The human gut microbiome is linked to many states of human health and disease1. The metabolic repertoire of the gut microbiome is vast, but the health implications of these bacterial pathways are poorly understood. In this study, we identify a link between members of the genus Veillonella and exercise performance. We observed an increase in Veillonella relative abundance in marathon runners postmarathon and isolated a strain of Veillonella atypica from stool samples. Inoculation of this strain into mice significantly increased exhaustive treadmill run time. Veillonella utilize lactate as their sole carbon source, which prompted us to perform a shotgun metagenomic analysis in a cohort of elite athletes, finding that every gene in a major pathway metabolizing lactate to propionate is at higher relative abundance postexercise. Using 13C3-labeled lactate in mice, we demonstrate that serum lactate crosses the epithelial barrier into the lumen of the gut. We also show that intrarectal instillation of propionate is sufficient to reproduce the increased treadmill run time performance observed with V. atypica gavage. Taken together, these studies reveal that V. atypica improves run time via its metabolic conversion of exercise-induced lactate into propionate, thereby identifying a natural, microbiome-encoded enzymatic process that enhances athletic performance.
Access optionsAccess options
Subscribe to Journal
Get full journal access for 1 year
only $18.75 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
All raw sequencing data have been uploaded to NCBI and SRA in the form of the BioProjects PRJNA472785 (16S) and PRJNA472768 (MGX). These are linked to associated BioSamples, which in turn are linked to the paired-end read files in the SRA, and correspond to the metadata in the Supplementary Information files.
Unless otherwise noted, all plots were generated in R version 3.4.1 with the ggplot2, dplyr, scales, grid and reshape2 packages28,37,38,39,40. Large-scale data analysis was done on AWS, utilizing machines running Ubuntu 16.04. Data curation methods were coded in python version 2.7.12. The Aether package utilized for analysis is available at https://github.com/kosticlab/aether.
Gilbert, J. A. et al. Current understanding of the human microbiome. Nat. Med. 24, 392–400 (2018).
Lax, S. et al. Longitudinal analysis of microbial interaction between humans and the indoor environment. Science 345, 1048–1052 (2014).
Rothschild, D. et al. Environment dominates over host genetics in shaping human gut microbiota. Nature 555, 210–215 (2018).
Dusko Ehrlich, S. & The MetaHIT Consortium. in Metagenomics of the Human Body 307–316 (Springer, 2011).
Petersen, L. M. et al. Community characteristics of the gut microbiomes of competitive cyclists. Microbiome 5, 98 (2017).
Clarke, S. F. et al. Exercise and associated dietary extremes impact on gut microbial diversity. Gut 63, 1913–1920 (2014).
Garvie, E. I. Bacterial lactate dehydrogenases. Microbiol. Rev. 44, 106–139 (1980).
Truong, D. T. et al. MetaPhlAn2 for enhanced metagenomic taxonomic profiling. Nat. Methods 12, 902–903 (2015).
Luber, J. M., Tierney, B. T., Cofer, E. M., Patel, C. J. & Kostic, A. D. Aether: leveraging linear programming for optimal cloud computing in genomics. Bioinformatics 34, 1565–1567 (2017).
Li, D., Liu, C.-M., Luo, R., Sadakane, K. & Lam, T.-W. MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics 31, 1674–1676 (2015).
Seemann, T. Prokka: rapid prokaryotic genome annotation. Bioinformatics 30, 2068–2069 (2014).
Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012).
Qin, J. et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature 490, 55–60 (2012).
Van den Bogert, B., Boekhorst, J., Smid, E. J., Zoetendal, E. G. & Kleerebezem, M. Draft genome sequence of Veillonella parvula HSIVP1, isolated from the human small intestine. Genome Announc. 1, e00977-13 (2013).
Ng, S. K. C. & Hamilton, I. R. Carbon dioxide fixation by Veillonella parvula M4 and its relation to propionic acid formation. Can. J. Microbiol. 19, 715–723 (1973).
Phypers, B. & Pierce, J. M. T. Lactate physiology in health and disease. Contin. Educ. Anaesth. Crit. Care Pain 6, 128–132 (2006).
Kimura, I. et al. Short-chain fatty acids and ketones directly regulate sympathetic nervous system via G protein-coupled receptor 41 (GPR41). Proc. Natl Acad. Sci. USA 108, 8030–8035 (2011).
Pluznick, J. A novel SCFA receptor, the microbiota, and blood pressure regulation. Gut Microbes 5, 202–207 (2014).
Pluznick, J. L. et al. Olfactory receptor responding to gut microbiota-derived signals plays a role in renin secretion and blood pressure regulation. Proc. Natl Acad. Sci. USA 110, 4410–4415 (2013).
Chambers, E. S. et al. Acute oral sodium propionate supplementation raises resting energy expenditure and lipid oxidation in fasted humans. Diabetes Obes. Metab. 20, 1034–1039 (2018).
Araghizadeh, F. & Abdelnaby, A. in Colorectal Surgery (eds Bailey, H. R., Billingham, R. P., Stamos, M. J. & Snyder, M. J.) 3–17 (Elsevier Health Sciences, 2012).
Callahan, B. J. et al. DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583 (2016).
McMurdie, P. J. & Holmes, S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8, e61217 (2013).
Jurkowski, J. E., Jones, N. L., Toews, C. J. & Sutton, J. R. Effects of menstrual cycle on blood lactate, O2 delivery, and performance during exercise. J. Appl. Physiol. 51, 1493–1499 (1981).
Pimentel, G. et al. Blood lactose after dairy product intake in healthy men. Br. J. Nutr. 118, 1070–1077 (2017).
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D & R Development Core Team. nlme: Linear and nonlinear mixed effects models. R package version 3.1-117 http://CRAN.R-project.org/package=nlme (2014).
Bolker, B. M. coefplot2: Coefficient Plots. R package version 0.1.3.3 http://r-forge.r-project.org/R/?group_id=1059 (2012).
Wickham, H. & Chang, W. ggplot2: An implementation of the grammar of graphics. R package version 1 http://CRAN.R-project.org/package=ggplot2 (2015).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
Caspi, R. et al. The MetaCyc database of metabolic pathways and enzymes and the BioCyc collection of pathway/genome databases. Nucleic Acids Res. 38, D473–D479 (2010).
Jones, E et al. SciPy: Open source scientific tools for python, http://www.scipy.org/ (2001).
Lemon, J. et al. plotrix: Various plotting functions. R package version 3.7 https://cran.r-project.org/package=plotrix (2007).
Robinson, O., Dylus, D. & Dessimoz, C. Phylo.io: interactive viewing and comparison of large phylogenetic trees on the web. Mol. Biol. Evol. 33, 2163–2166 (2016).
Kolde, R. Pheatmap: pretty heatmaps. R package version 61 (2012).
Fujisaka, S. et al. Diet, genetics, and the gut microbiome drive dynamic changes in plasma metabolites. Cell Rep. 22, 3072–3086 (2018).
R Development Core Team R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2013).
Wickham, H. & Francois, R. dplyr: A grammar of data manipulation. R package version 0.4.3 (2015).
Murrell, P. The grid graphics package. R. News 2, 14–19 (2002).
Wickham, H. reshape2: Flexibly reshape data: a reboot of the reshape package. R package version 1 (2012).
This work was funded by the Synthetic Biology platform at the Wyss Institute for Biologically Inspired Engineering at Harvard University; National Institutes of Health (NIH)/National Human Genome Research Institute grant T32 HG002295 (to J.M.L.; principal investigator: P. J. Park); NIH/National Institute of Diabetes and Digestive and Kidney Diseases grant T32 DK007260 (to T.A.C.; principal investigator: T. K. Blackwell); a National Science Foundation Graduate Research Fellowship Program fellowship (to J.M.L.); National Library of Medicine BIRT grant T15LM007092 (to B.T.T.); an AWS Research Credits for Education Grant (to J.M.L. and A.D.K.); a Smith Family Foundation Award for Excellence in Biomedical Research (to A.D.K.); an American Diabetes Association Pathway to Stop Diabetes Initiator Award (to A.D.K.) and NIH/National Institute of Diabetes and Digestive and Kidney Diseases Diabetes Research Center grant P30DK036836-30 (to J.M.L., T.A.C., T.M., B.T.T., L.-D.P., Z.Y., M.C.W., S.L. and A.D.K.; principal investigator: G. L. King). We acknowledge C. J. Patel and S. R. Stein for statistical advice, and S. Softic for assistance with tail vein injections.
J.S. and G.M.C. are co-founders of FitBiomics. Along with A.D.K., they hold equity in FitBiomics.
Peer review information: Joao Monteiro was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended Data Fig. 1 Statistical validations of association between Veillonella abundance and marathon running.
a, Histogram of P values (Wald Z-tests) for time coefficient from LOOCV models predicting 16S Veillonella abundance. The red line represents the P value for the model trained without any hold outs. b, Histogram of P values for time coefficients from 1,000 label permutations in GLMM models predicting Veillonella relative abundance. The red line represents the P value for the model trained without any label permutation.
a, 16S composition in control subjects. b, Veillonella relative abundance in control subjects.
Extended Data Fig. 3 Statistical validations of the association between Veillonella gavage and mouse treadmill run time.
a, Density plot of maximum run times in the AB/BA crossover study. A two-sided Shapiro–Wilk normality test on the maximum run times for each mouse in each treatment group resulted in P = 0.67, with the null hypothesis that the distribution of data is normal (n = 64). b, 95% confidence intervals for the coefficient effect on treadmill run time in AB/BA crossover (Wald Z-tests, n = 64). Center values are the regression estimate for each coefficient. Error bars represent the 95% confidence interval. c, Histogram of P values for the treatment coefficient from LOOCV models predicting treadmill run time. The red line represents the P value for the model trained without any hold outs (Wald Z-tests, n = 64). d, Histogram of P values for the treatment coefficient from 1,000 label permutations in GLMM models predicting treadmill run time. The red line represents the P value for the model trained without any label permutation (Wald Z-tests, n = 64 per permutation).
Each of the 32 facets (each representing an individual mouse) has six longitudinal treadmill run times plotted (three pre- and three post-treatment crossover). The shapes of the points represent the treatment sequence. Each mouse facet has two horizontal lines showing the mean run time when dosed with L. bulgaricus (light blue) or V. atypica (light red). Each facet has a GLMM fit to all data in a treatment sequence (green), a LOOCV GLMM fit trained on all mice except for the mouse the facet represents (red), and a GLMM fit showing the change in intercept related to random effect for each mouse (blue).
Difference in maximum run time between V. atypica and L. Bulgaricus gavage treatment periods, segregated into ‘responders’ and ‘non-responders’ to V. atypica treatment (n = 32).
a,b, Cytokines after V. atypica and L. bulgaricus gavage. Each mouse sample is represented as an individual point, with the central bar representing the mean and error bars representing s.e.m. (n = 64, 32 and 32 for baseline, L. bulgaricus and V. atypica, respectively). c,d, Cytokines after intrarectal propionate instillation. Each mouse sample is represented as an individual point, with the central bar representing the mean and error bars representing s.e.m. (n = 32, 16 and 16 for baseline, L. bulgaricus and V. atypica, respectively). P values were determined by one-way ANOVA followed by Tukey’s posthoc test.
a, Representative section of western blot showing GLUT4 abundance in pre-exercise states, as well as following L. bulgaricus and V. atypica gavage. A stain-free control was used to normalize the densitometry analysis shown. The experiment was performed once (n = 8). b, Fold-change in GLUT4 abundance. Each point represents an individual mouse sample, the centre bar represents the mean and error bars represent s.e.m. (n = 8).
a, Fraction of putative Veillonella relative abundance from metagenomics (calculated utilizing MetaPhlAn2) before and after exercise in rowers and runners. b, Significant alleles (calculated from pairwise ANOVA) that are present in each of the 87 samples. c, The aforementioned 396 significant alleles segregated by exercise state and sample. d, Histogram comparing non-redundant gene family size and annotation fraction.
Enzyme-resolution, log-transformed relative abundances of differentially abundant non-redundant gene families mapped by EC ID to methylmalonyl-CoA pathway components. a, Pathway in aggregate. b–i, Individual reactions in the pathway (n = 8). Data are represented as violin plots, which display the distribution of data as a rotated kernel density distribution.
a, Mice were gavaged either V. atypica or L. bulgaricus and, 5 h later, injected with sodium lactate (750 mg kg−1). Blood lactate was measured 5 min postinjection and every subsequent 10 min (n = 8). Points are means ± s.e.m. b, Area under the curve (AUC) was determined for each mouse and compared between treatments. Each mouse is represented as an individual point, with the central bar representing the mean and error bars representing s.e.m. (P = 0.72 by two-sided unpaired t-test, n = 8).
Unprocessed western blot