Limits...
Can (13)C stable isotope analysis uncover essential amino acid provisioning by termite-associated gut microbes?

Ayayee PA, Jones SC, Sabree ZL - PeerJ (2015)

Bottom Line: Gut-associated microbes of insects are postulated to provide a variety of nutritional functions including provisioning essential amino acids (EAAs).In this study, we investigated whether the eastern subterranean termite Reticulitermes flavipes sourced EAAs from its gut-associated microbiota. δ (13)CEAA data from termite carcass, termite gut filtrate and dietary (wood) samples were determined following (13)C stable isotope analysis.Despite this limitation, this study provides tentative data in support of hypothesized EAA provisioning by gut microbes, and also a baseline/framework upon which further work can be carried out to definitively verify this function.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Evolution, Ecology and Organismal Biology, The Ohio State University , Columbus, OH , USA.

ABSTRACT
Gut-associated microbes of insects are postulated to provide a variety of nutritional functions including provisioning essential amino acids (EAAs). Demonstrations of EAA provisioning in insect-gut microbial systems, nonetheless, are scant. In this study, we investigated whether the eastern subterranean termite Reticulitermes flavipes sourced EAAs from its gut-associated microbiota. δ (13)CEAA data from termite carcass, termite gut filtrate and dietary (wood) samples were determined following (13)C stable isotope analysis. Termite carcass samples (-27.0 ± 0.4‰, mean ± s.e.) were significantly different from termite gut filtrate samples (-27.53 ± 0.5‰), but not the wood diet (-26.0 ± 0.5‰) (F (2,64) = 6, P < 0.0052). δ (13)CEAA-offsets between termite samples and diet suggested possible non-dietary EAA input. Predictive modeling identified gut-associated bacteria and fungi, respectively as potential major and minor sources of EAAs in both termite carcass and gut filtrate samples, based on δ (13)CEAA data of four and three EAAs from representative bacteria, fungi and plant data. The wood diet, however, was classified as fungal rather than plant in origin by the model. This is attributed to fungal infestation of the wood diet in the termite colony. This lowers the confidence with which gut microbes (bacteria and fungi) can be attributed with being the source of EAA input to the termite host. Despite this limitation, this study provides tentative data in support of hypothesized EAA provisioning by gut microbes, and also a baseline/framework upon which further work can be carried out to definitively verify this function.

No MeSH data available.


Related in: MedlinePlus

Second discriminant analysis of termite, bacteria, fungi and plant samples.Predictive modeling (LDA) using δ13CEAA data based on three classifier groups (plants (n = 12), fungi (n = 9), and bacteria (n = 11)) and three predictor groups (termite carcass (n = 5), termite gut filtrate (n = 5), and wood diet (n = 3)) using the EAAs; lysine (Lys), phenylalanine (Phe), and valine (Val). Wilks’ lambda = 0.09, P < 0.0001; LD1 = 95.3%, LD2 = 4.6%. 95% confidence limits decision regions for each group/classifier are depicted as ellipses around the classifiers and the decision boundaries between the groups/classifiers as lines.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4556154&req=5

fig-3: Second discriminant analysis of termite, bacteria, fungi and plant samples.Predictive modeling (LDA) using δ13CEAA data based on three classifier groups (plants (n = 12), fungi (n = 9), and bacteria (n = 11)) and three predictor groups (termite carcass (n = 5), termite gut filtrate (n = 5), and wood diet (n = 3)) using the EAAs; lysine (Lys), phenylalanine (Phe), and valine (Val). Wilks’ lambda = 0.09, P < 0.0001; LD1 = 95.3%, LD2 = 4.6%. 95% confidence limits decision regions for each group/classifier are depicted as ellipses around the classifiers and the decision boundaries between the groups/classifiers as lines.

Mentions: The performance of the classification model without isoleucine from all samples was investigated, due to concerns about the influence of isoleucine on the displacement of samples in the LDA plot (Fig. 1). Bacteria, fungi and plant samples were correctly classified into distinct groups (F(8,54) = 28.9, P < 0.0001; Wilk’s lambda = 0.06, a test of appropriateness of classifiers in predicting group membership of predictors) (Table S3). As with the previous analysis, the test fungus, F. solani samples were similarly correctly classified as fungal in origin (Fig. 3), further validating the model and the classification in the absence of isoleucine. Omitting isoleucine from the model resulted in the placement of four termite gut filtrate samples within the 95% confidence limit decision region of the bacterial classifier, and the fifth one within the fungal classifier decision region (Fig. 3). Four termite carcass samples were classified as bacterial in origin, and three were located within the decision region of the bacterial classifier group. The fifth termite carcass sample was classified as fungal (Fig. 3). Classification of both termite carcass and termite gut filtrate samples in both models (Figs. 2 and 3) was essentially similar (Table S3). Omitting isoleucine in the second analysis, nonetheless, minimized the variance between samples and reduced the skewing of the samples within the LDA plot (Fig. 3). Based on the 13C-offset data and the results from both predictive model analyses, the hypothesis of gut microbial EAA is tentatively substantiated. Nonetheless, the sourcing of EAAs from extracellular wood-degrading fungi by termites in this study remains an additional/alternate possibility.


Can (13)C stable isotope analysis uncover essential amino acid provisioning by termite-associated gut microbes?

Ayayee PA, Jones SC, Sabree ZL - PeerJ (2015)

Second discriminant analysis of termite, bacteria, fungi and plant samples.Predictive modeling (LDA) using δ13CEAA data based on three classifier groups (plants (n = 12), fungi (n = 9), and bacteria (n = 11)) and three predictor groups (termite carcass (n = 5), termite gut filtrate (n = 5), and wood diet (n = 3)) using the EAAs; lysine (Lys), phenylalanine (Phe), and valine (Val). Wilks’ lambda = 0.09, P < 0.0001; LD1 = 95.3%, LD2 = 4.6%. 95% confidence limits decision regions for each group/classifier are depicted as ellipses around the classifiers and the decision boundaries between the groups/classifiers as lines.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4556154&req=5

fig-3: Second discriminant analysis of termite, bacteria, fungi and plant samples.Predictive modeling (LDA) using δ13CEAA data based on three classifier groups (plants (n = 12), fungi (n = 9), and bacteria (n = 11)) and three predictor groups (termite carcass (n = 5), termite gut filtrate (n = 5), and wood diet (n = 3)) using the EAAs; lysine (Lys), phenylalanine (Phe), and valine (Val). Wilks’ lambda = 0.09, P < 0.0001; LD1 = 95.3%, LD2 = 4.6%. 95% confidence limits decision regions for each group/classifier are depicted as ellipses around the classifiers and the decision boundaries between the groups/classifiers as lines.
Mentions: The performance of the classification model without isoleucine from all samples was investigated, due to concerns about the influence of isoleucine on the displacement of samples in the LDA plot (Fig. 1). Bacteria, fungi and plant samples were correctly classified into distinct groups (F(8,54) = 28.9, P < 0.0001; Wilk’s lambda = 0.06, a test of appropriateness of classifiers in predicting group membership of predictors) (Table S3). As with the previous analysis, the test fungus, F. solani samples were similarly correctly classified as fungal in origin (Fig. 3), further validating the model and the classification in the absence of isoleucine. Omitting isoleucine from the model resulted in the placement of four termite gut filtrate samples within the 95% confidence limit decision region of the bacterial classifier, and the fifth one within the fungal classifier decision region (Fig. 3). Four termite carcass samples were classified as bacterial in origin, and three were located within the decision region of the bacterial classifier group. The fifth termite carcass sample was classified as fungal (Fig. 3). Classification of both termite carcass and termite gut filtrate samples in both models (Figs. 2 and 3) was essentially similar (Table S3). Omitting isoleucine in the second analysis, nonetheless, minimized the variance between samples and reduced the skewing of the samples within the LDA plot (Fig. 3). Based on the 13C-offset data and the results from both predictive model analyses, the hypothesis of gut microbial EAA is tentatively substantiated. Nonetheless, the sourcing of EAAs from extracellular wood-degrading fungi by termites in this study remains an additional/alternate possibility.

Bottom Line: Gut-associated microbes of insects are postulated to provide a variety of nutritional functions including provisioning essential amino acids (EAAs).In this study, we investigated whether the eastern subterranean termite Reticulitermes flavipes sourced EAAs from its gut-associated microbiota. δ (13)CEAA data from termite carcass, termite gut filtrate and dietary (wood) samples were determined following (13)C stable isotope analysis.Despite this limitation, this study provides tentative data in support of hypothesized EAA provisioning by gut microbes, and also a baseline/framework upon which further work can be carried out to definitively verify this function.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Evolution, Ecology and Organismal Biology, The Ohio State University , Columbus, OH , USA.

ABSTRACT
Gut-associated microbes of insects are postulated to provide a variety of nutritional functions including provisioning essential amino acids (EAAs). Demonstrations of EAA provisioning in insect-gut microbial systems, nonetheless, are scant. In this study, we investigated whether the eastern subterranean termite Reticulitermes flavipes sourced EAAs from its gut-associated microbiota. δ (13)CEAA data from termite carcass, termite gut filtrate and dietary (wood) samples were determined following (13)C stable isotope analysis. Termite carcass samples (-27.0 ± 0.4‰, mean ± s.e.) were significantly different from termite gut filtrate samples (-27.53 ± 0.5‰), but not the wood diet (-26.0 ± 0.5‰) (F (2,64) = 6, P < 0.0052). δ (13)CEAA-offsets between termite samples and diet suggested possible non-dietary EAA input. Predictive modeling identified gut-associated bacteria and fungi, respectively as potential major and minor sources of EAAs in both termite carcass and gut filtrate samples, based on δ (13)CEAA data of four and three EAAs from representative bacteria, fungi and plant data. The wood diet, however, was classified as fungal rather than plant in origin by the model. This is attributed to fungal infestation of the wood diet in the termite colony. This lowers the confidence with which gut microbes (bacteria and fungi) can be attributed with being the source of EAA input to the termite host. Despite this limitation, this study provides tentative data in support of hypothesized EAA provisioning by gut microbes, and also a baseline/framework upon which further work can be carried out to definitively verify this function.

No MeSH data available.


Related in: MedlinePlus