMR prevalence across broad geographic regions with markedly different practices for regulating drug usage suggest that in some cases organic livestock practices have, at best, marginally reduced the prevalence of AMR. In other cases (e.g., free-range poultry), however, organic farms suffer from a notably high prevalence of AMR that warrants further investigation.
The trends presented here are consistent with previous research indicating high multidrug resistance in E.coli and Salmonella found in livestock 46,47 . Given a projected 14% increase in consumer demand for meat products by 2030 48 , AMR in livestock will continue to increase unless substantial management changes are implemented. Traditional interventions like stringent cleaning, antibiotics, and vaccines are critical for managing herd health and treating disease. In isolation, however, these costly and reactive approaches aimed at limiting pathogen proliferation can select for more virulent and resistant variants and ultimately ease their spread. The growing threat of antimicrobial resistance and consumer demands to reduce the use of antimicrobials in livestock emphasize the critical need to leverage non-pharmacological approaches to prevent and manage disease [49][50][51][52][53] .
Our results underscore the need for multidisciplinary and global approaches that blend organic farming principles and non-pharmacological interventions to reduce routine antibiotic use in livestock. In addition, research on interventions like bacteriophages, probiotics, and increased surveillance of antimicrobial resistance have shown promising results in reducing AMR 1,54 . Furthermore, collaborations among stakeholders (i.e., farmers, researchers, and policymakers in the animal health sector) could help disseminate information and best practices. Unfortunately, the industry continues to move in the opposite direction, particularly with the rapidly expanding trend toward growth of corporate-owned livestock farms that control both the dietary and pharmaceutical regimes of the animals 55 . The lack of regulations and transparency in these practices prevent a clear understanding of when and in what quantities antibiotics are provided in feed, for instance.
Future studies could help formulate scalable solutions for conventional farming practices with benefits for both agriculture and public health. Our review indicates that a key focal area includes a better understanding of how transition times and soil properties influence the prevalence, viability, and retention of pathogens -and the genes that harbor AMR 55,56 . Taken together, these results emphasize the inherent challenges to understanding links between drug use, livestock production practices, and drug resistance. Greater understanding of how, when, and where antimicrobials can be reduced in livestock production systems (e.g., by adopting a subset of select organic-based practices) without a concomitant increase in disease outbreaks would greatly enhance efforts to reduce the evolution of drug resistance and extend the 'shelf life' of these powerful biomedical tools. As our synthesis highlights, we are far from reaching such an understanding. We hope that by synthesizing these challenges, our study catalyzes future empirical research to address these gaps in knowledge.
Literature search strategy
Our initial goal was to examine studies that directly compared patterns of antibiotic resistance from organic and conventional farms within the same region (i.e., US: state; Africa: province; Canada: province; UK: county) and livestock species (e.g., cow, chicken). However, these initial search criteria were too restrictive and yielded only sixty-four studies. Therefore, we expanded the search to include studies that reported antimicrobial resistance from organic farms without always directly comparing their patterns alongside conventional counterparts.
We conducted literature searches for studies published between 2000 and 2022 using three electronic databases (PubMed, Web of Science, and PubAg). We used the following search terms, which we modified slightly for each database. Note, for brevity, we show abbreviated terms (e.g., "livestock names" reflects individual searches for sheep, goats, chickens, etc., and the name of the pathogen reflects individual searches for each pathogeni.e., Campylobacter, E. coli, Salmonella, etc. In the PubMed search, for example, we used the following terms: livestock name AND product OR livestock production OR livestock farm OR name of pathogen OR antimicrobial AND resistant OR agriculture OR conventional OR organic AND agriculture. The full search terms are provided in the supplementary methods S1. Finally, references from other literature reviews 42,42 and all other studies were screened for inclusion. Our search generated 1,836 studies for the initial screening (Fig. 1a).
Study selection
We reviewed all English-language and Portuguese-language articles that directly compared patterns of antimicrobial resistance (AMR) from chicken, turkey, cattle, pigs, and environmental samples from organic and conventional farms in a given geographic region. After screening the reference lists, we excluded reviews, unrelated topics, and book chapters (Fig. 1a). Following the search, all records were exported to Endnote's web citation manager 57 , and duplicates were removed. The records were then exported to a spreadsheet and organized by title, doi, authors, journal, year of publication, and abstract. Finally, the titles and abstracts were screened against the inclusion criteria.
The studies that met the eligibility criteria were retrieved in full text and were thoroughly reviewed. Seventytwo studies met our inclusion criteria (Fig. 1a). We attributed the reduction in sample size to two conditions: (1) our search terms covered general antimicrobial resistance and antimicrobial susceptibility topics and (2) our search focused only on articles written in English and Portuguese. As a way of assessing data quality in our review, we excluded records that did not clearly identify farm types as organic or conventional, gave no geographic information on study location, provided unclear resistance rates, or involved imported products. Data extraction results were stratified according to country name, antimicrobial resistance results, farm type, and pathogens.
Statistical analysis
All data analyses were conducted in R version 4.2.0 58 and QGIS version 3.24.0-Tisler 59. To examine differences in the prevalence of AMR on organic and conventional farms, we used generalized linear models (GLMs) with quasibinomial distributions and log link functions 60 . We built candidate models starting with the full model with all combinations of main effects among relevant biological and methodological factors, while avoiding overfitting. Specifically, to examine overall changes in the prevalence of AMR across the 19-year time frame included in this review, the full model included farm type (organic vs. conventional), country, study year, and their interaction as fixed effects. Then, to examine more fine-scale differences in the prevalence of AMR, the full model examined effects of farm type, host, pathogen, country, and their interactions. We excluded antimicrobial type because the large number of drugs covered in these studies led to overfitting the models.
Following Burnham and Anderson, we compared candidate models using Akaike's information criterion and ΔAIC (the difference in AIC values for the focal model and the model with the lowest AIC, i.e., the 'winning' model) 61 . We conducted model selection analyses using the aictab function in the R package AICcmodavg 62 . We also calculated the Akaike weight (ω), which further quantifies the probability that a model is the most appropriate model relative to the candidate models.ΔAIC less than two and a higher (ω) generally indicates that a model has substantial support while a suite of best models with low weights (ω ~ 0) indicates that no single variable plays a substantial role in mediating patterns of AMR 61 .
Using the Anova function in the R package car 63 , we assessed significance using Wald χ 2 statistics for the winning model. We also evaluated model fits with visual diagnostics, quantile-quantile plots, and residual-versuspredictor plots 64,65 . Note, the MIC (mean inhibitory concentration) values and MIC breakpoints were not used in this analysis; variations in methodologies and criteria used across laboratories and differing epidemiological contexts pose substantial challenges for standardization (in addition the other challenges outlined in the Discussion). Therefore, we focused on estimates of the prevalence of AMR as a relatively consistent measure of resistance across studies. To report the prevalence of AMR in foodborne pathogens, we calculated the pooled prevalence of resistance from each pathogen-drug combination 5,66 using the formula below: Pooled prevalence = Number of isolates resistant Total number of isolates tested
Figure 1 .
Figure 1.(a) Overview of PRISMA-based literature search results and categorization of studies.(b) After literature search and screening, 72 studies satisfied our inclusion criteria. These studies examined 109 farms and 61,299 bacterial isolates. The vast majority (46%) of these studies were in North America (n = 33), 36% were in Europe (n = 26), 14% were in Asia (n = 10), while Oceania and South America contributed to 3% (n = 2) and 1% (n = 1), respectively.
Figure 2 .
Figure 2. Temporal trends in the prevalence of antimicrobial resistance from different hosts (including environment samples) on organic and conventional farms. Note: 12 studies were excluded from this analysis because they did not report the year that the data were collected. Sixty surveys with complete sampling dates were included in the analysis. Globally, the prevalence of antimicrobial resistance was slightly lower on (a) organic farms (18%, n = 56) relative to conventional farms (28%, n = 53). However, antimicrobial resistance appears to be increasing on both (c) organic and (d) conventional farms. From 2001 to 2020, the prevalence of antimicrobial resistance in isolates from organic farms (n = 29,417) increased from 10% (CI: 8-13%) to 24% (CI: 20-28%), while the prevalence of drug resistance in isolates from conventional farms (n = 31,882) increased from 18% (CI: 15-23%) to 37% (CI: 32-43%). Examining host-specific patterns indicates that the prevalence of antimicrobial resistance from cattle, chicken, pig, and turkey isolates was higher on conventional farms as compared to organic farms. However, antimicrobial resistance was slightly higher in environmental samples collected from organic farms compared to conventional farms. Data represent the median ± the first and third quartile ranges. [URL] 4 .
Figure 4. Patterns of antimicrobial resistance in E. coli. The AMR prevalence is shown for the number of isolates (n) examined on organic and conventional farms in each geographic region. We included studies with at least 10 isolates.(a) Asia, n = 215, (b) Europe, n = 9,007, (c) North America, n = 23,845, (d) South America, n = 312, and (e) Oceania, n = 2,379. Data represent the mean ± 95% confidence intervals. The grey shading indicates antimicrobials classified as critically important; the unshaded region indicates highly important antimicrobials. For drug acronyms, see Supplementary TableS2.
Figure 5 .
Figure 5. Patterns of antimicrobial resistance in Salmonella and Campylobacter. The prevalence of antimicrobial resistance is shown for the number of isolates (n) examined on organic and conventional farms in each geographic region. We inlcuded studies with at least 10 isolates.(a) Asia: Salmonella, n = 845, (b) North America, Salmonella, n = 5,230 (c) Europe: Campylobacter, n = 1,214, (d) North America: Campylobacter, n = 14,607. Data represent the mean ± 95% confidence intervals. The grey shading indicates antimicrobials classified as critically important; the unshaded region indicates highly important antimicrobials. For drug acronyms, see Supplementary TableS2.
Figure 6 .
Figure 6. Patterns of antimicrobial resistance in Enterococcus. The prevalence of antimicrobial resistance is shown for the number of isolates (n) examined on organic and conventional farms in each geographic region. We included studies with at least 10 isolates.(a) Asia: Enterococcus, n = 104, (b) Europe: n = 568, (c) North America: n = 811, (d) Oceania: n = 706. Data represent the mean ± 95% confidence intervals. The grey shading indicates antimicrobials classified as critically important; the unshaded region indicates highly important antimicrobials. For drug acronyms, see Supplementary TableS2.
Figure 7 .
Figure 7. Patterns of antimicrobial resistance in S. aureus. The AMR prevalence is shown for the number of isolates (n) examined on organic and conventional farms in each geographic region. We included studies with at least 10 isolates.(a) Europe: n = 970, (b) North America: n = 486. Data represent the mean ± 95% confidence intervals. The grey shading indicates antimicrobials classified as critically important; the unshaded region indicates highly important antimicrobials. For drug acronyms, see Supplementary TableS2.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2023-04-13T13:11:20.130Z", "created": "2023-04-10T00:00:00.000", "id": "258082085", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.nature.com/articles/s41598-023-47862-7.pdf", "pdf_hash": "3871e65c56a452ec888cbeb1fc734c286c786ce6", "pdf_src": "PubMedCentral", "provenance": "peS2o-0075.json.gz:31794", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "8d4b31be4a7b2577d348ecc8842e2f7de9287e82", "year": 2023}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
SPRAY DEPOSITION ON COFFEE LEAVES FROM AIRBLAST SPRAYERS WITH AND WITHOUT ELECTROSTATIC CHARGE
Studies on the quality of applications of plant protection products on coffee crops are lacking. Thus, we studied spray deposition on coffee leaves and losses to the soil from hydropneumatic spraying at different spray volumes and with and without an electrostatic charge. The experiment was set up using randomized blocks in a factorial design (4 x 2 + 1). Spray deposition on the upper, middle and lower parts of the canopy and losses to the soil were evaluated using Brilliant Blue tracer. Applications were made at 200, 300, 400 and 500 L ha with a conventional airblast sprayer (axial fan type) and an airblast sprayer with directed air jets. Applications were also made with an electrostatic sprayer at 130 L ha. Electrostatic spraying resulted in greater spray deposition on the lower part of the coffee canopy compared to non-electrostatic spraying. On the lower and middle parts of the plants, the sprayer equipped with directed air ducts performed better than the sprayer with nozzles arranged along the lateral arcs (axial). The spray volume of the airblast sprayers without electrostatic charge (200 to 500 L ha) did not influence spray deposition on the plant leaves and losses to the soil, which were lower with the electrostatic sprayer.
INTRODUCTION
Coffee crop (Coffea arabica L.) presents several challenges to the application of plant protection products, principally canopy penetration and drift reduction. Plant architecture and large foliage area hinder spray coverage. The air currents from airblast sprayers can also increase the risk of spray drift that can reduce the effectiveness of the treatment and increase environmental contamination. Choosing the correct spray method and spray volume can improve spray deposition on the biological target.
Nevertheless, defining the correct spray volume is difficult given that low volumes can lead to inadequate coverage and high volumes can make applications more difficult, principally because of reduced operational capacity. According to Silva et al. (2008), there is insufficient information on appropriate spray distributions and spray volumes needed to effectively control pests and diseases in coffee crops. Cunha et al. (2005) observed that one of the causes of product loss is spray volume that is inappropriate for specific crop characteristics. The crop canopy is one of the most important of these characteristics (ROSELL POLO et al., 2009). According to Viana et al. (2010), uniform distribution within a specific diameter and using a specific number of droplets can lead to successful applications even at lower spray volumes. Other studies have achieved promising results using lower spray volumes on tree crops (BALAN et al., 2006;FERNANDES et al., 2010). Additionally, electrostatic spraying at lower volumes can improve leaf deposition and reduce drift losses (ZHAO et al., 2008;MASKI;DURAIRAJ, 2010). Some studies have demonstrated the advantages of electrostatic sprayers (MASKI;DURAIRAJ, 2010;DERKSEN et al., 2007;LARYEA;NO, 2005;XIONGKUI et al., 2011). Sasaki et al. (2013) evaluated a pneumatic backpack sprayer and also showed that the electrostatic system was efficient for spraying coffee plants. Electrostatic spraying increased spray deposition by 37%. Zheng et al. (2002) claimed that electrostatic spraying can improve the distribution and deposition of droplets on plants, decrease environmental contamination, reduce spray volumes and process costs and improve treatment effectiveness compared to conventional sprayers.
Nevertheless, Hislop (1988) showed that some electrostatic equipment fails to produce consistent control results because of charges that are insufficient to improve deposition or droplet sizes that are unsuitable for use with electrostatic charges. Bayer et al. (2011) worked with rice plants and showed that electrostatic spraying produced lower drop penetration in the interior of the crop and lower droplet densities compared to other spraying systems. Magno Júnior et al. (2011) showed that electrostatic spraying did not increase deposition on citrus crops.
Therefore, the objective of the current study was to evaluate spray deposition on coffee plants and losses to the soil resulting from hydropneumatic spraying at different volumes and with or without an electrostatic charge.
MATERIAL AND METHODS
The experiment was carried out in January 2014 on an eight-year-old coffee plantation (cultivar Catuai 144, spaced 3.8 x 0.65m) located in the municipality of Rio Paranaiba, MG, Brazil. The leaf area index (LAI) was 5.49 according to the methodology proposed by Favarin et al. (2002). The laboratory studies were conducted in the Federal University of Uberlândia.
Three types of trailer-mounted airblast sprayers were evaluated. The first (Sprayer A) was a Jacto Arbus 2000 Super Export model with a 2000 L tank, 36 nozzles mounted in two lateral arcs, a single air source driving all nozzles (conventional axial fan type sprayer), 150 L min -1 piston pump and an 850 mm axial fan with a capacity of 19 m 3 s -1 . The second (Sprayer B) was a Montana Maozinha Twister 1500 model with 1500 L tank, 32 nozzles mounted on eight adjustable air spouts (directed air jet sprayer), 90 L min -1 membrane pump and a 900 mm fan with a capacity of 13.8 m 3 s -1 . This sprayer has four air spouts on each side. The first and fourth spouts have three nozzles each and the second and third spouts have five nozzles each. The third sprayer (Sprayer C) was an Electrostatic Montana Maozinho model that is similar to sprayer B except that the air spouts are replaced with four SPE electrostatic devices on each side and positioned at 0.35, 1.10, 1.85 and 2.60 m above the ground. The system electrically charges the spray droplets by producing a high voltage (5000 V) electric field at the base of the spray jet as it exits the hollow cone nozzle. The charge is the result of an electric field produced by induction rings connected to a high voltage generator.
The sprayers were pulled by a 5425N John Deere tractor (57.4 kW). Table 1 shows the nozzles used in each treatment. Pressure was adjusted to achieve the desired spray volume. The trial was conducted in randomized blocks. The experimental plots consisted of four rows of coffee, 15 m long; however, only the two central rows were considered.
Nine treatments were performed (2 x 4 + 1) consisting of two sprayer types (Sprayers A and B), four spray volumes (200, 300, 400 and 500 L ha -1 ) and an additional sprayer (Sprayer C) with electrostatic equipment set at 130 L ha -1 (Table 1). Each treatment had four repetitions in which foliage deposition and losses to the soil were studied. The electrostatic sprayer was only tested at 130 L ha -1 because its biggest advantage over traditional equipment is a potential reduction in spray volume. Furthermore, according to the manufacturer, the electrostatic equipment does not perform well at high spray volumes given the difficulty of charging the droplets.
The MAG 1, MAG 2 and SPE 03 hollow cone nozzles used in this experiment are ceramic and angled at 80°. According to the manufacturers, the first two nozzles produce fine droplets and the third nozzle produces very fine droplets (at the pressure levels used in the experiment). The sprayers moved at a constant 2.2 m h -1 (7.8 km h -1 ) with a constant PTO of 540 rpm throughout all treatments.
Brilliant Blue tracer (300 g ha -1 ) was used in all treatments to determine leaf deposition on the upper, middle and lower parts of the canopy as well as spray run-off to the soil. Leaves were removed from plagiotropic branches that were internal and closer to the trunk of the coffee plant (0.20, 1.30 and 2.00 m above the soil). After removal, the leaves were placed in plastic bags and then stored in insulated containers. Ten leaves were collected for each repetition. The fungicide Azoxystrobin+Cyproconazole (750 mL ha -1 ) and paraffinic mineral oil (0.5% v/v) were applied with the tracer.
Two petri dishes (149.51 cm 2 ) were placed under the canopy (0.2 m from the stem) of each replicate to evaluate spray losses to the soil. In the laboratory, 100 mL of distilled water was added to each of the bags containing leaves and 40 mL was added to each petri dish. The resulting solutions were then removed and absorbance readings were made with a spectrophotometer (Biospectro SP-22) set at a wavelength of 630 nm. Leaf area was measured by digitalizing leaves and then analyzing them with the "Image Tool" program (University of Texas, Texas, USA). Absorbance data was transformed into concentration (mg L -1 ) using a calibration curve. Tracer mass was then divided by the foliage or petri dish area from each repetition to obtain deposition in µg cm -2 .
Environmental conditions were measured during the applications. Temperature varied from 23.2 to 29.4°C, relative humidity from 62% to 80% and wind speed from 0.8 to 1.2 m h -1 (3.0 to 4.4 km h -1 ).
Data assumptions were tested first. Homogeneity of variances and normality of residuals were tested by the Levene and Shapiro Wilk tests (SPSS statistical program, version 17.0), respectively. To meet the 0.01 level of significance, leaf deposition values from the lower part of the plants were transformed by the square root of x. The rest of the data was not transformed. The data were then submitted to analysis of variance. The averages were compared to each other by the Tukey test and compared to the additional treatment by the Dunnett test at the 0.05 significance level. Regression analysis was used to study the effect of spray volume; however, the resulting models were not significant To determine the effect of the three sprayers on tracer distribution uniformity, the canopy deposit variances (upper, middle and lower parts) among the three sprayers were tested by an F test (5% probability). For sprayers A and B, the average variance from the four spray volumes from each equipment type was used.
RESULTS AND DISCUSSION
Table 2 shows the spray deposition on the upper part of the coffee canopy and demonstrates that there was no significant relationship or a nondependent relationship between spray volume and sprayer type. There was also no significant interaction between these two factors and the additional treatment. In other words, there was no difference in spray deposition on the upper part of the canopy resulting from Sprayers A and B and the electrostatic sprayer. Note that the electrostatic sprayer used only 130 L ha -1 , while the conventional sprayers used 200 to 500 L ha -1 . Therefore, the electrostatic system produces the same foliar deposition with a lower spray volume and thus increases operational capacity and reduces application costs. Lower spray volumes allow greater treatment areas per tank and consequently reduce down time for refilling the tank. The upper part of the canopy is the farthest from the spray nozzles and is therefore most difficult to reach with the sprayer. Thus, none of three systems was superior in this regard. Not even electrostatic spraying, which creates attraction between the droplets and the target, increased foliar deposition given that the greater distance hinders the attraction between the droplets and the leaves. Increasing the height of the nozzles and thereby reducing the distance could improve deposition on the upper part of the canopy. Ferreira et al. (2013) evaluated spray droplet coverage on coffee with and without an extension arm for tall plants and had difficulty reaching the upper part of the plants, corroborating the findings of the current study.
Table 3 shows average spray deposition on the middle part of the coffee canopy. Sprayer B, with directed air jets, performed better than sprayer A with the conventional arrangement of nozzles positioned along an arc. The use of directed air jets on airblast sprayers is increasing given the demand for greater deposition. This change is improving spray quality and reducing losses (DEVEAU, 2009). The ability to change the direction of the air current and the angle that spray enters the vegetation provides greater uniformity and control over droplet distribution.
Deposition from the electrostatic equipment was better than Sprayer A at all spray volumes and better than Sprayer B at 200 L ha -1 . At higher volumes (300 to 500 L ha -1 ), Sprayer B produced similar deposition to that of the electrostatic equipment, thus demonstrating the potential spray volume reductions afforded by the electrostatic system. Spray volume did not produce significant differences in deposition. Consequently, a model was not fit to correlate deposition and volume. Averages followed by distinct letters in each column differ from each other (Tukey, 0.05). Averages followed by + differ from the additional treatment (Dunnett, 0.05). CV: Coefficient of Variation; MSD test : minimum significant difference for the additional treatment; MSD spr : minimum significant difference for a sprayer; F spr , F sv , F int , F int x test : F values calculated for sprayer factors, spray volume, interaction between factors and interaction between factors and the additional treatment; ns : not significant; *: significant at 0.05.
Table 4 shows the deposition on the lower part of the canopy. Again, Sprayer B, with directed air jets, performed better than Sprayer A. The electrostatic equipment produced better deposition than the two sprayers (A and B), regardless of spray volume. Lower leaves are closer to the point where the spray is emitted and to where the droplets are charged. This results in greater attraction between droplets and leaves, which reduces losses and increases deposition.
As in the present study, other authors have confirmed that in arboreal crops, good spray deposition is easier to achieve when the foliage has greater exposure to the spray nozzles (SCUDELER et al., 2004;RAMOS et al., 2007;FERNANDES et al., 2010;MIRANDA et al., 2012). The nozzles are closer to the lower part of the plant, which justifies the obtained results.
Once again, spray volume did not significantly affect spray deposition. Consequently, a model was not fit to correlate deposition and volume. This shows that lower volumes can be used and greater operational capacities achieved without affecting treatment quality. Applications greater than 500 L ha -1 are common in coffee crops; however, reducing these volumes is viable and yields significant gains in operational capacity.
There are few studies on coffee crop deposition; however, similar results can be gleaned from studies on citrus. Salyani and Farooq (2003), found no significant differences in leaf coverage with spray volumes from 250 to 3950 L ha -1 . In another study, Farooq and Salyani (2002) observed that spray deposition on orange trees was greater at a spray volume of 980 L ha -1 than at 250 L ha -1 . Nevertheless, they found little difference in spray coverage from 980 L ha -1 to 1945 L ha -1 . There were no differences in spray losses to the soil between sprayers A and B (Table 5) and no differences in spray volumes. All treatments without electrostatic charges produced greater losses than the electrostatic treatments. In general, electrostatic spraying allows greater deposition, mainly on the abaxial side of the leaves, which reduces runoff to the soil. In general, it is expected that up to a point, increases in spray volume will increase the spray retained by the leaves. After this point, the leaf surfaces will not be able to retain additional liquid and undesirable runoff will occur. Table 6 compares the variances in the tracer mass retained on the foliage throughout the entire plant. There were no differences between sprayers A and B. This demonstrates that these sprayers did not influence variations in deposition throughout the plant. Relative to the conventional sprayers (A and B), the electrostatic sprayer produced less uniform distribution across the entire plant because of the greater tracer concentration on the lower foliage. This data shows that charged droplets increase deposition on the lower foliage. In cases where it is desirable to increase distribution on the lower parts of the plant, without causing great variability throughout the entire plant, it may be necessary to find alternatives such as positioning the spray nozzles closer to the upper part of the plant.
CONCLUSIONS
Electrostatic spraying produced greater spray deposition on the lower part of the coffee canopy than did non-electrostatic spraying. In this region and in the middle part of the foliage, the sprayer with directed air jets performed better than the sprayer with nozzles positioned in lateral arcs. Deposition on the upper part of the plants was similar among the sprayer types and lower than deposition on the middle and lower parts.
Spray volume in the airblast sprayers without electric charge (200 to 500 L ha -1 ) did not influence spray deposition on the plants and losses to the soil.
The electrostatic sprayer reduced losses to the soil.
Table 2 .
Tracer deposition (µg cm -2 ) on the upper leaves of coffee plants resulting from different sprayer types and spray volumes.
Table 3 .
Tracer deposition (µg cm -2 ) on the middle leaves of coffee plants resulting from different sprayer types and spray volumes.
Table 4 .
Tracer deposition (µg cm -2 ) on the lower leaves of coffee plants resulting from different sprayer types and spray volumes. Averages followed by distinct letters in each column differ from each other (Tukey, 0.05). Averages followed by + differ from the additional treatment (Dunnett, 0.05). CV T : coefficient of variation of transformed data; CV NT : coefficient of variation of data not transformed; MSD Tspr : minimum significant difference for a sprayer (transformed data); F spr . F sv . F int . F int x test : F values calculated for sprayer factors, spray volume, interaction between factors and interaction between factors and the additional treatment; ns : not significant; *: significant at 0.05. Analysis of variance conducted on data transformed by the square root of x.
Table 5 .
Tracer deposition (ƞg cm -2 ) on petri dishes at ground level from different sprayer types and spray volumes. Averages followed by + differ from the additional treatment (Dunnett, 0.05). CV: coefficient of variation; MSD test : minimum significant difference for the additional treatment; F spr . F sv . F int. F int x test: F values calculated for sprayer factors, spray volume, interaction between factors and interaction between factors and the additional treatment; ns : not significant; *: significant at 0.05.
Table 6 .
Variances in the tracer mass retained throughout the coffee foliage after applications with different sprayer types. No significant difference between variances (F test, 5% probability).* Variances differ by the F test at 5% probability. ns
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2019-03-30T13:05:30.857Z", "created": "2015-09-10T00:00:00.000", "id": "85665211", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "http://www.seer.ufu.br/index.php/biosciencejournal/article/download/26876/17112/", "pdf_hash": "12a8a5b89c2b623f4c6c3af6b37392c0cea0f48c", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:39597", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "12a8a5b89c2b623f4c6c3af6b37392c0cea0f48c", "year": 2015}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
PROPAGATION THROUGH CUTTING TECHNIQUE OF SPECIES OCURRING IN THE LOWER SÃO FRANCISCO RIVER IN SERGIPE STATE WITH DIFFERENT CONCENTRATIONS OF INDOLBUTIRIC ACID 1
The objective of this work was to evaluate the feasibility of vegetative propagation through cutting technique of seven tree species with strong occurrence in the riparian forest of the Lower São Francisco River in Sergipe State, under different concentrations of indolbutiric acid at 0, 2500, and 5000 mg. L-1, for potentialization of its use in soil bioengineering technique. It was used a complete random block design with three replicates, and a total of twenty-one treatments. The evaluation period was 120 days for each species, and the data collection was made in intervals of fifteen days, in a total of eight evaluations for each species. The evaluated parameters were: Survival Rate, callus formation, and Root Dry matter Weight. Among the studied species, Schinus terebinthifolius Raddi presented the best results related to cutting technique mainly under the indolbutiric acid concentration of 2500 mg. L-1.
INTRODUCTION
Due to its length and complexity, the São Francisco River basin is divided into four sections: Upper, Middle, Sub-Middle, and Lower São Francisco. These sections are autonomous allowing a decentralized and Basinwide management model (ANA/GEF/PNUMA/ OEA, 2003). With the implementation of the cascade of dams and other public policies, especially in the sub-middle and lower São Francisco, impacts have been identified, such as the removal of riparian vegetation and accelerated erosion of the banks, causes and effects of changes in the dynamics of the river which gradually reduce productive land area (HOLANDA et al., 2005).
Control of bank erosion can be accomplished in several ways, such as the use of rockfill, which, though efficient, is quite expensive, precluding its use extensively throughout the river banks. In an attempt to solve the problem of the riverine population, many empirical solutions have been made, which do not produce the desired effect, cause problems for the recovery of riparian vegetation in addition to degrading the landscape (HOLANDA et al., 2009).
As an alternative to the empirical practices of the riverines and to the expensive bordering and rockfill biotechniques, the use of abundant raw material has been tested through experimental trials in the lower São Francisco through bioengineering techniques, providing a way of mitigating the problem that can be economically viable and has shown technical efficiency (HOLANDA et al., 2008).
Soil bioengineering is a biotechnology that consists in the use of living materials or inert plant substances, biotextiles, associated or not with rocks, concrete, or metals, presenting themselves to be environmentally sustainable for containment of erosion of embankments at various conditions of slope and soil grain size on the banks of water bodies like reservoirs, irrigation canals, and rivers (DURLO; SUTILI, 2005). According to Gray and Sotir (1996), the vegetation component which has great importance in these biotechnologies has been recognized not only for its aesthetic qualities, but also for its beneficial hydromechanical effects and protection against soil erosion. Unlike other technologies in which plants are merely an aesthetic component of design, in soil bioengineering they assume an important ecological, economic, and especially structural contribution (ARAUJO et al., 2005).
Currently, with regards to vegetative propagation, taking live cuttings from a matrix plant to grow new plants is one of the most widespread propagation methods of forest species. Cutting is the aseptic culture of shoot tips in nutrient media, with appropriate concentrations of auxin and cytokinin hormones used to promote the induction of growth, proliferation, and rooting (JONES; HATFIELD, 1976;VAZ;NEGUEROLES, 1979;JAMES;THURBON, 1981;SIMMONDS, 1983;WELANDER, 1983). Cutting presents some advantages such as: fast emergence of a reasonable quantity of seedlings from a single plant matrix; simple technique; absence of common problems found in to other processes of vegetative propagation (eg.incompatibility between grafts and rootstock), and low cost (MÉSEN et al., 1997;LANA et al., 2008;ALCANTARA et al., 2008).
The application of phytoregulators on species with rooting difficulty can compensate for low levels or even lack of endogenous auxin (WAGNER et al., 1989;PIO et al., 2004). Auxin induces rooting in many species thus providing a higher rate and speed of root formation, and greater quality, uniformity and survival of the root system.(BARBOSA et al., 1985;DUNN et al., 1996;TONIETTO et al., 1997;DIAS et al., 1999).
Little is known about cutting as a technique for vegetative propagation in application with native forest species, both at the experimental and commercial level (XAVIER et al., 2003). Initial results indicate the technical feasibility of vegetative propagation to produce seedlings of Schinus terebinthifolius (BAGGIO et al., 1989) and Mimosa caesalpiniaefolia Benth (LINS et al., 2006). However there is incipient information about the development of forest species regenerated through this form of propagation.
The objective of this work was to evaluate the feasibility of vegetative propagation through cutting technique of seven woody species with strong occurrence in the riparian forest of the Lower São Francisco River in Sergipe State, under different concentrations of Indolbutiric Acid at 0, 2500, and 5000 mg. L -1 , in order to be used as part of the soil bioengineering technique. For the collection of vegetative material, three vigorous trees from each species were selected based on their healthy appearance and built. The material was taken from the basal portion of the individuals, basal cuttings being prepared without leaves, which were taken in the preparation and with lengths between 15 and 20 cm and a diameter between 0.5 and 1.5 cm. After collection, the cuttings were placed in a bucket of water in order not to lose moisture, and then all cuttings were straight cut at the apex and at an angle at the base.
MATERIAL AND METHODS
Before planting, the cuttings were immersed in sodium hypochlorite 1%, remaining immersed for one minute, in a black plastic bag recipient of 750 ml. The planting was done on the same day as the collection. Irrigation was done twice daily, in the early morning and late afternoon. The substrate used was black soil, washed sand and coconut coir (2:1:1), which allows for greater moisture retention, nutrient availability, good aeration, allowing for future removal of roots. The base of the cuttings was immersed for 15 seconds at different concentrations of IBA (indolbutiric acid) phytoregulator: 0, 2500, and 5000 mg. L -1 . There were 30 cuttings in each plot, of which twenty-four were evaluated. Six additional cuttings were kept to replace eventual losses.
The evaluation period lasted for 100 days for each species. Data collection was carried out fortnightly, totaling eight assessments for each species. The evaluated parameters, which considered the total number of cuttings (live and dead), were: Survival Rate, Rooting Rate, Number of Roots Formed, Number of Calli Formed, and Root Dry Weight.
The Survival Rate is the number of live cuttings in relation to the total number of cuttings per treatment, given in percentage; Rooting Rate is the percentage number of cuttings that rooted in relation to total cuttings; Number of Roots Formed is determined by counting the number of roots that a cutting produce; Root Dry Weight is determined by collecting and drying the roots in an oven at 60ºC for 24 hours, and weighed on a balance accurate to 0.01g; Number of Calli Formed is determined by counting the Number of Calli that a cutting produced.
The experimental design was completely randomized with three replicates, totaling twenty-one treatments. Treatment effects and their interactions were tested using analysis of variance and significant differences between the means were carried out by using the Tukey test at 95% probability. Cuttings were selected for destructive evaluations, as was the randomized arrangement of treatments.
Survival Rate and Rooting Rate
Cuttings of S. terebinthifolius were the only ones to show significant reduced percentage at higher concentrations of IBA, suggesting phytotoxicity between the concentrations of 2500 and 5000 mg. L -1 (Table 1).
The total percentage of rooted cuttings, treated with IBA, varied from 0 to 66.8% (Table 2), showing a wide range, with a development less than desirable for the studied species, since rooting is closely linked to the survival rate. Comparing the concentrations of 2500 and 5000 mg. L -1 , there was significant rooting difference for the cuttings of S. terebinthifolius. This species, in the absence or presence of the hormone, showed different responses in relation to the others, at p < 0.05, presenting root emissions from the first evaluation. With the application of the hormone, a higher level of rooting was observed, with significant differences between the doses and a clear advantage at the concentration of 2500 mg. L -1 , which showed itself to be the most recommended for this parameter.
Number of Roots Formed and Number of Calli Formed
The most significant results for this parameter, as well as also for rooting, were shown by cuttings of S. terebinthifolius , showing that the application of indolbutiric acid (IBA) had a positive effect on the formation and root number (Table 3). The number of roots formed for cutting of S. terebinthifolius was significantly higher at a dose of 2500 mg. L -1 IBA (p < 0.05) to that observed in the other species.
S. terebinthifolius also presented better development regarding to the Number of Calli Formed, when compared with the other species, in all the treatments, reaching its greatest level at the dose of 2500 mg. L -1 (
Tabela 3 -Número de raízes formadas das espécies vegetais submetidas a diferentes concentrações de ácido indolbutírico (AIB).
*Means followed by the same lower case letters in a column and capital letters on the lines do not differ significantly to the level of 5% probability. CV = coefficient of variation.
In the absence of IBA, rooting of S. lutea and M. caesalpiniaefolia were similar for both, and very different when compared to the S. terebinthifolius which presented an increase of 12%.
Root Dry Weight
Root Dry Weight is associated to Rooting Rate and Number of Roots Formed. By determining the weight of dried roots, it is possible to estimate the Revista Árvore, Viçosa-MG, v.36, n.1, p.75-82, 2012 quality of the root system produced by the cutting. S. terebinthifolius was the species that had the highest mean dry weight (Table 5), as a result of higher rates of rooting and the greater number of roots produced by cuttings (Table 3). These results are in agreement with the results by Nicoloso et al. (1999) for Platanus acerifolia Ait., who observed that larger diameter cuttings presented a higher number of roots per cutting, therefore, presenting higher root dry matter.
DISCUSSION
When assessing the behavior of species in the absence of IBA (control), it was possible to observe the actual genetic potential of rooting. M. caesalpiniaefolia that showed less dependency on the use of this phytohormone when compared to other species, as also observed by Inoue and Putton (2007), who found that the behavior of the Survival Rate of cuttings of a species varies with the concentration of a plant growth regulator.
The use of different concentrations of IBA in the cuttings of the species S. lutea and M. caesalpiniaefolia did not promote a good callus formation, with no significant difference among the data. Fachinello et al. (2005) reported that callus formation is observed as a result of a trauma during the preparation of the cutting, however Hartman et al. (1997) asserts that rooting-inhibiting substances contained in adult plants may be the reason for no callus formation. Phytotoxicity by growth regulators is generally due to the fact that auxins, besides being promoters of rooting, also induce hormonal disruption of cuttings in higher concentrations, inhibiting rooting (FRANZON et al., 2004;HERRERA et al., 2004;FOCHESATO et al., 2006).
The species S. lutea, T. guianensis, L. sericeus, M. caesalpiniaefolia, and G. americana.)showed differences in Survival Rate, however, the use of hormone in the tested concentrations does not seem to be recommended, since there was no significant differences between treatments (p e" 0.05).
In the comparison of doses of IBA in each species S. terebinthifolius at the dose of 2500 mg. L -1 resulted in an increased level of rooting, but at the maximum dose of IBA a reduction of approximately 37% was observed when compared with the non-application of this hormone. Lima Filho and Santos (2009) emphasize that characteristics of the species and between species, such as vigor, rooting capacity, and survival under Tabela 5 -Peso de matéria seca da raiz de espécies vegetais submetidas a diferentes concentrações de AIB (ácido indolbutírico).
*Means followed by the lower case letters in a column and capital letters on the lines do not differ significantly to the level of 5% probability. CV = coefficient of variation.
Revista Árvore, Viçosa-MG, v.36, n.1, p.75-82, 2012 adverse conditions, are determinants for the high values of roots numbers. There was a greater level of rooting in absolute values for M. caesalpiniaefolia than in S. lutea species, but these were not significant at p < 0.05. Souza and Lima (2005) demonstrated that the application of IBA in cuttings of S. lutea tree reveals no significant effects on rooting.
The cuttings of the species I. marginata, T. guianensis, L. sericeus and G. americana showed no root formation under any of the treatments, leading to the dismissal of these cuttings from further evaluation. It is possible that the lack of roots in these species may have occurred due to the type of cutting used, since they were taken from the basal portion of the stem. The works of Couvillon (1988), Alvarenga andCarvalho (1983), andHartmann andKester (1990) conclude that the degree of lignification increases from the apex to the base of the stems, where the tissues present a higher degree of differentiation, hindering the resumption of the meristematic condition, essential for root initiation.
The use of different concentrations of IBA in the cuttings of the species S. lutea and M. caesalpiniaefolia did not promote a good callus formation, both showing a similar development, with no significant difference between the data. In agreement with Fachinello et al. (2005), callus formation is observed as a result of a trauma during the preparation of the cutting, however, the rooting-inhibiting substances contained in adult plants may be the reason for no callus formation (HARTMANN et al., 1997).
When comparing the variation of the hormonal doses in each species a greater rooting was verified in the presence of IBA, with a better performance of the S. terebinthifolius statistically significant at the 5% level, in disagreement with the rooting of the species S. lutea and M. caesalpiniaefolia. These results are in agreement with the observations by Nicoloso et al. (1999) who working with Platanus acerifolia Ait.verified that the cuttings of larger diameter presented a higher number of roots per cutting, than those of smaller diameter, therefore, presenting higher root dry matter.
The similarity in the absence of IBA to the dose of 5000 mg. L -1 , showed no significant results for the species M. caesalpiniaefolia and S. lutea in the number of roots formed, probably due to the interaction between IBA and substrate that influences the moisture content at the base of the cutting as mentioned by Lima et al.(2007). The absorption of water by the cutting is directly related to the degree of contact between it and the water film around the particles in the substrate, with higher absorption occurring with greater volume of water retained by the substrate (DAVIS et al., 1986).
Considering that the capacity of a cutting to emit roots is a function of endogenous factors and environmental conditions, maybe the pH of the substrate (6.2) may have positively influenced the production of roots of S. terebinthifolius . Depending on the plant species, pH between 5.5 and 6.5 is considered ideal for rooting cuttings (FACHINELLO et al., 1994). Coconut coir may also have contributed to improving the porosity of the substrate, allowing better gas exchange with the environment, and, consequently, contributing for a better performance of the cuttings of the species S. terebinthifolius. S. terebinthifolius also developed well and had greater values of Root Dry Weight at a 2500 mg. L -1 IBA concentration (Table 5), without significant difference between the control and the highest tested concentration. S. terebinthifolius and M. caesalpiniaefolia showed that increasing concentrations of this growth regulator caused a stimulation effect on the roots up to a certain value, from which, larger increases showed an inhibitory effect.
One factor that may have contributed to the low values of Root Dry Weight in the S. lutea species was the absence of leaves, which were taken in the preparation, as mostly recommended. Nevertheless, Bacarin et al. (1994) concluded that the greater dry weight material of roots from cuttings of Guava tree (Psidium guajava, L) is due to the increased use of photosynthetic material stored in the leaves, providing the survival and development of the cuttings.
CONCLUSIONS
The application of IBA allowed a direct relationship between Survival Rate and Rooting Rate, although the development of the species for the evaluated parameters presented differences on the concentrations of the hormone applied to the cutting. The species S. terebinthifolius showed the best results with the application of the hormone, especially at the concentration of 2500 mg. L -1 . The species M. caesalpiniaefolia
Table 1 -
Survival rate of plant species exposed to different concentrations of indolbutiric acid (IBA). Tabela 1 -Taxa de sobrevivência das espécies vegetais submetidas a diferentes concentrações de ácido indolbutírico (AIB). Means followed by the same lower case letters in a column and capital letters on the lines do not differ significantly by the Tukey test (p < 0.05). C. V. = coefficient of variation. *
Table 2 -
Rooting rate of plant species exposed to different concentrations of indolbutiric acid (IBA).
Tabela 2 -Taxa de enraizamento das espécies vegetais submetidas a diferentes concentrações de ácido indolbutírico (AIB).*Meansfollowed by the same lower case letters in a column and capital letters on the lines do not differ significantly by the Tukey test (p < 0.05). CV = coefficient of variation.
Table 3 -
Number of roots formed for the plant species exposed to different concentrations of indolbutiric acid (IBA).
Table 4 -
Number of Calli formed in plant species exposed to different concentrations of indolbutiric acid (IBA). Tabela 4 -Número de calos formados nas espécies vegetais submetidas a diferentes concentrações de ácido indolbutírico (AIB). Means followed by the same lower case letters in a column and capital letters on the lines do not differ significantly to the level of 5% probability. CV = coefficient of variation. *
Table 5 -
Root dry matter weight for plant species exposed to different concentrations of indolbutiric acid (IBA).
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2017-09-17T12:46:19.494Z", "created": "2012-02-01T00:00:00.000", "id": "29916202", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://www.scielo.br/j/rarv/a/rP8MZdLkbcLTHdp5Qrf7jrG/?format=pdf&lang=en", "pdf_hash": "95b16eb18a2d13d92d6443c476f670afdd363cd0", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:31145", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "95b16eb18a2d13d92d6443c476f670afdd363cd0", "year": 2012}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Genetic Analysis Studies in Potato (Solanum tuberosum L.) Genotypes for Tuber Yield and Yield Related Traits
Potato is a high potential food security crop in the world including Ethiopia. Genetic variability is the basis of all crop improvement programs. The study was conducted at Adet in 2018 with the objective of assessing the extent and pattern of genetic variability of potato genotypes for yield and yield related traits. A total of 36 potato genotypes were evaluated for 18 quantitative traits in simple lattice design with two replication. The analysis of variance revealed that highly significant (p ≤ 0.001) difference among the tested potato genotypes for all quantitative traits except average stem number per hill. The phenotypic coefficient of variation was ranged from 4.56 to 56.01% (specific gravity and unmarketable tuber yield t ha -1 respectively) and the genotypic coefficient of variation was ranged between 2.32 to 40.66% (specific gravity and late blight severity percentage respectively). The broad sense heritability was ranged from 25.93 to 97.05% (specific gravity and late blight severity percentage respectively) and the genetic advance as percent of mean was ranged from 2.44 to 82.64% (specific gravity and late blight severity percentage respectively). Days to attain 50% emergence, leaf area index, number of marketable tubers and total tubers per plant, marketable and total tuber yield t ha -1 and late blight severity percentage had high heritability with high genetic advance as percent of mean. Most of the traits had high phenotypic coefficient of variation and genotypic coefficient of variation; and coupled high heritability with high genetic advance as percent of mean. Traits having high heritability and high genetic advance as percent of means was effective for simple selection.
Introduction
The crop Potato (Solanum tuberosum L.) is one of the most important food crops worldwide including Ethiopia. It ranks third after rice and wheat in terms of human consumption [1]. According to FAO [2] report the total world potato production was 370,436,581 metric tons. China was by far the largest potato producer, accounting for 24.8% of world production [2]. In Ethiopia, during 2019/20 growing season more than 1 million small holders are engaged in potato production. The total area allocated for potato has reached 70,362.22 ha, total production of 924,728.361 tons produced [3]. Ethiopia ranked in 11 th in Africa and it covers 0.25% of the total world potato production [2]. Currently, Potato is a high potential food security crop in Ethiopia due to its high yield potential, nutritional quality, short growing period and wider adaptability [4]. On the other hand, the productivity of this crop in the country is very low (13.14 t ha -1) as compared to the world's average yield of 20.36 tons ha -1 [2,3]. The lower yield is attributed to many biotic and abiotic factors, such as poor agronomic practices, lack of high-quality and improved planting material, high cost of improved seed tubers, disease and pest problems [5,6].
The use of local tuber seed and varieties with low genetic variability are the major constraints of low yield in potato. Breeders should take the challenge to provide food at cheaper rate to the millions of hungry people in developing countries by increasing the production of potato per unit area and per unit time. To initiate any breeding program to this direction, presence of enough genetic variability in the population for yield and yield related traits should be considered as pre requisite element. Moreover, application of perfect breeding method is dependent on estimation of genetic gain of the characters for successful selection as to develop desirable traits suggested by Johnson et al. [7].
Plant breeding or crop improvement depends upon the magnitude of genetic variability and extent to which the desirable characters are heritable. Genetic variability is the basis of all crop improvement programs. Sufficient genetic variability, if present, can be exploited for developing superior cultivar or varieties. The total variability can be partitioned into heritable and non-heritable components with the help of genetic parameters like genotypic and phenotypic coefficients of variation, heritability and genetic advance. Parameters of genotypic and phenotypic coefficients of variation (GCV and PCV) are useful in detecting the amount of variability present in the available genotypes. Heritability and genetic advance help in determining the influence of environment expression of the characters and the extent to which improvement is possible after selection [8]. High heritability alone is not enough to make efficient selection in segregating generation, unless the information is accompanied for substantial amount of genetic advance [7]. Rahman [9] also reported that knowledge on the nature of variability and association of yield with its components is of great impotence for identification of superior parents in any breeding program.
In Ethiopia, potato breeding method depends on conventional breeding method such as introduction of potato germplams from International Potato Center (CIP) every year. Thus indicated that introduced potato genotypes needs to be characterized and evaluated, because quantitative traits are strongly influenced by environmental factors. Knowing the nature of genetic variability and diversity of genotypes is essential to use as a base material for further breeding program and to meet the diversified goals of plant breeding such as for increasing tuber yield, wider adaptation, desirable quality, pest and disease resistance. Therefore, the objective of the present study is to assess the extent and pattern of genetic variability of potato genotypes for agronomic, yield and tuber quality traits.
Description of the Study Area
The experiment was conducted at Adet Agricultural Research Center's experimental station in Northwestern Ethiopia. It is nearly 450 km away from Addis Ababa and 42 km from the Capital City of Amhara Regional State Bahir Dar. Geographically, it is located at 11°16'N latitude and 37°29'E longitude at an altitude of 2240 meter above sea level. The mean annual rain fall is 869 mm and the mean annual temperature is 18.56°C [10]. The soil type of the study area is Nitosol soil.
Experimental Design, Treatments and Procedures
A total of 36 potato genotypes consisting of 33 advanced genotypes were introduced from International Potato Center (CIP) and three recently nationally released potato varieties as standard checks were used ( Table 1). All of the 36 genotypes were planted at Adet Agricultural Research Center on station during the main rainy cropping season in 2018. The genotypes arranged in simple lattice design with two replications and each gross plot were 3 m x 3 m = 9 m 2 consisting of four rows, which accommodated 10 plants per row and thus 40 plants per plot. The net plot size is 1.5 m x 2.4 m=3.6 m 2 . The spacing between rows and plants were 0.75 m and 0.30 m, respectively. The spacing between plots and adjacent replications were 1 m and 1.5 m, respectively. The experimental field was cultivated to a depth of 25-30 cm by a tractor and ridges were made manually after leveling. Fertilizer application was made as per the specific recommendation for the location, in which NPS as a source of phosphorus was applied at a rate of 180 kg /ha and Urea as a source of nitrogen was applied at rate of 117 kg/ha. NPS was applied once during planting in the rows, while urea was applied in split application half at emergence and half at 50% flowering as a side dress application [11]. All other agronomic practices such as weeding, cultivation and spraying Redomil chemical were kept uniform for all treatments in each plot. The two middle rows were used for data collection.
All tested potato genotypes was introduced from CIP (International Potato Center) & the released varieties were from Adet Agricultural Research Center.
Phonological, Growth, Tuber Yield and Yield Related
Traits Was Collected as Follows Days to 50% emergence: -the numbers of days from planting to the emergence of 50% of plants in each plot was recorded.
Days to 50% flowering: -was recorded as actual number of days taken from emergence to the days at which 50% of the plants in each plot produced flowers.
Days to maturity: -was recorded by counting days from emergence to days on which more than 90% of the plant in each plot get yellow.
Plant height in cm: -The height of five plants in each plot was measured in centimeter from the ground surface to the tip of the main stem and averaged to get the mean plant height.
Number of stem per plant: -It was recorded as the average stem count of five hills or plant per plot at 50% flowering. Only stems that were emerged independently above the soil as single stems were considered as main stems.
Leaf area index (LAI):-To determine leaf area index, five plants (hills) were used from each plot. Individual leaf area of the potato plants was estimated from individual leaf length by using the formula developed by Firman et al. [12] and leaf area index were determined by dividing the total leaf area of a plant by the ground area covered by a plant.
Number of marketable tubers per plant: -Number of tubers harvested from five plants (hills) which counted as marketable after sorting tubers which have greater or equal to 20 g weight, free from disease and insect attack. The average number of marketable tubers were counted and registered.
Number of unmarketable tubers per plant: -The tubers that are sorted as diseased, insect attacked and small-sized (< 20 g) from five plants as indicated in the above were recorded as unmarketable tuber number. The average number of unmarketable tubers were counted and registered.
Total tuber number per hill: -the total number of tubers produced per plant was recorded or it was recorded by the sum of both marketable and unmarketable tubers number per plant.
Average tuber weight (g tuber -1 ):-It was determined by dividing the total fresh tuber weight to the respective total tubers number which was harvested from five plants (hills).
Marketable tuber yield (t ha -1 ):-The total tuber weight which were free from diseases, insect pests, and greater than or equal to 20 g in weight determined from the net plot area and were converted to tons per hectare.
Unmarketable tuber yield (t ha-1 ):-was determined by weighting tubers that were sorted out as diseased, insect attack and small-sized (< 20 g) from the net plot area and converted to tons per hectare.
Total tuber yield (t ha -1 ):-This was determined as the sum of the weights of marketable and unmarketable tubers from the net plot area and converted to tons per hectare.
Tuber Quality Attributes Was Calculated as Follows
Tuber dry matter content (TDMC) (%):-Five fresh tubers were randomly taken from each plot, washed, weighed and sliced at harvest, dried for seven days under sun and finally in oven at 75°C for 72 hours until a constant weight attained and dry matter percent calculated according to William et al. [13] formula.
Specific gravity of tubers (SG):-was determined by the weight in air and in water method. Five kg tuber of all shapes and sizes were randomly taken from each plot. The tubers were washed with water. Then after the sample were first weighed in air and then re-weighed suspended in water. Specific gravity was calculated according to Kleinkopf et al. [14] formula.
The percentage of starch was calculated from the specific gravity, a formula developed by Talburt et al. [15].
Starch (%) =17.546 + 199.07 × (SG-1.0988). Specific gravity (SG) was determined as indicated above by the weight in air and weight in water method.
Total soluble solids (°Brix):-The Brix of the raw potato samples was determined using a method as described by Pardo et al. [16] using hand refractometer. The Brix was measured in the juice obtained after washing, crushing and extracting juice of the tuber samples.
Disease Data
Assessment of severity of late blight under field conditions in percent was recorded on a plot basis taking into account the number of plants developing disease symptoms in a leaf and/or many leaves and plants free from disease following the procedures of Heinfnings [17].
Analysis of Variance
The collected data were subjected to analysis of variance (ANOVA) for Simple Lattice by SAS (Statistical Analysis Software) version (9.0). Duncan Multiple Range Test (DMRT) was used to compare means at 5% and 1% level of significance.
Phenotypic and Genotypic Variances
The phenotypic and genotypic variability of each quantitative trait was estimated as genotypic and phenotypic variance components and coefficient of variation. The phenotypic and genotypic variances were estimated according to the method suggested by Singh et al. [18] as follows: Genotypic variance (σ 2 g) =
& %&
Where: σ 2 g = genotypic variance, MSg = mean square due to genotype, MSe = environmental variance (error mean Genotypes for Tuber Yield and Yield Related Traits square) and r = number of replications.
Coefficient of variation at phenotypic, genotypic and environmental levels was estimated by using the formula, adopted by Burton et al. [19] as follows: Where: PCV= Phenotypic coefficient of variation, GCV= Genotypic coefficient of variation, x̄ = population mean of the character being evaluated. PCV and GCV values were categorized as low (0-10%), moderate (10-20%), and high (>20%) as suggested by Sivasubramanian et al. [20].
Broad Sense Heritability (H 2 b)
Broad sense heritability was estimated based on the formula given by Allard and Falconer et al. [21,22] as follows:
Estimation of Genetic Advance and Genetic Advance as Percent of Mean
Genetic advance and genetic advance as percent of means were estimated as described by Allard [21] and Johnson et al. [7] as follows: Genetic Advance (GA) = K σ p H 2 b Where: K= the standardized selection differential at 5% (2.063), σp = phenotypic standard deviation and, Where: GA= genetic advance, and x̄ = mean of population. The GA as percent of mean was categorized as low (0-10%), moderate (10-20%) and high (>20%) as suggested by Johnson et al. [7].
Analysis of Variance
The result of Analysis of variance showed that there is highly significant (p ≤ 0.001) difference among the tested potato genotypes for all traits except average stem number per hill/plant ( Table 2). The findings on variance for tuber yield and its components indicates the existence of substantial amount of variability for most of the traits in experimental material studied. This provides an opportunity for a breeder to select best genotypes for their better tuber yield and other yield related traits. Many authors also reported the existence of significant variation among potato genotypes for different traits. Addisu Fekadu [24] reported that, highly significant difference among potato genotypes with respect to days to emergence, days to flowering, days to maturity, plant height, number of stem per plant, tuber number per plant and tuber yield (Kg) per plant. Similarly, highly significant difference for plant height, leaf area index, average tuber number per plant, average tuber weight (g/tuber), dry matter content (%) and total tuber yield (t/ha) was reported by Rahman et al. [9,25,26].
Estimates of Variance Components
The variability components (genotypic and phenotypic variance and coefficient of variations, heritability in broad sense and genetic advance as percent of mean) were estimated for seventeen traits and results are presented below in Table 3. However, the results excluded the one trait (stem number per hill/ plant) because of absence of significant difference at both 1% and 5% level of significant.
Phenotypic and Genotypic Coefficient of Variation
The result of analysis of phenotypic coefficient of variation (PCV) was relatively greater than the genotypic coefficient of variation (GCV) for all traits. It is due to presence of substantial influence of environmental factors besides the genetic variation for expression of these traits. The Phenotypic coefficient of variation (PCV) and genotypic coefficient of variation (GCV) values can be categorized as low (<10%), moderate (10-20%), and high (>20%) by Sivasubramanian, S. et al. [20]. Based on these categories most of the traits such as marketable tuber number per hill (36.19 and 30.90%), un marketable tuber number per hill (44.57 and 26.51%), total tuber number per hill (24.59 and 20.87%), marketable tuber yield t ha -1 (34.84 and 32.59%), un marketable tuber yield t ha -1 (56.01 and 37.81%), total tuber yield t ha -1 (32.26 and 30.40%), starch content percentage (31.83 and 23.44%) and late blight severity percentage (41.28 and 40.66%) had high PCV and GCV (>20%), respectively. However, days to attain 50% flowering (5.28 and 4.66%), days to maturity (5.38 and 5.18%) and specific gravity (4.56 and 2.32%) showed low PCV and GCV (<10%) respectively. Moderate PCV and GCV (10-20%) was observed in days to attain 50% emergence (16.80 to 16.28), plant height in cm (13.16 to 10.96%), leaf area index (19.81 to 17.13%) and total soluble solid (19.26 to 13.26). The highest magnitude of PCV and GCV was observed for unmarketable tuber yield per hectare (56.01 and 37.81%) while the lowest PCV and GCV was observed in specific gravity (4.56 and 2.32%) respectively. Average tuber weight (g/tuber) had high PCV (25.56%) and moderate GCV (18.97%). Moderate PCV (14.6%) and low GVC (8.64%) was observed in dry matter content percentage (Table 3). In agreement with this result, high PCV and GCV for tuber yield per plant, hectare, number of tuber per plant has been reported by Rahman et al. [9,27,28].
Estimate of Broad Sense Heritability and Genetic
Advance The estimated broad sense heritability and genetic advance for 17 quantitative traits was presented in Table 3. The minimum and maximum heritability and genetic advance value ranged from 25.93 to 97.05 and 0.03 to 49.24 for late blight severity percentage and specific gravity, respectively. The heritability was categorized as low (0 -40%), medium (40 -59%), moderately high (60-79%) and very high (> 80%) as suggested by [23]. Based on these categories traits such as days to attain 50% emergence, days to maturity, marketable tuber yield t ha -1 , total tuber yield t ha -1 and late blight severity percentage showed very high heritability (>80). The highest heritability was recorded (97.05%) for late blight severity percentage followed by days to attain 50% emergence (93.96%) and days to maturity (92.57%) ( Table 3) In accordance with this result, high heritability for marketable tuber yield, total tuber yield has been reported by Rahman et al. [9,27,28,30]. Similarly, higher heritability for late blight severity percentage (91.02%) was reported by Mohammed W. [29].
The genetic advance as percent mean (GAM) was categorized as low (0 -10%), moderate (10 -20%) and high (>20%) as suggested by Johnson et al. [7]. Accordingly, most of the traits showed high GAM (>20%). However, days to maturity (10.28%) and dry matter content (10.71%), total soluble solid (18.82%) and plant height (18.83) had moderate GA (10-20%). High genetic advance was obtained from late blight percentage (82.64%) and low genetic advance was obtained in specific gravity (2.44%) followed by days to attain 50% flowering (8.48%) ( Table 3). In agreement with this result, the highest GAM was recorded for marketable tuber yield and total tuber yield has been reported by Rahman et al. [9,28]. The higher genetic advance as percent of mean for late blight intensity and severity percentage (96.31 and 85.63 respectively) was reported by Mohammed [29]. Medium GAM for plant height, dry matter content percentage was reported by Rahman [9]. Most of the traits coupled medium to very high heritability with high genetic advance except specific gravity and dry matter content percentage coupled with low heritability and low to medium genetic advance respectively (Table 3). Traits with high heritability couple with high GAM indicated additive gene action for the expression these traits and effective for simple selection while traits with low heritability couple with low GAM indicated non-additive gene action for the expression of these traits. According to Panigrahi et al. [30] report high heritability coupled with high genetic advance in total tuber yield and marketable tuber yield was found indicating the influence of additive gene effect on these characters. High heritability coupled with high genetic advance for marketable tuber yield and total tuber yield was reported by Rahman et al. [9,27,28,30]. Similarly, high heritability coupled with high GAM for leaf area index, number of tuber per plant was reported by Rahman et al. [9,27].
Conclusions
The tested potato genotypes in the current study area showed statistically high significant difference at (P ≤0.001) level of significance revealing presence of substantial amount of genetic variability. It confirms a positive response for the effectiveness of selection based on the traits with high and medium PCV and GCV values for trait of interest improvement. Most of the traits had high PCV and GCV; and coupled high heritability with high GAM. Traits with high heritability coupled with high genetic advance as percent of mean is also important for simple selection.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-12-09T17:54:31.775Z", "created": "2021-10-29T00:00:00.000", "id": "244987605", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GOLD", "oa_url": "https://article.sciencepublishinggroup.com/pdf/10.11648.j.aff.20211005.15.pdf", "pdf_hash": "129f981d2be5a4f720eec0c21f5a0db7895e474d", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:31736", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "05546fcc59d730d178653378c72174ee89e87f33", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Occurrence of Cymbidium Mosaic Virus in Dendrobium Orchids in Kerala and Its Management through Meristem Culture
Orchids are infected by many plant viruses among which Cymbidium mosaic virus (CymMV) is the most prevalent virus. In Kerala, the virus was detected in almost all orchid growing nurseries with varying severity. Commonly observed symptoms of CymMV included irregular mottling and mosaic patterns on leaves. The presence of this virus was confirmed by using DAC-ELISA using specific antibody of CymMV. It was confirmed by TEM analysis to show flexuous rod shaped particles of 4001000nm, which is the characteristic feature of Potex viruses in general. The plants were found to acquire disease through mechanical means as the virus is highly mechanically transmissible. Production of virus free plants using meristem culture was followed as other treatment technique fail to cure the disease effectively and without involving high cost.
INTRODUCTION
Orchid is one of the most important ornamental flowers in the world. It is popular around the world for the aesthetic value of its inflorescence, with wide variety of colours and long vase life. In India, orchids are grown for gardens and cut flower industry. Orchids have been reported to be infected by more than 50 different viruses (Chang et al., 2005;& Zettler et al., 1990). Among the different orchid infecting viruses Cymbidium mosaic virus (CymMV) (Genus: Potexvirus, family: Flexiviridae) and Odontoglossum ringspot tobamovirus (ORSV) (Genus: Tobamovirus, family: Virgaviridae) have been reported as the most prevalent and economically important worldwide (Zettler et al., 1990;Wong et al., 1994;& Sherpa et al., 2004). CymMV and ORSV are estimated to have coinfected about 14% of cultivated orchids worldwide (Wong et al., 1994). Wey et al. (2001) found that the wild species and hybrid plants of Phalaenopsis, Doritaenopsis and Doritis collected from Taiwan, Japan, UK and USA were found to be infected with CymMV and/or ORSV.
Kerala has the optimum climatic conditions for growth of orchids and production of good quality cut flowers. This has attracted many growers to turn to cultivation of orchids for cut flower production. Not much virus infections are reported from the native orchids. But most virus infections are detected in imported orchid plants, which may often escape undetected during multiplication and mass production. This is posing serious threat to cut flower industry as the quality of the flower is directly affected.
Most of the growers are not aware of this possible threat as viral diseases in orchids were not of much significance until recently. Moreover it is often misunderstood for physiological or nutritional disorders. The viral diseases of orchids are of quarantine importance as it can cause severe damage to the cut flower industry and to the diversity of indigenous varieties. Not much study has been conducted in this area for identification and effective prevention and management of this disease, hence this study has been proposed.
The CymMV is a stable virus RNA virus belonging to the group of flexuous rodshaped potexviruses which are approximately 475-490 nm in length (Frowd & Tremaine 1977;& Steinhart & Oshiro 1990). The virus maintains high concentration in plant tissues (Hu et al., 1992). The orchid plants readily take up CymMV and ORSV infection when inoculated of through cut/slash/rub techniques, with minor differences in the rate of spread within the plant (Hu et al., 1994).
CymMV is easily transmitted from plant to plant while repotting and harvesting. Proper handling and sanitation during harvesting is critical in preventing the spread of virus. Sodium hypochlorite, skim milk, ethanol, Agriboom and Physan have been used by orchid growers in Hawaii to inactivate the viruses on cutting tools. NaOH solution with less than 10% conc. was found to inactivate CymMV and ORSV without causing any phytotoxicity. The common skimmed milk at 30% concentration inactivated both the viruses in local lesion indicator host plant Chenopodium amaranticolor (Hu et al., 1994).
Sample collection
A survey was conducted in the year 2015-18 among the major orchid growers in five districts of Kerala for the incidence of virus diseases. The overall incidence of virus diseases in orchids were calculated by taking a random sample of 100 plants and finding out the percent of infected plants against the total number of plants surveyed. Plants showing typical symptoms of orchid viruses such as mosaic, streak, ringspot, leaf spots, necrosis etc., were considered.
Per cent disease incidence was calculated as Disease incidence = Number of plants infected X 100 Total number of plants The intensity of disease was scored based on the ratings observed from the plants and vulnerability index (V. I.) calculated using the following formula.
Etiology and epidemiology of the viral disease
The primary source of virus diseases in dendrobium orchids were studied during the survey. The plants kept in greenhouse at College of Agriculture, Vellayani were observed for the presence of insect and non insect vectors. Mechanical sap transmission methods such slash, piercing and using carborundum were tested on dendrobium orchids. The chances of co-infection of more than one virus at a time was studied in dendrobium. The possibility of sap transmission of orchid viruses were studied on varieties/cultivars native and cultivated dendrobium and non-dendrobium orchids. The overall change in virus titer in dendrobium plants due to change in climatic factors were observed to find the correlation between environmental conditions and virus titer in plants. The plants that were maintained in insect proof polyhouse of grower were selected for this purpose. The plants were maintained in plastic pots at a fixed location inside the polyhouse. The third fully opened leaf was taken at three month interval. Virus titer was found out by DAC-ELISA method using specific antibody and reading the absorbance value at 405nm. The corresponding minimum and maximum temperature and RH were noted.
Morphological characterization
Electron microscopy was done to detect the presence of viral particles in leaves of infected Dendrobium by leaf dip preparation. The samples were negatively stained with 2% uranyl acetate (pH-4.5) and then examined under JEM-1011 transmission electron microscope at Plant Advanced research center for virology, Division of Plant Pathology, Indian Agricultural Research Institute, New Delhi. Digital images of the virus particles were captured by Gatan CCD DV 300 W1 camera which was interfaced with the microscope.
Elimination of virus through meristem culture
Meristem culture was carried out for elimination of virus using tissue culture techniques. Chemicals having antiviral properties were also supplemented to the culture media for virus elimination. The plants produced by meristem culture were indexed for CymMV disease. Dendrobium plants that were tested positive for CymMV maintained in the greenhouse were selected as mother plants for tissue culture. Newly emerged shoots (2-3 cm length) were excised for obtaining meristem tissue for tissue culture. The shoots were washed with Tween-20 (0.05%v/v) followed by plain distilled water. Surface sterilization of shoots was done using 1% sodium hypochlorite for 3-5 minutes followed by 70% ethanol for 2 minutes. The tissues were then washed serially in sterile distilled water 2 -3 times and placed on sterile tissue paper disc to absorb remaining water from the surface.
The outer whorls of leaf and adjoining tissue were peeled out using sterile forceps and rest of the tissue were cut off using sterile scalpel until the meristem was observed. The tissue were given a final cut and placed into culture establishment medium using sterile forceps.
Murashige and Skoog (1962), medium was used as basal medium to initiate callus in the culture. Two strengths of MS i.e., full strength and half strength, both supplemented with different concentrations of BAP (6benzylaminopurine) and NAA (Naphthalene acetic acid) were used to initiate callus from the meristem. Shooting medium (Knudson's agar) was supplemented with 1mg/l NAA and 2mg/l BAP supplemented with 10 mg/l activated charcoal and 20g/l mashed banana for shoot regeneration. Rooting of regenerated shoots were done in orchid multiplication medium with 2mg/l IBA and 1mg/l NAA supplemented with 10 mg/l activated charcoal and 20g/l mashed banana. The plantlets were maintained in orchid multiplication medium with 1mg/l IBA and 2mg/l BAP supplemented with 10 mg/l activated charcoal until the plantlets were ready to be hardened. The culture bottles were placed in culture grow rooms at 24±1 o C under cool white light (10000 lux) for 7 hrs. (Table 1).
The plantlets that have attained proper root and shoot growth were planted into seedling trays containing mixture of coirpith : charcoal : broken tiles ( 1: 1: 1 ). The plants were placed in greenhouse at 75 per cent shade and watered twice a day for 3 months until properly hardened. The hardened plantlets were transferred to hanging pots. The plantlets thus produced were observed for typical symptoms such as mosaic, streaks and mottling followed by immunological assay (DAC-ELISA) using CymMV specific antibodies.
RESULTS
The results of survey was conducted from the period of 2015 to 2019 among different orchid growers in five districts of Kerala viz., Thiruvananthapuram, Alappuzha, Kottayam, Ernakulam and Thrissur are listed in Table 2. A total of 9500 plants were surveyed from four districts of Kerala and among them CymMV was found to be widely prevalent in all locations surveyed. All the locations surveyed had disease incidence not less than 5 per cent. The highest disease incidence of 100 per cent was recorded in Thiruvananthapuram district.
The samples from Thiruvananthapuram recorded highest V. I. of 84.62 among 100 plants surveyed. The lowest V. I. of 5.61 and 12.25 was recorded from samples collected from Thrissur district.
CymMV produced varying symptoms in dendrobium plants depending on the age of the plant, nutrition, abiotic factors like temperature, light intensity and available moisture. Symptom expression also varied in different species of orchid. Commonly observed symptoms include; irregular mottling and mosaic patterns on leaves. Plants appear generally weak with less number of leaves. Floral parts are also affected, with smaller sized and reduced number of flowers per peduncle. Flower colour break or necrotic symptoms were not observed in Dendrobium orchids anywhere during the survey. No visible symptoms were observed in stem and roots (Plate 1).
Other symptoms of CymMV in Dendrobium include general yellowing and sunken pits in older leaves and narrow streak and spindle like mosaic pattern in younger leaves. The infection can remain latent in plants and can be carried over through vegetative propagation or during tissue culture. Detection DAC-ELISA carried out on leaf samples showing symptoms of viral infection confirmed the presence of CymMV. In CymMV infected samples, the highest virus titer value of 3.260 was observed on fully opened new leaves of infected dendrobium plants against 0.022 in healthy samples. The least value of 0.211 was observed in tender stem tissue of CymMV infected plants against 0.0240 in healthy. The movement and of CymMV in dendrobium when artificially inoculated was found to be quickest in inoculated leaves and time of infection to reach to root tips were 3 months post inoculation (Table 3).
Etiology and epidemiology of the viral disease
The primary source of CymMV and ORSV infection at every location surveyed during the period of study was the imported planting materials. A major share of import was made from Thailand, followed by Taiwan, Singapore and Malaysia. Secondary spread of CymMV in every location surveyed was through infected tools such as secatures, horticultural knives and blades that were used to prune orchids. None of the growers or importers were found to disinfect the tools after intercultural operations during the period of study. The plants that were vegetatively propagated through suckers and side shoots from infected plants were found to cause secondary spread of CymMV in Trivandrum and Alappuzha. Use of infected mother plants for tissue culture was found to cause spread of CymMV in Ernakulam.
The overall virus titer was found to vary seasonally and the value was found to be higher during cooler months of the year. The virus titer at 405µm was found to be the highest during the November to January (2.835, which was 142.54 times the negative control) where there was lower RH (70 per cent) and lower night temperature (22.1 o C). (Table 4).
Morphological characterization
Transmission electron microscopy analysis of symptomatic leaf samples of dendrobium by leaf dip method showed the presence of CymMV and ORSV. The characteristic flexuous rod shaped particles of CymMV was observed at 80000 -100000x magnification in mosaic affected leaf sample. The particles were single, non enveloped and 400 -1000 nm in length, thus confirming it as CymMV (Plate 2).
Elimination of virus through meristem culture
The meristem tissue of 1 -1.5 mm was used for culturing for disease free plants. The callus initiation medium was supplemented with 3g/l ascorbic acid and 10 mg/ l activated charcoal to reduce tissue darkening due to production of phenolics. The callus development from cultured meristem began after 4 weeks of culturing. Protocorm like bodies were found to emerge from developing callus in 3 weeks after callus initiation (Plate 3) followed by development of leaf primodia in 8 weeks of callus initiation. The callus was then transferred to Knudsons agar for shooting at this stage. The shooting media was supplemented with 10mg/l activated charcoal and 20g/l mashed ripe banana for better growth of developing tissue.
At 4 months after culturing, the shoot development was profuse (2 -3 cm) and the plantlets were separated and 2 -4 plantlets were placed in single culture flask containing orchid multiplication media for rooting.
Proper root development of 1.5 -3 cm was achieved in 7 months after first culturing of meristem. The plantlets at the stage of 2-4 cm shoot length and 2-3 cm root length were deflasked and hardened (Plate 4). The plants hardened for 5 months at 75 per cent shade had 4 leaves and shoot length of 6 cm and root length of 8 cm.
The plants observed during the period of hardening did not produce any externally visible symptoms of virus diseases such as mosaic, streaks, mottling or ringspots. The DAC-ELISA conducted on developing plants during period of observation also gave negative reaction to specific antibodies of CymMV and ORSV. The plants hence produced by meristem culture of CymMV infected mother plants were found to be free from virus infection.
DISCUSSION
A survey conducted during 2015-2019 in Dendrobium sp. among major orchid growers in five districts of Kerala and every location had incidence of CymMV of not less than 5 per cent. The highest disease incidence of CymMV of 100 per cent and V. I. of 84.62 was recorded in Thiruvananthapuram district. Pant et al. (2010), similarly surveyed and collected samples, mostly Cymbidium sp. from many regions under Sikkim and Darjeeling hills during 2007-08 for detection and analysis of orchid viruses. Mixed infection of CymMV and ORSV was detected in most Cymbidium sp. and hybrids, in 20 different orchid nurseries from Sikkim and 3 nurseries from Darjeeling hills. A similar survey was conducted by Bhai et al. (2003), in vanilla growing parts of Kerala and Tamil Nadu and reported the incidence of CymMV in V. planifolia plants.
CymMV produced varying symptoms in dendrobium plants with common symptom like slightly yellowish mosaic patterns, with overall appearance of weak plants with less number of leaves. Necrotic flower buds, reduced flower size and reduced number of flowers per peduncle was observed in case of severe infection. The disease symptoms of CymMV described by Hue et al. (1993), is almost similar to those observed during the survey such as irregular mosaic, chlorotic and sunken patches on leaves and colour break and necrosis on flowers. Sherpa et al. (2004), also describes symptoms of CymMV as chlorotic or necrotic sunken patches on orchid leaves and flowers; infected flowers turned necrotic and exhibited deformation and color breaking.
The presence of CymMV was confirmed with TEM observation of morphological character from leaf dip preparation and flexuous rod shaped monopartite particles of 400-1000nm in length, which was typical to that of most Potex viruses. Pant et al. (2010), also observed flexuous particles of CymMV measuring 450-500 x 13 nm from the species of Arides odoratum, Calanthe sp. Eria sp. Cymbidium sp., D. nobile, Epidendrum sp. and Rhynchostylus retusa. Similar was the result of Lawson (1970), where flexuous rod shaped particles (800 -900µm) of CymMV was observed form infected samples of Cattleya sp. For production of virus free planting material of dendrobium, MS medium was found to be best for callus initiation and development of dendrobium and followed by Knudsons agar during later stages of development. The meristem cultured plantlets after seven months of culturing showed negative results for CymMV when tested using DAC-ELISA. Pradhan et al. (2016), similarly used different strengths of MS medium for production of virus free planting matrial from Cymbidium aliofolium pods. Similarly the hardened plants also tested negative for CymMV after tissue culture, even though the mother plants was infected.
CONCLUSION
The present study focuses on surveying major orchid importers and growers of different parts of Kerala to detect the presence of orchid infecting viruses. Cymbidium mosaic virus was found to be most prevalent orchid virus in all regions of Kerala. The presence of CymMV was confirmed by using ELISA and TEM methods. The virus being systemic in nature and highly mechanically transmissible the best method to produce virus free plant stock was by tissueculture method. Here meristem culture method was employed successfully to produce virus free dendrobium plantlets which again was confirmed by immunological method.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2021-03-29T13:04:13.781Z", "created": "2021-02-28T00:00:00.000", "id": "232390196", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://doi.org/10.18782/2582-2845.7723", "pdf_hash": "77d104de2bcdac90582c6a2a2c126e499d37f039", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:37117", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "77d104de2bcdac90582c6a2a2c126e499d37f039", "year": 2021}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Semen quality, lipid peroxidation, and seminal plasma antioxidant status in horses with different intensities of physical exercise
The aim of this study was to compare markers of semen quality, sperm membrane damage, and the seminal plasma antioxidant activity in warmblood stallions with and without sport workload stress. Four stallions were used for breeding only (control) and four both for breeding and competition in jumping. Semen samples were collected at 14-day intervals (from June to August) from each stallion (5 ejaculates per stallion). Immediately after sperm collection, a conventional examination of the ejaculate was processed. Catalytic activities of enzymes aspartate aminotransferase, alanin aminotransferase, glutathione peroxidase, superoxide dismutase and indicator of lipoperoxidation F2α isoprostanes were measured in samples of seminal plasma. Contrary to basic semen quality indicators, the values of seminal plasma pH, aspartate aminotransferase and alanin aminotransferase were significantly (P < 0.05) impaired in the physically stressed stallions. Also, the level of F2α isoprostanes and the activity of superoxide dismutase were significantly (P < 0.05) increased by stress. The antioxidant activities of superoxide dismutase and glutathion peroxidase increased during the monitored period and reflected changes in F2α isoprostane concentration. We can conclude that even the conventional basic sperm indicators stay within the reference ranges of the biochemical indicators of seminal plasma such as pH or AST/ALT activity may be negatively influenced by sport workload stress. Increased concentrations of F2α isoprostanes indicate that lipoperoxidation can be a mechanism of cell membrane destabilization, which is counteracted by an increase of antioxidant enzyme activities. This is the first report of oxidative stress symptoms in normospermic equine semen in relation to stallion sport workload. F2α isoprostanes, antioxidative enzymes, sperm, stallions Reproductive capability of stallions is considerably influenced by the quality of their sperm. Indicators such as percentage of live sperm, motility and others are regularly tested (Věžník et al. 2004) although they do not necessarily correlate with the fertility of stallions (Magistrini et al. 1996). Thus, basic examination is supplemented by other methods focusing on sperm quality observation. Due to disrupted human male fertility, since the 1990s much attention has been paid to the loss of functionality and integrity of sperm membranes by reactive oxygen species – ROS (Aitken and Baker 2004). Although lowlevel ROS generation appears to be important in the regulation of the physiological function of sperm (Sanocka et al. 1997), an increase in ROS non-regulated by the production of antioxidants causes oxidative stress (Aitken 2006). Such situation occurs in processes that increase the tissue and cell oxygen requirements, e.g. physical exercise (Avellini et al. 1999). Spermatozoa have a higher unsaturated fatty acid and sterol content. Hence, they are more susceptible to oxidative stress due to lipid peroxidation sperm membranes in the presence of ROS (Aziz et al. 2004). F2α isoprostanes, the end-products of lipid peroxidation, is a specific and quantitative marker of oxidative stress (Khosrowbeygi and ACTA VET. BRNO 2013, 82: 031–035; doi:10.2754/avb201382010031 Address for correspondence: MVDr. Helena Härtlová, CSc. Department. of Veterinary Science, Faculty of Agrobiology Food and Natural Resources, Czech University of Life Sciences Prague 165 21 Prague 6, Czech Republic Phone +420 224 382 799 Fax: +420 234 381 841 E-mail [URL]/ Zarghami 2005). Sperm damage through oxidative stress results in increased membrane permeability to enzymes and other substances, and therefore, reduced metabolic activity of sperm (Storey 1997). Changes in the activity of enzymes such as aspartate aminotransferase (AST) or alanine aminotransferase (ALT) in stallion semen plasma are associated with defects of sperm membranes (Colebrander et al. 1992). Enzymatic and non-enzymatic antioxidants play a very important role in prevention of the effects of ROS. Lack of these antioxidants increases vulnerability of tissues and cells to oxygen reactive forms and increases the risk of oxidative stress (Aitken and Baker 2004). The amount of cytosol in spermatozoa is limited, thereby limiting the antioxidant capacity. Antioxidant enzymes in seminal plasma such as glutathione peroxidase (GPx), superoxide dismutase (SOD), glutathione reductase and catalase therefore play a major role in protecting spermatozoa against lipid peroxidation (Baumer and Bal l 2000). The aim of this study was to compare markers of semen quality, sperm membrane damage, and the seminal plasma antioxidant capacity in stallions with different intensities of physical exercise. Materials and Methods Eight fertile warmblood stallions between 6 and 10 years of age with a body weight of 550 ± 30kg were used for the experiment. Their diet was balanced with the requirements of the NRC (2007) and water was available ad libitum. Four stallions were used for breeding only (non-stressed control group), and four horses were used for both breeding as well as standard training and competition in jumping (stressed group). Semen samples were collected with an artificial vagina (Hanover model, Minitüb, Germany) at 14-day intervals (from June to August) from each stallion (5 ejaculates per stallion). Immediately after sperm collection, conventional examination of the ejaculate (volume, colour, motility, concentration and the hypo-osmotic swelling test) was performed in accordance with Věžník et al. (2004) in the laboratory of the Municipal Stud Farm in Tlumačov. After pH detection, the raw sperm was centrifuged at 2000 × g for 10 min at 4 °C. Samples of seminal plasma were stored at – 70 °C. The catalytic activities of ALT (alanin aminotrasferase) and AST (aspartate aminotrasferase) enzymes were measured by commercial assay kits (Randox Laboratories Ltd, UK). The seminal plasma concentration of F2α isoprostanes was determined by enzyme immunoassay using the commercially available 8-isoprostane ACETM Elisa kit (Cayman Chemical Company, Ann Arbor, MI, USA). The activity of glutathione peroxidase (GPx) was determined in accordance with Paglia and Valentin (1967) using commercial kits RS 505 Glutathion Peroxidase Ransel (Randox Laboratories Ltd, UK). The activity of superoxide dismutase (SOD) was measured by colorimetric assay using a commercially available colorimetric method SD 125 Superoxid dismutase Ransod (Randox Laboratories Ltd, UK). Statistical analysis of the data was performed by two-way analysis of variance with treatment and time interactions using the GLM procedure of SAS (SAS Institute Inc. 2003). Differences were considered significant with P < 0.05.
Reproductive capability of stallions is considerably influenced by the quality of their sperm. Indicators such as percentage of live sperm, motility and others are regularly tested (Věžník et al. 2004) although they do not necessarily correlate with the fertility of stallions (Magistrini et al. 1996). Thus, basic examination is supplemented by other methods focusing on sperm quality observation. Due to disrupted human male fertility, since the 1990s much attention has been paid to the loss of functionality and integrity of sperm membranes by reactive oxygen species -ROS (Aitken and Baker 2004). Although lowlevel ROS generation appears to be important in the regulation of the physiological function of sperm (Sanocka et al. 1997), an increase in ROS non-regulated by the production of antioxidants causes oxidative stress (Aitken 2006). Such situation occurs in processes that increase the tissue and cell oxygen requirements, e.g.physical exercise (Avellini et al. 1999). Spermatozoa have a higher unsaturated fatty acid and sterol content. Hence, they are more susceptible to oxidative stress due to lipid peroxidation sperm membranes in the presence of ROS (Aziz et al. 2004). F 2α isoprostanes, the end-products of lipid peroxidation, is a specific and quantitative marker of oxidative stress (Khosrowbeygi and Zarghami 2005). Sperm damage through oxidative stress results in increased membrane permeability to enzymes and other substances, and therefore, reduced metabolic activity of sperm (Storey 1997). Changes in the activity of enzymes such as aspartate aminotransferase (AST) or alanine aminotransferase (ALT) in stallion semen plasma are associated with defects of sperm membranes (Colebrander et al. 1992).
Enzymatic and non-enzymatic antioxidants play a very important role in prevention of the effects of ROS. Lack of these antioxidants increases vulnerability of tissues and cells to oxygen reactive forms and increases the risk of oxidative stress (Aitken and Baker 2004). The amount of cytosol in spermatozoa is limited, thereby limiting the antioxidant capacity. Antioxidant enzymes in seminal plasma such as glutathione peroxidase (GPx), superoxide dismutase (SOD), glutathione reductase and catalase therefore play a major role in protecting spermatozoa against lipid peroxidation (Baumer and Ball 2000).
The aim of this study was to compare markers of semen quality, sperm membrane damage, and the seminal plasma antioxidant capacity in stallions with different intensities of physical exercise.
Materials and Methods
Eight fertile warmblood stallions between 6 and 10 years of age with a body weight of 550 ± 30kg were used for the experiment. Their diet was balanced with the requirements of the NRC (2007) and water was available ad libitum. Four stallions were used for breeding only (non-stressed control group), and four horses were used for both breeding as well as standard training and competition in jumping (stressed group).
Semen samples were collected with an artificial vagina (Hanover model, Minitüb, Germany) at 14-day intervals (from June to August) from each stallion (5 ejaculates per stallion). Immediately after sperm collection, conventional examination of the ejaculate (volume, colour, motility, concentration and the hypo-osmotic swelling test) was performed in accordance with Věžník et al. (2004) in the laboratory of the Municipal Stud Farm in Tlumačov. After pH detection, the raw sperm was centrifuged at 2000 × g for 10 min at 4 °C. Samples of seminal plasma were stored at -70 °C.
The catalytic activities of ALT (alanin aminotrasferase) and AST (aspartate aminotrasferase) enzymes were measured by commercial assay kits (Randox Laboratories Ltd, UK). The seminal plasma concentration of F 2α isoprostanes was determined by enzyme immunoassay using the commercially available 8-isoprostane ACE TM Elisa kit (Cayman Chemical Company, Ann Arbor, MI, USA). The activity of glutathione peroxidase (GPx) was determined in accordance with Paglia and Valentin (1967) using commercial kits RS 505 Glutathion Peroxidase -Ransel (Randox Laboratories Ltd, UK). The activity of superoxide dismutase (SOD) was measured by colorimetric assay using a commercially available colorimetric method SD 125 Superoxid dismutase -Ransod (Randox Laboratories Ltd, UK).
Statistical analysis of the data was performed by two-way analysis of variance with treatment and time interactions using the GLM procedure of SAS (SAS Institute Inc. 2003). Differences were considered significant with P < 0.05.
Results
No major differences were found between the two stallion groups comparing the average rates of conventional qualitative and quantitative indicators.
The average pH of seminal plasma was significantly (P < 0.05) influenced by both the term of collection as well as the physical exercise stress (Table 1). Except for the last collection, when the average pH of seminal plasma in the physically stressed stallions significantly (P < 0.05) increased, this indicator was higher in non-stressed stallions. The activities of AST and ALT enzymes in the seminal plasma were significantly (P < 0.05) influenced by the group of animals; tending to be higher in the physically stressed stallions than in the non-stressed ones, particularly in case of AST (Table 1).
Also, the concentration of F 2α isoprostanes in the seminal plasma differed significantly (P < 0.05) between the groups of animals. Despite the large variability of individual data, the average values of F 2α isoprostanes tended to be higher in the physically stressed group (Table 2).
On the contrary, the GPx activity was influenced by the collection term only. The average concentrations were higher in the second half of the experiment (3 rd -5 th sampling, Table 2). The SOD activity was significantly (P < 0.05) influenced by both the stallion groups and the term of collection as well. This was due to its significant increase in the exercised groups at the end of the experiment (5 th sampling, Table 2).
Discussion
All of the conventional semen indicators were within the reference ranges (Juhasz et al. 2000;Věžník et al. 2004) and did not differ significantly between the groups of stallions. Thus, all the monitored stallions met the conditions of applicability for artificial insemination. Nevertheless, conventional assessment of semen immediately after collection does not capture 100% of the spermatozoa fertility, as confirmed by differences e.g. in spermatozoa viability tests (Colebrandner et al. 2003). The reason may be changes in some biochemical indicators of seminal plasma (Podstawski et al. 2007).
The average pH of semen in both groups of stallions was lower than the reference range (Věžník et al. 2004), which could negatively influence spermatozoa quality (Mocé and Graham 2008). Stallions under a workload tended to have a lower pH than those without a workload. Their pH levelled itself out toward that of the non-stressed group just at the end of the monitored period, i.e. the end of the sport season. Stallions under a workload also showed significantly higher activities of AST and ALT. Values of AST and ALT activities were generally higher than those reported by Věžník et al. (2004) and values in the stressed group even slightly exceeded the range reported by Pesch et al. (2006). It is generally accepted that increased activities of these intracellular enzymes in seminal plasma correlate with defects of the spermatozoa membranes (Katila 2001). Therefore, it is obvious that although the monitored groups of stallions did not differ significantly in classic semen quality indicators, the evaluation of seminal plasma pH and AST or ALT revealed differences between these groups to the detriment of the working stallions.
Concentrations of F 2α isoprostanes were significantly influenced in horses of the experimental group. Predominantly higher concentrations of F 2α isoprostanes in the seminal plasma were observed in the exercise-stressed stallions. F 2α isoprostanes are stable end-products of lipid peroxidation of the spermatozoa membrane and therefore can be associated with overproduction of ROS during exercise (Morrow and Roberts 1997;Kirschvink et al. 2002).
The organism is protected from overproduction of ROS by the antioxidant system (Baumer and Ball 2000). In our experiment, the activity of GPx in group of stressed stallions was significantly increased simultaneously with the greatest increase of the F2α isoprostanes concentration. The activity of SOD significantly increased at the following (5 th ) collection. This is in agreement with our assumption that the defense systems will be enhanced in the semen of stallions with higher concentrations of F2α isoprostanes.
Our data show that even the conventional basic sperm indicators within the reference ranges for biochemical indicators of the stallion seminal plasma such as pH or AST/ALT activity may be negatively influenced by sport workload stress. Increased concentrations of F2α isoprostanes indicate that lipoperoxidation can be a mechanism of cell membrane destabilization, which is counteracted by an increase of antioxidant enzyme activities.
Table 1 .
Average values of pH, aspartate aminotransferase, alanin aminotransferase activity in seminal plasma of physically stressed and non-stressed stallions in single collections.*significant (P ≤ 0.05), NS -non significant, a ,b in the line with common superscript do not differ significantly as determined by Scheffe's test, 1, 2 in the column with common superscript do not differ significantly as determined by Scheffe's test, SEM -standard error of the mean, AST -aspartate aminotransferase, ALT -alanin aminotransferase *
Table 2 .
Average values of antioxidant enzymes glutathione peroxidase and superoxiddismutase and of F 2α isoprostanes in seminal plasma of physically stressed or non-stressed stallions in single collections.
**significant (P ≤ 0.05), NS -non significant, a ,b in the line with common superscript do not differ significantly as determined by Scheffe's test, 1, 2 in the column with common superscript do not differ significantly as determined by Scheffe's test, SEM -standard error of the mean, GPx -glutathione peroxidase, SOD -superoxiddismutase
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2018-12-06T23:55:22.644Z", "created": "2013-05-26T00:00:00.000", "id": "55795771", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "HYBRID", "oa_url": "https://actavet.vfu.cz/media/pdf/avb_2013082010031.pdf", "pdf_hash": "5392f7fbbe3aeaceff6b1bdcc75076e4142ca31c", "pdf_src": "Anansi", "provenance": "peS2o-0075.json.gz:38405", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "5392f7fbbe3aeaceff6b1bdcc75076e4142ca31c", "year": 2013}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Pistils may be required for anthocyanin synthesis —— the whole-transcriptome analysis of mutant and normal capitula of Chrysanthemum morifolium
Background Chrysanthemum morifolium is one of the most economically important and popular floricultural crops in Asteraceae. Chrysanthemums have many different flower colors and shapes. However, the molecular mechanism controlling the development of chrysanthemum floral colors and shapes is still an enigma. We obtained a cut chrysanthemum variety with mutant capitula in which the ray florets became green and the inside pistils became vegetative buds, while normal capitula have many rounds of purple ray florets and few disc florets. Results We conducted whole-transcriptome analysis of differentially expressed genes (DEGs) between the mutant and normal capitula using third-generation and second-generation sequencing techniques. We identified DEGs between the mutant and normal capitula to reveal important regulators underlying their differential development. Regulatory genes involved in the photoperiod pathway and the control of floral organ identification as well as important functional genes in the anthocyanin synthesis pathway were also identified. Therefore, a list of candidate genes for studying flower development and anthocyanin synthesis in chrysanthemums was generated. Qualitative analysis of pigments in the florets of normal and mutant capitula revealed anthocyanins were synthesized and accumulated in the florets of normal capitula, but not in the florets of mutant capitula. It was indicated that pistils may be required for anthocyanin synthesis in chrysanthemums. Conclusions These results will help to elucidate the molecular mechanisms of floral organ development and will contribute to the development of techniques for studying flower shape and color regulation to promote breeding in chrysanthemum.
Introduction
Chrysanthemum morifolium is one of the most economically important and popular floricultural crops in Asteraceae, and ranks second in the cut flower trade after rose [1].
The head-like inflorescence (capitulum) resembling a single large flower is the main ornamental part of C. morifolium, and is also regarded as the key innovation behind the evolutionary success of the Asteraceae [2]. The typical capitulum of chrysanthemum is formed by two morphologically distinct florets: the marginal ray florets and the central disc florets. Ray florets have ligulate and zygomorphic colorful corollas (petals) and aborted stamens, which function in attracting pollinators. The disc florets have radially symmetrical colorless corollas and their fertile pollens are hermaphroditic and are used for reproduction in chrysanthemum (Additional file 1). The colors and shapes of the flowers are the most visible and amazing products of evolution, and also connect humans to nature [3].
Flowering is a key developmental process in the plant life cycle that is very complex and is controlled by endogenous factors and environmental cues. As currently understood, flower development contains three phases: flowering determination, flower evocation, and floral organ development [4]. In Arabidopsis, tremendous progress has been achieved toward understanding the molecular mechanisms involved in flower development [5,6].
ABCE models have revealed that A-class together with E-class genes specify sepal identity, A-class, B-class and E-class genes specify petals, B-class, C-class and E-class genes determine stamens, and C-class and E-class genes determine carpel/gynoecium organ identity [7]. With the notable exception of A-class genes, all of these genes belong to the MADS-box family of transcription factors including the AP1, AP3, PI, AG and SEP genes.
The splendid colors presented mainly by flower petals have enabled plants to constantly develop new showy traits and prosper throughout millions of years of evolution.
Anthocyanins and carotenoids are the two major groups of pigments generated in plant petals. Anthocyanins are accumulated in the vacuoles of petal epidermal cells and confer orange-to-violet colors in flowers [8]. Beside attracting pollinators, anthocyanins also protect against damage from UV irradiation [9]. Anthocyanins provide chrysanthemum ray florets with colorful bright colors to attract pollinators, which improves the success rate of cross pollination between different species or varieties and promotes cultivar groups with large varieties of flower types in C. morifolium. Anthocyanins enhance the ornamental value of chrysanthemums, and many cut flower and pot flower varieties with bright colors are on sale every year to satisfy the demand of the market. Understanding the mechanism of anthocyanin biosynthesis and its regulation will contribute to the cultivation and improvement of new color varieties of chrysanthemums.
In chrysanthemums, a few regulatory genes involved in flower development have been isolated, such as MADS-box, TCP, and WUS-like genes [10][11][12]16]. Some important functional genes and transcription factors involved in the anthocyanin biosynthesis pathway have also been characterized including ANS, F3′H, F3H and MYB-like genes [13][14][15][16]. However, chrysanthemums have complex capitula containing two morphologically distinct florets and through a long period of breeding a variety of flower shapes and diverse colors have been created. The mechanism involved in chrysanthemum flower evolution and development is extremely complicated and little is known about it.
The development of RNA-seq technology has greatly improved transcriptomic studies in chrysanthemums [1]. However, the read-length offered by second-generation highthroughput sequencing platforms is much shorter than the typical length of a eukaryotic mRNA. Additionally, the differences in transcript abundance and the presence of different isoforms make the assembly of transcriptomes from short reads extremely challenging [17].
Despite these problems, Hideki Hirakawa et al. performed de novo whole-genome assembly in C. seticuspe using the Illumina sequencing platform and Chi Song et al.
sequenced the diploid C. nankingense genome using the Oxford Nanopore long-read technology; however, no more than 40% of the transcriptome sequencing reads from C. morifolium can be mapped to each of these two genome sequences, probably because of the extreme variation in chromosome ploidy and biological characteristics [18,19]. Thirdgeneration sequencing technology has dramatically increased the length of sequencing reads, which enables the precise location and sequencing of repetitive regions and isoforms within a single read.
Recently, we obtained a mutant plant of the cut chrysanthemum variety C. morifolium 'ZY' with both mutant and normal capitula. In the mutant capitula the ray florets became green and the inside pistils became vegetative buds, while the normal capitula had many rounds of purple ray florets and few disc florets. In this study, we analyzed a mixed sample of normal and mutant flowers, leaves, stems and roots of 'ZY' with the single-molecule long-read sequencing technology from Pacific Biosciences (PacBio). Based on the results, transcriptional sequencing and analysis of the mutant and normal capitula were performed using second-generation sequencing technology and RNA-Seq quantification analysis.
Thus, we combined third-generation and second-generation sequencing to generate a more complete/full-length C. morifolium transcriptome.
Based on the transcriptional sequencing and analysis, we identified differentially expressed genes (DEGs) between mutant and normal capitula to reveal important regulators controlling their differential development. Regulatory genes involved in the photoperiod pathway and the control of floral organ identification as well as important functional genes in the anthocyanin synthesis pathway were also identified to create a list of candidate genes for studying flower development and anthocyanin synthesis in chrysanthemums. These results will be helpful for elucidating the molecular mechanisms of floral organ development and will contribute to the development of techniques for studying flower shape and color regulation, as well as breeding and molecular biology in chrysanthemum.
Sequencing and assembly
As shown in Figure 1, in the mutant capitula the ray florets became green and the inside pistils became vegetative buds, while the normal capitula had many rounds of purple ray florets and few disc florets. We analyzed a mixed sample of normal and mutant capitula, leaves, stems and roots of 'ZY' using PacBio sequencing and then analyzed the normal and mutant capitula separately using the Illumina paired-end sequencing technology. The resulting sequences were assembled into 130,097 isoforms with an N50 of 3013 bp and average length of 2510 bp (Table 1).
Gene annotation and functional classification
A total of 118,589 isoforms were annotated by BLAST in at least one of the four databases searched (nr, Swiss-Prot, KOG, and KEGG), leaving 11,508 (8.85%) isoforms without annotation. In total, 118,043, 101,048, 87,630, and 54,245 isoforms were annotated in the nr, Swiss-Prot, KOG, and KEGG databases, respectively. Gene Ontology (GO) has three ontologies-molecular functions, cellular components, and biological processes-that facilitate gene annotation and analysis. A total of 36,144 isoforms were classified into 47 functional categories, including 19, 17 and 11 GO categories in the biochemical process, cellular component and molecular function ontologies, respectively.
The dominant categories in the biochemical process, molecular function and cellular component ontologies were 'metabolic process' (20,871), 'catalytic activity' (22,818) and 'cell' (11,887), respectively, indicating that numerous metabolic activities were activated during the development of chrysanthemum capitula and that this process was regulated by a wide range of genes that interacted within cells. In addition, we observed a high percentage of genes from the 'cellular process', 'binding', and 'cell part' categories, but few from 'locomotion', 'transcription factor activity, protein binding', and 'extracellular matrix (2,089 members). In addition, 1,339 isoforms were assigned to 'Plant hormone signal transduction'.
Comparison of the transcriptomes of normal and mutant capitula
The set of isoforms common to normal and mutant capitula In total, 124,284 isoforms were shared by normal and mutant capitula ( Figure 5A). By contrast, 3269 and 955 isoforms showed specific expression in normal and mutant capitula, respectively. Therefore, more genes were expressed in normal capitula than in mutant capitula, because the pistils were mutated to vegetative buds in mutant capitula.
DEGs between mutant and normal capitula
The transcriptomes of the normal and mutant capitula were compared, and the resulting reads were mapped to the reference transcriptome. A total of 35,419 DEGs (8,232 upregulated and 27,187 down-regulated in the mutant capitula relative to the normal capitula) were identified between the normal and mutant capitula ( Figure 5B). The correlation coefficient of gene expression between the normal and mutant capitula was 0.8897, which was determined by an algorithm developed from the correlation scatter plot.
A total of 131 DEGs were specifically expressed in the mutant capitula relative to the normal capitula, including TCP1 and AP2/ERF domain-containing genes. Conversely, 2,132 genes were not expressed in the mutant capitula but were specifically expressed in the normal capitula, including some important transcription factor genes (MYB, GRAS, and BTF3 genes), ubiquitin-conjugating enzyme genes, zinc finger protein genes and many genes without annotations. Annotation information on the DEGs specifically expressed in the mutant and normal capitula is provided in Additional files 2 and 3, respectively. These genes may play important roles during the flower development process especially during pistil determination and development in chrysanthemums. Additionally, these genes should be important candidate regulators in chrysanthemum flower development. Therefore, functional research should be conducted on these genes in the future.
GO and KEGG pathway enrichment analyses were conducted on the DEGs to identify differences in biological processes and pathways between the mutant and normal capitula.
It was found that 256 genes enriched in the 'reproduction' term (GO:0000003) in biochemical processes were all remarkably down-regulated in the mutant capitula relative to the normal capitula, among which 11 genes were specifically expressed in the normal capitula including WD40 and UBA1-like protein-encoding genes. These genes may play important roles in the regulatory pathways related to reproduction in chrysanthemums (Additional file 4).
In total, 6,733, 7,216, and 3,879 DEGs were enriched in the biological process, molecular function and cellular component categories, respectively (Additional files 5-7).
In the biological process category, the dominant terms were 'metabolic process'
Identification of genes involved in the photoperiod pathway in chrysanthemum
As a typical short-day plant, chrysanthemum can flower in response to a single short day. identified in this study, which is an upstream regulatory gene of LFY. Interestingly, its expression level was significantly up-regulated in the mutant capitula relative to the normal capitula. As an 'A-class'-like gene, the expression of AP1 is directly activated by LFY [20,21]. Homologs of AP1 were identified, which were all remarkably up-regulated in the mutant capitula relative to the normal capitula.
As another 'A-class' gene, AP2 is not a MADS-box transcription factor. Two homologs of AP2 were identified, both of which were up-regulated in the normal capitula relative to the mutant capitula. In most core eudicot species, B-class genes include three different lineages: PI, euAP3 and TM6; however, TM6-like genes seem to have been lost in Arabidopsis and Antirrhinum [22]. In this study, homologs of PI and AP3 were identified, but the expression of TM6-like genes was not detected. In contrast, the expression of TM6like genes was detected in chrysanthemums in earlier studies [23,24]. We also identified homologs of the C-class gene AG and E-like MADS-box genes in this study. We found that Interestingly, all four DFR homologs and one CHI homolog showed extremely low expression levels (close to 0) in the mutant capitula relative to the normal capitula. Realtime PCR showed that the CHI homolog was not expressed in the mutant capitula but was expressed in the normal capitula. Annotation information for these genes in the anthocyanin biosynthetic pathway is listed in Additional file 12.
The pigment types and contents in flowers determine the variety of flower colors.
Qualitative analysis of pigments in the florets of normal and mutant capitula was performed by HPLC. Anthocyanins were detected in the florets of normal capitula, but not in the florets of mutant capitula ( Figure 8B). As shown in Figure 8A, the detectable anthocyanins mainly included delphinidin, cyanidin, petunidin, pelargonidin, peonidin, and malvidin. Figure 9).
Discussion
After the pistils mutated to become vegetative buds, anthocyanins were not synthesized, which indicated pistils may be required for anthocyanin synthesis in chrysanthemums As one of the major flower pigment groups in higher plants, anthocyanin synthesis and accumulation is an integral part of flower development in most plant species and is tightly linked with petal cell expansion. It is thought that the activation of the anthocyanin synthesis pathway during petal development requires both environmental and endogenous signals.
In Petunia hybrida, gibberellins (GAs), sugars and light were revealed to be required to induce the transcription of anthocyanin synthesis genes and accumulate pigment in the developing corolla. Research results have also indicated that GAs, sugars and light are involved in the regulation of various pathways to complete the entire flower development process [25]. David Weiss presented evidence that in the early stages of Petunia hybrida flower development, GAs produced by the anthers controlled anthocyanin synthesis and accumulation in the corolla by activating the transcription of related genes in the anthocyanin synthesis pathway, and the removal of stamens at the early stage of flower development inhibited anthocyanin synthesis in corollas [26]. However, in C. morifolium, the capitula of many varieties, such as those with the 'Pingpang' shape, have only ray florets, because of the long period of double flower breeding. Interestingly, in C. morifolium 'ZY', most capitula have ray florets without disc florets. Therefore, in chrysanthemums, anthocyanin synthesis may be associated with pistils.
In the mutant capitula, the pistils of ray florets mutated to become vegetative buds and the corollas of ray florets became green because of a lack of anthocyanins. When the pistils of ray florets mutated into vegetative buds, anthocyanin synthesis in the corolla may have been blocked. Therefore, it was hypothesized that pistils may be required for anthocyanin synthesis in chrysanthemums.
Many transcription factors are implicated in the developmental regulation of florets in chrysanthemums
In this study, 963 of 3,921 detected important transcription factors genes were dramatically differentially expressed between the normal and mutant capitula, including members of the ERF, C2H2, MYB, bHLH and WRKY transcription factor families. This C2H2 zinc finger protein genes functioning as transcription factors play important roles in many biological processes related to plant growth and development, hormone signaling and stress responses [27]. In many plants, C2H2 zinc finger protein genes are involved in salt, cold, drought and oxidative tolerance, light stress and pathogen defense [28].
Additionally, some members participate in the developmental regulation of flowers. For example, SIZF2 controls flower and leaf shape in A. thaliana. SUPERMAN (SUP) determines the boundary between the stamen and carpel whorls, suppresses class B gene expression and promotes stem cell termination in the fourth whorl of A. thaliana flowers [28][29][30]. In this study, 35 C2H2 genes showed no expression in the mutant capitula, which indicated that these C2H2 zinc finger protein genes may have important functions in the determination and growth regulation of pistils or anthocyanin synthesis and accumulation in the ray floret corolla in chrysanthemums.
As one of the largest transcription factor families, basic leucine zipper (bZIP) family transcription factors play key roles in controlling plant development and stress responses [31]. bZIP genes have been reported to be involved in various flower developmental processes in plants, including pollen development, the floral transition and flower initiation [32,33]. In this study, the expression deficiency of six bZIP genes in mutant capitula suggested that these bZIPs are also important regulators of ray floret development in chrysanthemums.
MYB transcription factors comprise one of the largest and most diverse transcription factor families in the plant kingdom, exist widely in eukaryotes, and play essential roles in a wide range of physiological and biochemical processes controlling plant growth and development [34,35]. bHLH transcription factors possess the highly conserved bHLH domain including a basic region and a HLH region, and play important roles in plant growth and development, metabolic regulation, and responses to environmental changes. The regulatory functions of bHLH transcription factors in active secondary metabolism especially anthocyanin synthesis have been a focus of research [36].
It was revealed that R2R3-MYB, bHLH, and WD40 proteins form a ternary complex called the MBW complex that regulates the anthocyanin biosynthesis pathway in model plants [37]. We found that five bHLH genes and seven MYB genes showed no expression in the mutant capitula, which indicated that these genes are essential regulators of normal
Carpels may originate from vegetative buds
The female reproductive organ, named the carpel, is specific to the angiosperms, or flowering plants. The carpel is probably the major factor contributing to the success of angiosperms, because it allowed angiosperms to diversify from an unconfirmed, possibly gymnosperm-like ancestor to form the more than 300,000 species alive today [39]. As in most species, the chrysanthemum carpel is differentiated into stigma, style, and ovary tissues. Flower developmental mechanism studies show that flowers develop from a floral apex. On the flanks of the floral apex, floral organ development starts in a centripetal sequence: the outermost organs are initiated first and the innermost organs are initiated last. However, the evolutionary sequence of floral organs is reversed. The ovules in the center of the flower are the evolutionarily oldest organs and the perianth organs at the periphery are the youngest [40]. Each organ has its own evolutionary history.
In this study, the pistils in the mutant capitula were mutated and became vegetative buds.
In the early stages, the vegetative buds mutated from the pistils looked very much like normal pistils ( Figure 1). As the mutant florets grew, the two seemingly bifurcate stigma became two tender leaves and the center part between the two seemingly bifurcate stigma grew into a shoot apex. Here we present a bold hypothesis; that the pistil is derived from a vegetative bud. The predecessor of the stigma may be tender leaves developed from the phyllopodium of the vegetative bud; thus, the ovule would have evolved out of the apical meristem of the vegetative bud. This hypothesis supports the inference that the ovule did not originate from a phyllome, and each ovule represents a short shoot [41,42]. Additionally, the dorsiventral gene expression in the outer integument of the ovule indicates that the outer integument of ovules has a leaf-derived origin, which is consistent with the inference that the stigma is derived from leaves [43,44].
Conclusions
In our study, comparative transcriptome analysis revealed significant differences in gene expression and signaling pathways between the mutant and normal capitula. We identified DEGs between the mutant and normal capitula to reveal important regulators underlying their differential development. The transcription factors specifically expressed in the normal capitula in this study are important candidate genes to explore the regulatory molecular mechanisms in chrysanthemum flower development. Regulatory genes involved in the photoperiod pathway and the control of floral organ identification as well as important functional genes in the anthocyanin synthesis pathway were also identified. Qualitative analysis of pigments in the florets of normal and mutant capitula revealed anthocyanins were synthesized and accumulated in the florets of normal capitula, but not in the florets of mutant capitula. It was indicated that pistils may be required for anthocyanin synthesis in chrysanthemums. This study represents the first step in exploring the molecular mechanism of floral organ development and will contribute to the development of techniques for studying flower shape and color regulation to promote breeding in chrysanthemum.
Plant materials and RNA extraction
The tissues (normal and mutant capitula) used in this study were obtained from a cut- RNA quantity and quality assessment was performed using a NanoDropND2000 instrument (Thermo Scientific).
Library construction, PacBio sequencing and Data processing
Library construction and PacBio sequencing were conducted with the PacBio Sequel system (PacBio, CA, USA). Briefly, 1 μg of total RNA extracted from each of the five tissues was equally pooled together and the mRNA was enriched by Oligo (dt) magnetic beads. Second, the low-quality isoforms were further corrected using Illumina short reads obtained from the same samples with the LoRDEC tool (version 0.8) [45]. Then, the final transcriptome isoform sequences were filtered by removing the redundant sequences with the CD-HIT-v4.6.7 software using a threshold of 0.99 identities.
Illumina sequencing
The total RNA isolated from the normal and mutant capitula was used for Illumina sequencing on an Illumina HiSeq™ 2000 system (Illumina, San Diego, CA, USA). We purified the poly (A) mRNAs, fragmented them into small pieces, and then synthesized the first-and second-strand cDNAs.
After the double-stranded cDNAs were purified and resolved for end reparation and poly (A) tail addition, the short fragments were connected with sequencing adapters. Briefly, a cDNA library with average insert sizes of 300-500 bp was created and cDNA sequencing was performed using the Illumina HiSeq™ 2000 system, with paired end 2×100 nt multiplexes. [38]. GO annotation was also performed with the Blast2GO software [46]. Isoforms ranking in the 20 highest scores and no shorter than 33 HSP (High-scoring Segment Pair) hits were selected for Blast2GO analysis. Then, functional classification of the isoforms was performed using the WEGO software [47].
Analysis of chrysanthemum transcriptome sequencing results
RNA-Seq quantification analysis using the number of reads per kilobase of exon model per million mapped reads (RPKM) was performed to calculate the expression level of each isoform [39]. The chrysanthemum transcriptome of the samples was used as a reference for the screening and analysis of DEGs. A rigorous algorithm was created based on the method of Audic et al. to screen the DEGs [48]. The false discovery rate (FDR) was used to affirm the threshold of the P-value in multiple tests and analyses [41]. The absolute value of log2 (ratio) ≥ 2 and FDR < 0.05 were used as the thresholds to determine significant differences in gene expression [49]. Only the DEGs with a minimum of a twofold change in expression level were used in the differential gene expression analysis.
Alternative splicing detection
To analyze alternative splicing events in the transcript isoforms, the Coding Genome reconstruction Tool (Cogent) was employed to partition the transcripts into gene families based on k-mer similarity and then each family was reconstructed into a coding reference genome with De Bruijn graph methods [50]. The alternative splicing events of the transcript isoforms were analyzed using the SUPPA tool [51].
Gene expression analysis based on qRT-PCR
Total RNA was extracted from the normal and mutant capitula as described above. First, we treated the total RNA with DNase (Promega, USA), and then subjected it to reverse transcription into cDNA using a reverse transcription system (Tiangen, China [52].
Qualitative analysis of pigments in chrysanthemum flowers
The analysis of anthocyanin profiles in the normal and mutant capitula was performed using high pressure liquid chromatography. A 1.0-10.0 g sample was ground into fine powder in liquid N2 and homogenized into 50 ml anthocyanin extracts [ethyl alcohol: distilled water: hydrochloric acid (2:1:1, v/v/v)] assisted by sonication at 20 °C for 30 min [53]. Then, the mixture was heated in boiling water for 1 h using a water bath and centrifuged at 16000 ×g for 10 min at 20 °C. min. The injection volume was 20 μL and the photodiode array detector was set at 530 nm for anthocyanins [54]. Three biological replicates of each sample type were analyzed.
==
Domain: Agricultural And Food Sciences Biology
|
[{"added": "2020-05-28T09:17:57.450Z", "created": "2020-05-26T00:00:00.000", "id": "234701490", "metadata": {"extfieldsofstudy": ["Biology"], "oa_license": "CCBY", "oa_status": "GREEN", "oa_url": "https://bmcgenomdata.biomedcentral.com/track/pdf/10.1186/s12863-021-00959-2", "pdf_hash": "0aa1bc8fcb3e20bb63305d72b3186cbac0f74f2c", "pdf_src": "MergedPDFExtraction", "provenance": "peS2o-0075.json.gz:30722", "s2fieldsofstudy": ["Agricultural And Food Sciences"], "sha1": "554204ca811be17b402a5b67361d7d5313e85feb", "year": 2020}, "source": "pes2o/s2orc", "version": "v3-fos-license"}]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.