Diet rapidly and reproducibly alters the human gut microbiome

Journal name:
Nature
Volume:
505,
Pages:
559–563
Date published:
DOI:
doi:10.1038/nature12820
Received
Accepted
Published online

Long-term dietary intake influences the structure and activity of the trillions of microorganisms residing in the human gut1, 2, 3, 4, 5, but it remains unclear how rapidly and reproducibly the human gut microbiome responds to short-term macronutrient change. Here we show that the short-term consumption of diets composed entirely of animal or plant products alters microbial community structure and overwhelms inter-individual differences in microbial gene expression. The animal-based diet increased the abundance of bile-tolerant microorganisms (Alistipes, Bilophila and Bacteroides) and decreased the levels of Firmicutes that metabolize dietary plant polysaccharides (Roseburia, Eubacterium rectale and Ruminococcus bromii). Microbial activity mirrored differences between herbivorous and carnivorous mammals2, reflecting trade-offs between carbohydrate and protein fermentation. Foodborne microbes from both diets transiently colonized the gut, including bacteria, fungi and even viruses. Finally, increases in the abundance and activity of Bilophila wadsworthia on the animal-based diet support a link between dietary fat, bile acids and the outgrowth of microorganisms capable of triggering inflammatory bowel disease6. In concert, these results demonstrate that the gut microbiome can rapidly respond to altered diet, potentially facilitating the diversity of human dietary lifestyles.

At a glance

Figures

  1. Short-term diet alters the gut microbiota.
    Figure 1: Short-term diet alters the gut microbiota.

    ae, Ten subjects were tracked across each diet arm. a, Fibre intake on the plant-based diet rose from a median baseline value of 9.3±2.1 to 25.6±1.1g per 1,000kcal (P = 0.007; two-sided Wilcoxon signed-rank test), but was negligible on the animal-based diet (P = 0.005). b, Daily fat intake doubled on the animal-based diet from a baseline of 32.5±2.2% to 69.5±0.4% kcal (P = 0.005), but dropped on the plant-based diet to 22.1±1.7% kcal (P = 0.02). c, Protein intake rose on the animal-based diet to 30.1±0.5% kcal from a baseline level of 16.2±1.3% kcal (P = 0.005), and decreased on the plant-based diet to 10.0±0.3% kcal (P = 0.005). d, Within-sample species diversity (α diversity, Shannon diversity index), did not significantly change during either diet. e, The similarity of each individual’s gut microbiota to their baseline communities (β diversity, Jensen–Shannon distance) decreased on the animal-based diet (dates with q<0.05 identified with asterisks; Bonferroni-corrected, two-sided Mann–Whitney U test). Community differences were apparent 1 day after a tracing dye showed the animal-based diet reached the gut (blue arrows depict appearance of food dyes added to first and last diet day meals; Extended Data Fig. 3a).

  2. Bacterial cluster responses to diet arms.
    Figure 2: Bacterial cluster responses to diet arms.

    Cluster log2 fold changes on each diet arm were computed relative to baseline samples across all subjects and are drawn as circles. Clusters with significant (Sig.) fold changes on the animal-based diet are coloured in red, and clusters with significant fold changes on both the plant- and animal-based diets are coloured in both red and green. Uncoloured clusters exhibited no significant (Insig.) fold change on either the animal- or plant-based diet (q<0.05, two-sided Wilcoxon signed-rank test). Bacterial membership in the clusters with the three largest positive and negative fold changes on the animal-based diet are also displayed and coloured by phylum: Firmicutes (purple), Bacteroidetes (blue), Proteobacteria (green), Tenericutes (red) and Verrucomicrobia (grey). Multiple operational taxonomic units (OTUs) with the same name are counted in parentheses.

  3. Diet alters microbial activity and gene expression.
    Figure 3: Diet alters microbial activity and gene expression.

    a, b, Faecal concentrations of SCFAs from carbohydrate (a) and amino acid (b) fermentation (*P<0.05, two-sided Mann–Whitney U test; n = 9–11 faecal samples per diet arm; Supplementary Table 11). ce, The animal-based diet was associated with significant increases in gene expression (normalized to reads per kilobase per million mapped (RPKM); n = 13–21 data sets per diet arm) among glutamine amidotransferases (KEGG orthologous group K08681, vitamin B6 metabolism) (c), methyltransferases (K00599, polycyclic aromatic hydrocarbon degradation) (d) and β-lactamases (K01467) (e). f, Hierarchical clustering of gut microbial gene expression profiles collected on the animal-based (red) and plant-based (green) diets. Expression profile similarity was significantly associated with diet (P<0.003; two-sided Fisher’s exact test excluding replicate samples), despite inter-individual variation that preceded the diet (Extended Data Fig. 6a, b). g, h, Enrichment on animal-based diet (red) and plant-based diet (green) for expression of genes involved in amino acid metabolism (g) and central metabolism (h). Numbers indicate the mean fold change between the two diets for each KEGG orthologous group assigned to a given enzymatic reaction (Supplementary Table 17). Enrichment patterns on the animal- and plant-based diets agree perfectly with patterns observed in carnivorous and herbivorous mammals, respectively2 (P<0.001, Binomial test). GDH, glutamate dehydrogenase; Glu Dx, glutamate decarboxylase; ODx, oxaloacetate decarboxylase; PEPCx, phosphoenolpyruvate carboxylase; PEPCk, PEP carboxykinase; PPDk, pyruvate, orthophosphate dikinase; PTS, phosphotransferase system; Pyr Cx, pyruvate carboxylase; SSADH, succinate-semialdehyde dehydrogenase. Note that Pyr Cx is represented by two groups, which showed divergent fold changes. ch, *P<0.05, Student’s t-test. Values in panels ae are mean±standard error of the mean (s.e.m.).

  4. Foodborne microbes are detectable in the distal gut.
    Figure 4: Foodborne microbes are detectable in the distal gut.

    a, Common bacteria and fungi associated with the animal-based diet menu items, as measured by 16S rRNA and ITS gene sequencing, respectively. Taxa are identified on the genus (g) and species (s) level. A full list of foodborne fungi and bacteria on the animal-based diet can be found in Supplementary Table 21. Foods on the plant-based diet were dominated by matches to the Streptophyta, which derive from chloroplasts within plant matter (Extended Data Fig. 7a). be, Faecal RNA transcripts were significantly enriched (q<0.1, Kruskal–Wallis test; n = 6–10 samples per diet arm) for several food-associated microbes on the animal-based diet relative to baseline (BL) periods, including Lactococcus lactis (b), Staphylococcus carnosus (c), Pediococcus acidilactici (d) and a Penicillium sp. (e). A complete table of taxa with significant expression differences can be found in Supplementary Table 22. f, Fungal concentrations in faeces before and 1–2 days after the animal-based diet were also measured using culture media selective for fungal growth (plate count agar with milk, salt and chloramphenicol). Post-diet faecal samples exhibit significantly higher fungal concentrations than baseline samples (P<0.02; two-sided Mann–Whitney U test; n = 7–10 samples per diet arm). c.f.u., colony-forming units. g, Rubus chlorotic mottle virus transcripts increase on the plant-based diet (q<0.1, Kruskal–Wallis test; n = 6–10 samples per diet arm). bg, Bar charts all display mean±s.e.m.

  5. Changes in the faecal concentration of bile acids and biomarkers for Bilophila on the animal-based diet.
    Figure 5: Changes in the faecal concentration of bile acids and biomarkers for Bilophila on the animal-based diet.

    a, DCA, a secondary bile acid known to promote DNA damage and hepatic carcinomas26, accumulates significantly on the animal-based diet (P<0.01, two-sided Wilcoxon signed-rank test; see Supplementary Table 23 for the diet response of other secondary bile acids). b, RNA-seq data also supports increased microbial metabolism of bile acids on the animal-based diet, as we observed significantly increased expression of microbial bile salt hydrolases (K01442) during that diet arm (*q<0.05, **q<0.01, Kruskal–Wallis test; normalized to RPKM; n = 8–21 samples per diet arm). c, Total faecal bile acid concentrations also increase significantly on the animal-based diet, relative to the preceding baseline period (P<0.05, two-sided Wilcoxon signed-rank test), but do not change on the plant-based diet (Extended Data Fig. 9). Bile acids have been shown to cause inflammatory bowel disease in mice by stimulating the growth of the bacterium Bilophila6, which is known to reduce sulphite to hydrogen sulphide via the sulphite reductase enzyme DsrA (Extended Data Fig. 10). d, e, Quantitative polymerase chain reaction (PCR) showed a significant increase in microbial DNA coding for dsrA on the animal-based diet (P<0.05; two-sided Wilcoxon signed-rank test) (d), and RNA-seq identified a significant increase in sulphite reductase expression (*q<0.05, **q<0.01, Kruskal–Wallis test; n = 8-21 samples/diet arm) (e). b, e, Bar graphs display mean±s.e.m.

  6. Study design.
    Extended Data Fig. 1: Study design.

    a, b, The plant-based (a) and animal-based (b) diets were fed to subjects for five consecutive days. All dates are defined relative to the start of these diet arms (day 0). Study volunteers were observed for 4 days before each diet (the baseline period, days −4 to −1) and for 6 days after each diet arm (the washout period, days 5 to 10) in order to measure subjects’ eating habits and assess their recovery from each diet arm. Subjects were instructed to eat normally during both the baseline and washout periods. Stool samples were collected daily on both diet arms and 16S rRNA and fungal ITS sequencing was performed on all available samples. Subjects also kept daily diet logs. Several analyses (RNA-seq, SCFAs and bile acids) were performed primarily using only two samples per person per diet (that is, a baseline and diet arm comparison). Comparative sampling did not always occur using exactly the same study days owing to limited sample availability for some subjects. Because we expected the animal-based diet to promote ketogenesis, we only measured urinary ketones on the animal-based diet. To test the hypothesis that microbes from fermented foods on the animal-based diet survived transit through the gastrointestinal tract, we cultured bacteria and fungi before and after the animal-based diet.

  7. A vegetarian/'s microbiota.
    Extended Data Fig. 2: A vegetarian’s microbiota.

    ac, One of the study subjects is a lifelong vegetarian (subject 6). a, Relative abundances of Prevotella and Bacteroides are shown across the plant-based diet for subject 6 (orange circles), as well as for all other subjects (green circles). Consecutive daily samples from subject 6 are linked by dashed lines. For reference, median baseline abundances are depicted using larger circles. b, Relative abundances are also shown for samples taken on the animal-based diet. Labelled points correspond to diet days where subject 6’s gut microbiota exhibited an increase in the relative abundance of Bacteroides. c, A principal-coordinates-based characterization of overall community structure for subject 6, as well as all other subjects. QIIME30 was used to compute microbial β diversity with the Bray–Curtis, unweighted UniFrac and weighted UniFrac statistics. Sample similarities were projected onto two dimensions using principal coordinates analysis. Top, when coloured by subject, samples from subject 6 (green triangles) partition apart from the other subjects’ samples. Bottom, of all of subject 6’s diet samples, the ones most similar to the other subjects’ are the samples taken while consuming the animal-based diet.

  8. Subject physiology across diet arms.
    Extended Data Fig. 3: Subject physiology across diet arms.

    a, Gastrointestinal motility, as measured by the initial appearance of a non-absorbable dye added to the first and last lunch of each diet. The median time until dye appearance is indicated with red arrows. Subject motility was significantly lower (P<0.05, Mann–Whitney U test) on the animal-based diet (median transit time of 1.5 days) than on the plant-based one (1.0 days). b, Range (shaded boxes) and median (solid line) of subjects’ weights over time. Subjects’ weight did not change significantly on the plant-based diet relative to baseline periods, but did decrease significantly on the animal-based diet (asterisks denote q<0.05, Bonferroni-corrected Mann–Whitney U test). Subjects lost a median of 1.6% and 2.5% of body weight by days 3 and 4, respectively, of the animal-based diet arm. c, Measurements of subjects’ urinary ketone levels. Individual subjects are shown with black dots, and median values are connected with a black solid line. Urinary ketone readings were taken from day 0 of the animal-based diet onwards. Ketone levels were compared to the readings on day 0, and asterisks denote days with significant ketone increases (q<0.05, Bonferroni-corrected Mann–Whitney U test; significance tests were not carried out for days on which less than four subjects reported their readings.). All subjects on the animal-based diet showed elevated levels of ketones in their urine by day 2 of the diet (≥15mgdl−1 as compared to 0mgdl−1 during initial readings), indicating that they experienced ketonuria during the diet arm. This metabolic state is characterized by the restricted availability of glucose and the compensatory extraction of energy from fat tissue56.

  9. Baseline Prevotella abundance is associated with long-term fibre intake.
    Extended Data Fig. 4: Baseline Prevotella abundance is associated with long-term fibre intake.

    Prevotella fractions were computed by summing the fractional 16S rRNA abundance of all OTUs whose genus name was Prevotella. Daily intake of dietary fibre over the previous year was estimated using the Diet History Questionnaire32 (variable name “TOTAL_DIETARY_FIBER_G_NDSR”). There is a significant positive correlation between subjects’ baseline Prevotella abundance and their long-term dietary fibre intake (Spearman’s ρ = 0.78, P = 0.008).

  10. Significant correlations between SCFAs and cluster abundances across subjects.
    Extended Data Fig. 5: Significant correlations between SCFAs and cluster abundances across subjects.

    SCFAs are drawn in rectangles and coloured maroon or green if they are produced from amino acid or carbohydrate fermentation, respectively. Clusters whose members include known bile-tolerant or amino-acid-fermenting bacteria15, 16 are coloured maroon, whereas clusters including known saccharolytic bacteria3 are coloured green. Uncoloured clusters and SCFAs are not associated with saccharolytic or putrefactive pathways. Significant positive and negative correlations are shown with black arrows and grey arrows, respectively (q<0.05; Spearman correlation).

  11. Inter-individual microbial community variation according to diet and sequencing technique.
    Extended Data Fig. 6: Inter-individual microbial community variation according to diet and sequencing technique.

    a, b, To measure the degree to which diet influences inter-individual differences in gut microbial gene expression, we clustered RNA-seq profiles from baseline (a) and diet (b) periods. Dots indicate pairs of samples that cluster by subject. The potential for diet to partition samples was measured by splitting trees at the arrowed branches and testing the significance of the resulting 2×2 contingency table (diet versus partition; Fisher’s exact test). To avoid skewed significance values caused by non-independent samples, we only clustered a single sample per subject, per diet period. In the case of multiple baseline samples, the sample closest to the diet intervention was used. In the case of multiple diet samples, the last sample during the diet intervention was kept. A single sample was randomly chosen if there were multiple samples from the same person on the same day. No association between diet and partitioning was found for partitions I–VI (P>0.05). However, a significant association was observed for partition VII (P = 0.003). c, To determine whether diet affects inter-individual differences in gut microbial community structure, we hierarchically clustered 16S rRNA data from the last day of each diet arm. Samples grouped weakly by diet: sub-trees partitioned at the arrowed node showed a minor enrichment for plant-based diet samples in one sub-tree and animal-based diet samples in the other (P = 0.07; Fisher’s exact test). Still, samples from five subjects grouped by individual, not diet (indicated by black nodes), indicating that diet does not reproducibly overcome inter-individual differences in gut microbial community structure.

  12. Food-associated microbes and their enteric abundance over time.
    Extended Data Fig. 7: Food-associated microbes and their enteric abundance over time.

    a, Major bacterial and fungal taxa found in plant-based diet menu items were determined using 16S rRNA and ITS sequencing, respectively, at the species (s), genus (g) and order level (o). The majority of 16S rRNA gene sequences are Streptophyta, representing chloroplasts from the ingested plant matter. b, One of the fungi from a, Candida sp., showed a significance increase in faecal abundance on the plant-based diet (P<0.05, Wilcoxon signed-rank test). c, Levels of bacteria and fungi associated with the animal-based diet are plotted over the plant- and animal-based diet arms. Taxa are identified on the genus (g) and species (s) level. The abundance of foodborne bacteria was near our detection limit by 16S rRNA gene sequencing; to minimize resulting measurement errors, we have plotted the fraction of samples in which bacteria are present or absent. Lactococcus lactis, Pediococcus acidilactici and Staphylococcus-associated reads all show significantly increased abundance on the animal-based diet (P<0.05, Wilcoxon signed-rank test). Fungal concentrations were measured using ITS sequencing and are plotted in terms of log-fractional abundance. Significant increases in Penicillium-related fungi were observed, along with significant decreases in the concentration of Debaryomyces and a Candida sp. (P<0.05, Wilcoxon signed-rank test). One possible explanation for the surprising decrease in the concentration of food-associated fungi is that the more than tenfold increase in Penicillium levels lowered the relative abundance of all other fungi, even those that increased in terms of absolute abundance.

  13. Eukaryotic and viral taxa detected via RNA-seq.
    Extended Data Fig. 8: Eukaryotic and viral taxa detected via RNA-seq.

    a, Identified plant and other viruses. The most common virus is a DNA virus (lambda phage) and may be an artefact of the sequencing process. b, Identified fungi, protists and other eukaryotes. Taxa that were re-annotated using manually curated BLAST searches are indicated with asterisks and their original taxonomic assignments are shown in parentheses (see Methods for more details).

  14. Faecal bile acid concentrations on baseline, plant- and animal-based diets.
    Extended Data Fig. 9: Faecal bile acid concentrations on baseline, plant- and animal-based diets.

    a, b, Median bulk bile acid concentrations are shown for all individuals on the plant-based (a) and animal-based (b) diets (error bars denote median absolute deviations). For detailed experimental protocols, see Methods. Bile acid levels did not significantly change on the plant-based diet relative to baseline levels (P>0.1, Mann–Whitney U test). However, bile acid levels trended upwards on the animal-based diet, rising from 1.48μmol per 100mg dry stool during the baseline period to 2.37μmol per 100mg dry stool (P<0.10, Mann–Whitney U test).

  15. The dissimilatory sulphate reduction pathway.
    Extended Data Fig. 10: The dissimilatory sulphate reduction pathway.

    a, Microbes reduce sulphate to hydrogen sulphide by first converting sulphate to adenosine 5′-phosphosulphate (APS) via the enzyme ATP sulphurylase (Sat). Next, APS is reduced to sulphite by the enzyme APS reductase (Apr). Finally, the end product hydrogen sulphide is reached by reducing sulphite through the enzyme sulphite reductase (DsrA). This last step of the pathway can be performed by Bilophila and is thought to contribute to intestinal inflammation6. b, No significant changes in apr gene abundance were observed on any diet (P>0.05, Mann–Whitney U test; n = 10 samples per diet arm). Values are mean±s.e.m. However, dsrA abundance increased on the animal-based diet (Fig. 5d). NS, not significant.

Accession codes

Referenced accessions

Gene Expression Omnibus

References

  1. Wu, G. D. et al. Linking long-term dietary patterns with gut microbial enterotypes. Science 334, 105108 (2011)
  2. Muegge, B. D. et al. Diet drives convergence in gut microbiome functions across mammalian phylogeny and within humans. Science 332, 970974 (2011)
  3. Duncan, S. H. et al. Reduced dietary intake of carbohydrates by obese subjects results in decreased concentrations of butyrate and butyrate-producing bacteria in feces. Appl. Environ. Microbiol. 73, 10731078 (2007)
  4. Ley, R. E., Turnbaugh, P. J., Klein, S. & Gordon, J. I. Microbial ecology: human gut microbes associated with obesity. Nature 444, 10221023 (2006)
  5. Walker, A. W. et al. Dominant and diet-responsive groups of bacteria within the human colonic microbiota. ISME J. 5, 220230 (2011)
  6. Devkota, S. et al. Dietary-fat-induced taurocholic acid promotes pathobiont expansion and colitis in Il10−/− mice. Nature 487, 104108 (2012)
  7. Turnbaugh, P. J. et al. The effect of diet on the human gut microbiome: a metagenomic analysis in humanized gnotobiotic mice. Sci. Transl. Med. 1, 6ra14 (2009)
  8. Turnbaugh, P. J. et al. An obesity-associated gut microbiome with increased capacity for energy harvest. Nature 444, 10271031 (2006)
  9. Faith, J. J., McNulty, N. P., Rey, F. E. & Gordon, J. I. Predicting a human gut microbiota’s response to diet in gnotobiotic mice. Science 333, 101104 (2011)
  10. Russell, W. R. et al. High-protein, reduced-carbohydrate weight-loss diets promote metabolite profiles likely to be detrimental to colonic health. Am. J. Clin. Nutr. 93, 10621072 (2011)
  11. Cordain, L. et al. Plant-animal subsistence ratios and macronutrient energy estimations in worldwide hunter-gatherer diets. Am. J. Clin. Nutr. 71, 682692 (2000)
  12. Arumugam, M. et al. Enterotypes of the human gut microbiome. Nature 473, 174180 (2011)
  13. De Filippo, C. et al. Impact of diet in shaping gut microbiota revealed by a comparative study in children from Europe and rural Africa. Proc. Natl Acad. Sci. USA 107, 1469114696 (2010)
  14. Reddy, B. S. Diet and excretion of bile acids. Cancer Res. 41, 37663768 (1981)
  15. Smith, E. A. & Macfarlane, G. T. Enumeration of amino acid fermenting bacteria in the human large intestine: effects of pH and starch on peptide metabolism and dissimilation of amino acids. FEMS Microbiol. Ecol. 25, 355368 (1998)
  16. Smith, E. A. & Macfarlane, G. T. Enumeration of human colonic bacteria producing phenolic and indolic compounds: effects of pH, carbohydrate availability and retention time on dissimilatory aromatic amino acid metabolism. J. Appl. Bacteriol. 81, 288302 (1996)
  17. Sinha, R. et al. High concentrations of the carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) occur in chicken but are dependent on the cooking method. Cancer Res. 55, 45164519 (1995)
  18. Langille, M. G. I. et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene sequences. Nature Biotechnol. 31, 814821 (2013)
  19. Kanehisa, M. & Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 28, 2730 (2000)
  20. Pittard, J. & Wallace, B. J. Distribution and function of genes concerned with aromatic biosynthesis in Escherichia coli. J. Bacteriol. 91, 14941508 (1966)
  21. Hawkes, K., O’Connell, J. F. & Jones, N. G. Hunting income patterns among the Hadza: big game, common goods, foraging goals and the evolution of the human diet. Philos. Trans. R. Soc. Lond. B Biol. Sci. 334, 243250 (1991)
  22. Bourdichon, F., Berger, B. & Casaregola, S. Safety demonstration of microbial food cultures (MFC) in fermented food products. Bull. Int. Dairy Fed. 455, 166 (2012)
  23. Nychas, G. J. & Arkoudelos, J. S. Staphylococci: their role in fermented sausages. Soc. Appl. Bacteriol. Symp. Ser. 19, 167S188S (1990)
  24. McGavin, W. J. & Macfarlane, S. A. Rubus chlorotic mottle virus, a new sobemovirus infecting raspberry and bramble. Virus Res. 139, 1013 (2009)
  25. Zhang, T. et al. RNA viral community in human feces: prevalence of plant pathogenic viruses. PLoS Biol. 4, e3 (2006)
  26. Yoshimoto, S. et al. Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome. Nature 499, 97101 (2013)
  27. Ridlon, J. M., Kang, D. J. & Hylemon, P. B. Bile salt biotransformations by human intestinal bacteria. J. Lipid Res. 47, 241259 (2006)
  28. Islam, K. B. et al. Bile acid is a host factor that regulates the composition of the cecal microbiota in rats. Gastroenterology 141, 17731781 (2011)
  29. Maurice, C. F., Haiser, H. J. & Turnbaugh, P. J. Xenobiotics shape the physiology and gene expression of the active human gut microbiome. Cell 152, 3950 (2013)
  30. Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nature Methods 7, 335336 (2010)
  31. Lewis, S. J. & Heaton, K. W. Stool form scale as a useful guide to intestinal transit time. Scand. J. Gastroenterol. 32, 920924 (1997)
  32. National Institutes of Health Diet History Questionnaire Version 2.0 (National Institutes of Health, Applied Research Program, National Cancer Institute, 2010)
  33. Caporaso, J. G. et al. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl Acad. Sci. USA 108 (suppl. 1). 45164522 (2011)
  34. Caporaso, J. G. et al. Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J. 6, 16211624 (2012)
  35. DeSantis, T. Z. et al. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 72, 50695072 (2006)
  36. Jones, E. et al. SciPy: open source scientific tools for Python. (2001)
  37. McKinney, W. Data structures for statistical computing in Python. Proc. 9th Python Sci. Conf. 5156. (2010)
  38. Strimmer, K. fdrtool: a versatile R package for estimating local and tail areabased false discovery rates. Bioinformatics 24, 14611462 (2008)
  39. Friedman, J. & Alm, E. J. Inferring correlation networks from genomic survey data. PLoS Comput. Biol. 8, e1002687 (2012)
  40. Turnbaugh, P. J. et al. Organismal, genetic, and transcriptional variation in the deeply sequenced gut microbiomes of identical twins. Proc. Natl Acad. Sci. USA 107, 75037508 (2010)
  41. Rey, F. E. et al. Dissecting the in vivo metabolic potential of two human gut acetogens. J. Biol. Chem. 285, 2208222090 (2010)
  42. Nelson, K. E. et al. A catalog of reference genomes from the human microbiome. Science 328, 994999 (2010)
  43. Ning, Z., Cox, A. J. & Mullikin, J. C. SSAHA: a fast search method for large DNA databases. Genome Res. 11, 17251729 (2001)
  44. Abubucker, S. et al. Metabolic reconstruction for metagenomic data and its application to the human microbiome. PLOS Comput. Biol. 8, e1002358 (2012)
  45. Segata, N. et al. Metagenomic biomarker discovery and explanation. Genome Biol. 12, R60 (2011)
  46. Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nature Methods 9, 357359 (2012)
  47. Markowitz, V. M. et al. IMG: the Integrated Microbial Genomes database and comparative analysis system. Nucleic Acids Res. 40, D115D122 (2012)
  48. Martin, J. et al. Optimizing read mapping to reference genomes to determine composition and species prevalence in microbial communities. PLoS ONE 7, e36427 (2012)
  49. Deplancke, B. et al. Molecular ecological analysis of the succession and diversity of sulfate-reducing bacteria in the mouse gastrointestinal tract. Appl. Environ. Microbiol. 66, 21662174 (2000)
  50. Stewart, J. A., Chadwick, V. S. & Murray, A. Carriage, quantification, and predominance of methanogens and sulfate-reducing bacteria in faecal samples. Lett. Appl. Microbiol. 43, 5863 (2006)
  51. Porter, J. L. et al. Accurate enzymatic measurement of fecal bile acids in patients with malabsorption. J. Lab. Clin. Med. 141, 411418 (2003)
  52. Setchell, K. D., Lawson, A. M., Tanida, N. & Sjovall, J. General methods for the analysis of metabolic profiles of bile acids and related compounds in feces. J. Lipid Res. 24, 10851100 (1983)
  53. Schoch, C. L. et al. Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for Fungi. Proc. Natl Acad. Sci. USA 109, 62416246 (2012)
  54. Gardes, M. & Bruns, T. D. ITS primers with enhanced specificity for basidiomycetes application to the identification of mycorrhizae and rusts. Mol. Ecol. 2, 113118 (1993)
  55. White, T. J., Bruns, T., Lee, S. & Taylor, J. in PCR Protocols: A Guide to Methods and Applications (eds Gelfand, D. H., Innis, M. A., Shinsky, J. J. & White, T. J.) 315–322. (1990)
  56. Walker, H. K., Hall, W. D., Hurst, J. W., Comstock, J. P. & Garber, A. J. Ketonuria 3rd edn (Butterworths, 1990)

Download references

Author information

Affiliations

  1. FAS Center for Systems Biology, Harvard University, Cambridge, Massachusetts 02138, USA

    • Lawrence A. David,
    • Corinne F. Maurice,
    • Rachel N. Carmody,
    • David B. Gootenberg,
    • Julie E. Button,
    • Benjamin E. Wolfe,
    • Rachel J. Dutton &
    • Peter J. Turnbaugh
  2. Society of Fellows, Harvard University, Cambridge, Massachusetts 02138, USA

    • Lawrence A. David
  3. Division of Endocrinology, Children’s Hospital Boston, Harvard Medical School, Boston, Massachusetts 02115, USA

    • Alisha V. Ling &
    • Sudha B. Biddinger
  4. Department of Bioengineering & Therapeutic Sciences and the California Institute for Quantitative Biosciences, University of California, San Francisco, San Francisco, California 94158, USA

    • A. Sloan Devlin,
    • Yug Varma &
    • Michael A. Fischbach
  5. Present address: Molecular Genetics & Microbiology and Institute for Genome Sciences & Policy, Duke University, Durham, North Carolina 27708, USA.

    • Lawrence A. David

Contributions

L.A.D., R.J.D. and P.J.T. designed the study, and developed and prepared the diets. L.A.D., C.F.M., R.N.C., D.B.G., J.E.B., B.E.W. and P.J.T. performed the experimental work. A.V.L., A.S.D., Y.V., M.A.F. and S.B.B. conducted bile acid analyses. L.A.D. and P.J.T. performed computational analyses. L.A.D. and P.J.T. prepared the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

RNA-seq data have been deposited in the Gene Expression Omnibus under accession GSE46761; 16S and ITS rRNA gene sequencing reads have been deposited in MG-RAST under accession 6248.

Author details

Extended data figures and tables

Extended Data Figures

  1. Extended Data Figure 1: Study design. (435 KB)

    a, b, The plant-based (a) and animal-based (b) diets were fed to subjects for five consecutive days. All dates are defined relative to the start of these diet arms (day 0). Study volunteers were observed for 4 days before each diet (the baseline period, days −4 to −1) and for 6 days after each diet arm (the washout period, days 5 to 10) in order to measure subjects’ eating habits and assess their recovery from each diet arm. Subjects were instructed to eat normally during both the baseline and washout periods. Stool samples were collected daily on both diet arms and 16S rRNA and fungal ITS sequencing was performed on all available samples. Subjects also kept daily diet logs. Several analyses (RNA-seq, SCFAs and bile acids) were performed primarily using only two samples per person per diet (that is, a baseline and diet arm comparison). Comparative sampling did not always occur using exactly the same study days owing to limited sample availability for some subjects. Because we expected the animal-based diet to promote ketogenesis, we only measured urinary ketones on the animal-based diet. To test the hypothesis that microbes from fermented foods on the animal-based diet survived transit through the gastrointestinal tract, we cultured bacteria and fungi before and after the animal-based diet.

  2. Extended Data Figure 2: A vegetarian’s microbiota. (359 KB)

    ac, One of the study subjects is a lifelong vegetarian (subject 6). a, Relative abundances of Prevotella and Bacteroides are shown across the plant-based diet for subject 6 (orange circles), as well as for all other subjects (green circles). Consecutive daily samples from subject 6 are linked by dashed lines. For reference, median baseline abundances are depicted using larger circles. b, Relative abundances are also shown for samples taken on the animal-based diet. Labelled points correspond to diet days where subject 6’s gut microbiota exhibited an increase in the relative abundance of Bacteroides. c, A principal-coordinates-based characterization of overall community structure for subject 6, as well as all other subjects. QIIME30 was used to compute microbial β diversity with the Bray–Curtis, unweighted UniFrac and weighted UniFrac statistics. Sample similarities were projected onto two dimensions using principal coordinates analysis. Top, when coloured by subject, samples from subject 6 (green triangles) partition apart from the other subjects’ samples. Bottom, of all of subject 6’s diet samples, the ones most similar to the other subjects’ are the samples taken while consuming the animal-based diet.

  3. Extended Data Figure 3: Subject physiology across diet arms. (348 KB)

    a, Gastrointestinal motility, as measured by the initial appearance of a non-absorbable dye added to the first and last lunch of each diet. The median time until dye appearance is indicated with red arrows. Subject motility was significantly lower (P<0.05, Mann–Whitney U test) on the animal-based diet (median transit time of 1.5 days) than on the plant-based one (1.0 days). b, Range (shaded boxes) and median (solid line) of subjects’ weights over time. Subjects’ weight did not change significantly on the plant-based diet relative to baseline periods, but did decrease significantly on the animal-based diet (asterisks denote q<0.05, Bonferroni-corrected Mann–Whitney U test). Subjects lost a median of 1.6% and 2.5% of body weight by days 3 and 4, respectively, of the animal-based diet arm. c, Measurements of subjects’ urinary ketone levels. Individual subjects are shown with black dots, and median values are connected with a black solid line. Urinary ketone readings were taken from day 0 of the animal-based diet onwards. Ketone levels were compared to the readings on day 0, and asterisks denote days with significant ketone increases (q<0.05, Bonferroni-corrected Mann–Whitney U test; significance tests were not carried out for days on which less than four subjects reported their readings.). All subjects on the animal-based diet showed elevated levels of ketones in their urine by day 2 of the diet (≥15mgdl−1 as compared to 0mgdl−1 during initial readings), indicating that they experienced ketonuria during the diet arm. This metabolic state is characterized by the restricted availability of glucose and the compensatory extraction of energy from fat tissue56.

  4. Extended Data Figure 4: Baseline Prevotella abundance is associated with long-term fibre intake. (239 KB)

    Prevotella fractions were computed by summing the fractional 16S rRNA abundance of all OTUs whose genus name was Prevotella. Daily intake of dietary fibre over the previous year was estimated using the Diet History Questionnaire32 (variable name “TOTAL_DIETARY_FIBER_G_NDSR”). There is a significant positive correlation between subjects’ baseline Prevotella abundance and their long-term dietary fibre intake (Spearman’s ρ = 0.78, P = 0.008).

  5. Extended Data Figure 5: Significant correlations between SCFAs and cluster abundances across subjects. (189 KB)

    SCFAs are drawn in rectangles and coloured maroon or green if they are produced from amino acid or carbohydrate fermentation, respectively. Clusters whose members include known bile-tolerant or amino-acid-fermenting bacteria15, 16 are coloured maroon, whereas clusters including known saccharolytic bacteria3 are coloured green. Uncoloured clusters and SCFAs are not associated with saccharolytic or putrefactive pathways. Significant positive and negative correlations are shown with black arrows and grey arrows, respectively (q<0.05; Spearman correlation).

  6. Extended Data Figure 6: Inter-individual microbial community variation according to diet and sequencing technique. (377 KB)

    a, b, To measure the degree to which diet influences inter-individual differences in gut microbial gene expression, we clustered RNA-seq profiles from baseline (a) and diet (b) periods. Dots indicate pairs of samples that cluster by subject. The potential for diet to partition samples was measured by splitting trees at the arrowed branches and testing the significance of the resulting 2×2 contingency table (diet versus partition; Fisher’s exact test). To avoid skewed significance values caused by non-independent samples, we only clustered a single sample per subject, per diet period. In the case of multiple baseline samples, the sample closest to the diet intervention was used. In the case of multiple diet samples, the last sample during the diet intervention was kept. A single sample was randomly chosen if there were multiple samples from the same person on the same day. No association between diet and partitioning was found for partitions I–VI (P>0.05). However, a significant association was observed for partition VII (P = 0.003). c, To determine whether diet affects inter-individual differences in gut microbial community structure, we hierarchically clustered 16S rRNA data from the last day of each diet arm. Samples grouped weakly by diet: sub-trees partitioned at the arrowed node showed a minor enrichment for plant-based diet samples in one sub-tree and animal-based diet samples in the other (P = 0.07; Fisher’s exact test). Still, samples from five subjects grouped by individual, not diet (indicated by black nodes), indicating that diet does not reproducibly overcome inter-individual differences in gut microbial community structure.

  7. Extended Data Figure 7: Food-associated microbes and their enteric abundance over time. (636 KB)

    a, Major bacterial and fungal taxa found in plant-based diet menu items were determined using 16S rRNA and ITS sequencing, respectively, at the species (s), genus (g) and order level (o). The majority of 16S rRNA gene sequences are Streptophyta, representing chloroplasts from the ingested plant matter. b, One of the fungi from a, Candida sp., showed a significance increase in faecal abundance on the plant-based diet (P<0.05, Wilcoxon signed-rank test). c, Levels of bacteria and fungi associated with the animal-based diet are plotted over the plant- and animal-based diet arms. Taxa are identified on the genus (g) and species (s) level. The abundance of foodborne bacteria was near our detection limit by 16S rRNA gene sequencing; to minimize resulting measurement errors, we have plotted the fraction of samples in which bacteria are present or absent. Lactococcus lactis, Pediococcus acidilactici and Staphylococcus-associated reads all show significantly increased abundance on the animal-based diet (P<0.05, Wilcoxon signed-rank test). Fungal concentrations were measured using ITS sequencing and are plotted in terms of log-fractional abundance. Significant increases in Penicillium-related fungi were observed, along with significant decreases in the concentration of Debaryomyces and a Candida sp. (P<0.05, Wilcoxon signed-rank test). One possible explanation for the surprising decrease in the concentration of food-associated fungi is that the more than tenfold increase in Penicillium levels lowered the relative abundance of all other fungi, even those that increased in terms of absolute abundance.

  8. Extended Data Figure 8: Eukaryotic and viral taxa detected via RNA-seq. (365 KB)

    a, Identified plant and other viruses. The most common virus is a DNA virus (lambda phage) and may be an artefact of the sequencing process. b, Identified fungi, protists and other eukaryotes. Taxa that were re-annotated using manually curated BLAST searches are indicated with asterisks and their original taxonomic assignments are shown in parentheses (see Methods for more details).

  9. Extended Data Figure 9: Faecal bile acid concentrations on baseline, plant- and animal-based diets. (94 KB)

    a, b, Median bulk bile acid concentrations are shown for all individuals on the plant-based (a) and animal-based (b) diets (error bars denote median absolute deviations). For detailed experimental protocols, see Methods. Bile acid levels did not significantly change on the plant-based diet relative to baseline levels (P>0.1, Mann–Whitney U test). However, bile acid levels trended upwards on the animal-based diet, rising from 1.48μmol per 100mg dry stool during the baseline period to 2.37μmol per 100mg dry stool (P<0.10, Mann–Whitney U test).

  10. Extended Data Figure 10: The dissimilatory sulphate reduction pathway. (109 KB)

    a, Microbes reduce sulphate to hydrogen sulphide by first converting sulphate to adenosine 5′-phosphosulphate (APS) via the enzyme ATP sulphurylase (Sat). Next, APS is reduced to sulphite by the enzyme APS reductase (Apr). Finally, the end product hydrogen sulphide is reached by reducing sulphite through the enzyme sulphite reductase (DsrA). This last step of the pathway can be performed by Bilophila and is thought to contribute to intestinal inflammation6. b, No significant changes in apr gene abundance were observed on any diet (P>0.05, Mann–Whitney U test; n = 10 samples per diet arm). Values are mean±s.e.m. However, dsrA abundance increased on the animal-based diet (Fig. 5d). NS, not significant.

Supplementary information

PDF files

  1. Supplementary Information (132 KB)

    This file contains the Supplementary Discussion.

Excel files

  1. Supplementary Data (133 KB)

    This file contains Supplementary Tables 1-23.

Additional data