Gut dysbiosis might underlie the pathogenesis of type 1 diabetes. In mice of the non-obese diabetic (NOD) strain, we found that key features of disease correlated inversely with blood and fecal concentrations of the microbial metabolites acetate and butyrate. We therefore fed NOD mice specialized diets designed to release large amounts of acetate or butyrate after bacterial fermentation in the colon. Each diet provided a high degree of protection from diabetes, even when administered after breakdown of immunotolerance. Feeding mice a combined acetate- and butyrate-yielding diet provided complete protection, which suggested that acetate and butyrate might operate through distinct mechanisms. Acetate markedly decreased the frequency of autoreactive T cells in lymphoid tissues, through effects on B cells and their ability to expand populations of autoreactive T cells. A diet containing butyrate boosted the number and function of regulatory T cells, whereas acetate- and butyrate-yielding diets enhanced gut integrity and decreased serum concentration of diabetogenic cytokines such as IL-21. Medicinal foods or metabolites might represent an effective and natural approach for countering the numerous immunological defects that contribute to T cell–dependent autoimmune diseases.
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.
Gene Expression Omnibus
Atkinson, M.A., Eisenbarth, G.S. & Michels, A.W. Type 1 diabetes. Lancet 383, 69–82 (2014).
Vatanen, T. et al. Variation in microbiome LPS immunogenicity contributes to autoimmunity in humans. Cell 165, 842–853 (2016).
Livanos, A.E. et al. Antibiotic-mediated gut microbiome perturbation accelerates development of type 1 diabetes in mice. Nat. Microbiol. 1, 16140 (2016).
Thorburn, A.N., Macia, L. & Mackay, C.R. Diet, metabolites, and “western-lifestyle” inflammatory diseases. Immunity 40, 833–842 (2014).
Maslowski, K.M. & Mackay, C.R. Diet, gut microbiota and immune responses. Nat. Immunol. 12, 5–9 (2011).
Wen, L. et al. Innate immunity and intestinal microbiota in the development of type 1 diabetes. Nature 455, 1109–1113 (2008).
Markle, J.G. et al. Sex differences in the gut microbiome drive hormone-dependent regulation of autoimmunity. Science 339, 1084–1088 (2013).
Smith, P.M. et al. The microbial metabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science 341, 569–573 (2013).
Furusawa, Y. et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature 504, 446–450 (2013).
Arpaia, N. et al. Metabolites produced by commensal bacteria promote peripheral regulatory T-cell generation. Nature 504, 451–455 (2013).
Fukuda, S. et al. Bifidobacteria can protect from enteropathogenic infection through production of acetate. Nature 469, 543–547 (2011).
Maslowski, K.M. et al. Regulation of inflammatory responses by gut microbiota and chemoattractant receptor GPR43. Nature 461, 1282–1286 (2009).
Zhang, Y. et al. Genome-wide DNA methylation analysis identifies hypomethylated genes regulated by FOXP3 in human regulatory T cells. Blood 122, 2823–2836 (2013).
Sakaguchi, S. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat. Immunol. 6, 345–352 (2005).
Turley, S.J., Lee, J.W., Dutton-Swain, N., Mathis, D. & Benoist, C. Endocrine self and gut non-self intersect in the pancreatic lymph nodes. Proc. Natl. Acad. Sci. USA 102, 17729–17733 (2005).
Vaarala, O., Atkinson, M.A. & Neu, J. The “perfect storm” for type 1 diabetes: the complex interplay between intestinal microbiota, gut permeability, and mucosal immunity. Diabetes 57, 2555–2562 (2008).
de Goffau, M.C. et al. Fecal microbiota composition differs between children with beta-cell autoimmunity and those without. Diabetes 62, 1238–1244 (2013).
Endesfelder, D. et al. Compromised gut microbiota networks in children with anti-islet cell autoimmunity. Diabetes 63, 2006–2014 (2014).
Tan, J. et al. Dietary fiber and bacterial SCFA enhance oral tolerance and protect against food allergy through diverse cellular pathways. Cell Rep. 15, 2809–2824 (2016).
Topping, D.L. & Clifton, P.M. Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides. Physiol. Rev. 81, 1031–1064 (2001).
Clarke, J.M., Topping, D.L., Bird, A.R., Young, G.P. & Cobiac, L. Effects of high-amylose maize starch and butyrylated high-amylose maize starch on azoxymethane-induced intestinal cancer in rats. Carcinogenesis 29, 2190–2194 (2008).
Perry, R.J. et al. Acetate mediates a microbiome-brain-beta-cell axis to promote metabolic syndrome. Nature 534, 213–217 (2016).
Amrani, A. et al. Progression of autoimmune diabetes driven by avidity maturation of a T-cell population. Nature 406, 739–742 (2000).
Marino, E., Tan, B., Binge, L., Mackay, C.R. & Grey, S.T. B-cell cross-presentation of autologous antigen precipitates diabetes. Diabetes 61, 2893–2905 (2012).
Brummel, R., Roberts, T.L., Stacey, K.J. & Lenert, P. Higher-order CpG-DNA stimulation reveals distinct activation requirements for marginal zone and follicular B cells in lupus mice. Eur. J. Immunol. 36, 1951–1962 (2006).
van den Elsen, P.J., Holling, T.M., Kuipers, H.F. & van der Stoep, N. Transcriptional regulation of antigen presentation. Curr. Opin. Immunol. 16, 67–75 (2004).
Martins, G. & Calame, K. Regulation and functions of Blimp-1 in T and B lymphocytes. Annu. Rev. Immunol. 26, 133–169 (2008).
Saemann, M.D. et al. Bacterial metabolite interference with maturation of human monocyte-derived dendritic cells. J. Leukoc. Biol. 71, 238–246 (2002).
Thorburn, A.N. et al. Evidence that asthma is a developmental origin disease influenced by maternal diet and bacterial metabolites. Nat. Commun. 6, 7320 (2015).
Vignali, D.A., Collison, L.W. & Workman, C.J. How regulatory T cells work. Nat. Rev. Immunol. 8, 523–532 (2008).
Shi, G. et al. Free fatty acid receptor 2, a candidate target for type 1 diabetes, induces cell apoptosis through ERK signaling. J. Mol. Endocrinol. 53, 367–380 (2014).
Macia, L. et al. Metabolite-sensing receptors GPR43 and GPR109A facilitate dietary fibre-induced gut homeostasis through regulation of the inflammasome. Nat. Commun. 6, 6734 (2015).
Sutherland, A.P. et al. Interleukin-21 is required for the development of type 1 diabetes in NOD mice. Diabetes 58, 1144–1155 (2009).
Spolski, R., Kashyap, M., Robinson, C., Yu, Z. & Leonard, W.J. IL-21 signaling is critical for the development of type I diabetes in the NOD mouse. Proc. Natl. Acad. Sci. USA 105, 14028–14033 (2008).
Feehily, C. & Karatzas, K.A. Role of glutamate metabolism in bacterial responses towards acid and other stresses. J. Appl. Microbiol. 114, 11–24 (2013).
Marino, E. et al. CD4+CD25+ T-cells control autoimmunity in the absence of B-cells. Diabetes 58, 1568–1577 (2009).
Buhlmann, J.E. et al. In the absence of a CD40 signal, B cells are tolerogenic. Immunity 2, 645–653 (1995).
Kostic, A.D. et al. The dynamics of the human infant gut microbiome in development and in progression toward type 1 diabetes. Cell Host Microbe 17, 260–273 (2015).
Macfarlane, S. & Macfarlane, G.T. Regulation of short-chain fatty acid production. Proc. Nutr. Soc. 62, 67–72 (2003).
Atarashi, K. et al. Treg induction by a rationally selected mixture of Clostridia strains from the human microbiota. Nature 500, 232–236 (2013).
Chen, X. et al. Requirement for the histone deacetylase Hdac3 for the inflammatory gene expression program in macrophages. Proc. Natl. Acad. Sci. USA 109, E2865–E2874 (2012).
White, C.A. et al. Histone deacetylase inhibitors upregulate B cell microRNAs that silence AID and Blimp-1 expression for epigenetic modulation of antibody and autoantibody responses. J. Immunol. 193, 5933–5950 (2014).
Reddy, P. et al. Histone deacetylase inhibition modulates indoleamine 2,3-dioxygenase-dependent DC functions and regulates experimental graft-versus-host disease in mice. J. Clin. Invest. 118, 2562–2573 (2008).
Le Leu, R.K. et al. Butyrylated starch intake can prevent red meat-induced O6-methyl-2-deoxyguanosine adducts in human rectal tissue: a randomised clinical trial. Br. J. Nutr. 114, 220–230 (2015).
Bajka, B.H., Topping, D.L., Cobiac, L. & Clarke, J.M. Butyrylated starch is less susceptible to enzymic hydrolysis and increases large-bowel butyrate more than high-amylose maize starch in the rat. Br. J. Nutr. 96, 276–282 (2006).
Round, J.L. & Mazmanian, S.K. Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota. Proc. Natl. Acad. Sci. USA 107, 12204–12209 (2010).
Marino, E. et al. BAFF regulates activation of self-reactive T cells through B-cell dependent mechanisms and mediates protection in NOD mice. Eur. J. Immunol. 44, 983–993 (2014).
Yi, W. et al. Targeted regulation of self-peptide presentation prevents type I diabetes in mice without disrupting general immunocompetence. J. Clin. Invest. 120, 1324–1336 (2010).
Ashelford, K.E., Chuzhanova, N.A., Fry, J.C., Jones, A.J. & Weightman, A.J. At least 1 in 20 16S rRNA sequence records currently held in public repositories is estimated to contain substantial anomalies. Appl. Environ. Microbiol. 71, 7724–7736 (2005).
Bragg, L., Stone, G., Imelfort, M., Hugenholtz, P. & Tyson, G.W. Fast, accurate error-correction of amplicon pyrosequences using Acacia. Nat. Methods 9, 425–426 (2012).
DeSantis, T.Z. et al. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 72, 5069–5072 (2006).
Fadrosh, D.W. et al. An improved dual-indexing approach for multiplexed 16S rRNA gene sequencing on the Illumina MiSeq platform. Microbiome 2, 6 (2014).
Mariño, E. et al. Marginal-zone B-cells of nonobese diabetic mice expand with diabetes onset, invade the pancreatic lymph nodes, and present autoantigen to diabetogenic T-cells. Diabetes 57, 395–404 (2008).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Liao, Y., Smyth, G.K. & Shi, W. The Subread aligner: fast, accurate and scalable read mapping by seed-and-vote. Nucleic Acids Res. 41, e108 (2013).
Robinson, M.D., McCarthy, D.J. & Smyth, G.K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140 (2010).
Polo, J.M. et al. A molecular roadmap of reprogramming somatic cells into iPS cells. Cell 151, 1617–1632 (2012).
We thank S. Akira (Osaka University) for Myd88−/− mice; D.R. Powell, S. Archer, Deltagen, B. Scherer, H. Thomas, G. Jhala, L. Mason, H.Y. Goh, M. Udugama, F. Lim, Y.A. Leong, C. Ang Kim Lian, C. McKenzie and Monash Animal Services for assistance; the NIH Tetramer Facility for the IGRP and BDC2.5 tetramers; and the Monash Histology Platform of Monash University. Supported by the Juvenile Diabetes Research Foundation (3-2013-94), the Diabetes Australia Research Trust (Y14M1-MARE), and the Australian National Health and Medical Research Council (program grant 1016953 and Australia Fellowship 606947).
The authors declare no competing financial interests.
Integrated supplementary information
(a) T1D incidence in germ-free (GF) NOD.MyD88−/− mice (n=11) vs. specific pathogen-free (SPF) NOD.MyD88−/− mice (n=14); ****P<0.0001, Mantel-Cox log rank test. Concentrations of acetate, butyrate and propionate in feces of (b) 5-10 week-old female SPF and GF NOD and NOD.MyD88−/− mice, and (c) NP-fed age-matched female vs male NOD and C57BL/6 mice, n ≥ 4. (d) Body weights of 15 week-old female NOD mice (n=5 per group) fed with NP, HAMS, HAMSA or HAMSB diets. (e) Energy intake measured from NOD mice fed different diets. Panels (b,c,e) were analyzed by one-way or (d) two-way ANOVA with Bonferroni’s multiple comparisons test. ****P<0.0001, ***P<0.001, **P<0.01, *P<0.05. For all graphs, data represent mean ± SEM; each symbol represents an individual mouse. All data are representatives of three independent experiments. (f) T1D incidence in female NOD mice fed for 5 weeks with HAMSA diet starting at age 5 weeks (n=11) or 10 weeks (n=15). Blue arrows in the graph represent the start and end time point on HAMSA diet. NS= not significant (5 weeks vs 10 weeks). (g) T1D incidence in HAMSA- and HAMSB-fed female NOD mice starting from birth to 30 weeks of age (n=10 per group). Mantel-Cox log rank test.
(a) Gating strategy used for identification of autoreactive IGRP tetramer+ CD8+ T cells and BDC2.5 tetramer+ CD4+ T cells. TUM and hu CLIP were used as tetramer controls respectively. Data is representative from 10,000 events acquired from the total viable lymphocyte gate. Representative plots and cumulative data showing frequency and number of PLN and splenic (b) autoreactive IGRP tetramer+ CD8+ gated from total CD8+ T cells and (c) BDC2.5 tetramer+ CD4+ gated from total CD4+ T cells from 15 week-old female NOD mice fed NP, HAMS, HAMSA or HAMSB diet. Data from HAMS and HAMSA for spleen in b is the same as that is shown in Figure 3a. Numbers shown within the plots are representative of the frequency of tetramer+ T cells. (d) Representative plots and cumulative data showing frequency and number of IGRP tetramer+ CD8+ T cells in 6 week-old NOD8.3 mice. All data represent mean ± SEM. ****P<0.0001, **P<0.001, *P<0.05 (NP vs HAMS; HAMSA vs HAMS; HAMSB vs HAMS). Data analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test. Data is representative of at least two independent experiments, n ≥ 2.
(a) Gating strategy used for splenic B cell subsets identification. (b) Cumulative data showing frequency (upper panel) and numbers (lower panel) of TR B cells, T2-MZ B cells, FO B cells and MZ B cells. ****P<0.0001, **P<0.01, *P<0.05 (NP vs HAMS; HAMSA vs HAMS; HAMSB vs HAMS). Data shown is from three independent experiments, n ≥ 4. (c) Mean fluorescence index (MFI) +/- SD of surface protein expression for MHCI and CD86 on a per-cell basis on splenic IgM+B220+ B cells (NOD mice fed with diets), n ≥ 5. ***P<0.001, **P<0.01, *P<0.05, NS (HAMSA vs HAMS; HAMSB vs HAMS). (d) Real time PCR showing expression of CD86 and Il12 on sorted splenic CD21high CD23low (MZ B) and CD21mid CD23high (FO B) cells (gated from total IgM+B220+ B cells), n=3. **P<0.01 (HAMSA vs NP; HAMSB vs NP). (e) Real time PCR showing fold change in expression of C80/B7.1, C80/B7.2, B2M and PRDM1 relative to β-actin in splenic IgM+B220+ B cells from of 15 week-old HAMS-, HAMSA-, or HAMSB-fed female NOD mice, n ≥ 4. **P<0.01, *P<0.05, NS (HAMSA vs HAMS; HAMSB vs HAMS). Data in panels b-e are representative of at least three experiments analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test.
(a) Gene expression profile between purified splenic IgM+B220+ B cells from HAMSA (x-axis) against HAMSB (y-axis), relative to HAMS contrast. Circles in red represent log2 fold-change expressed genes with FDR < 0.05 for differential expression test between HAMSA and HAMSB diets. (b) Cumulative data showing frequency of CFSE-labeled NOD.8.3 CD8+ T as CFSE−Vβ8.1/8.3+CD8+ T cells in PLN. **P=0.0042 (HAMSA vs HAMS); NS (NP vs HAMS and HAMSB vs HAMS). (c) Cumulative data showing frequency of CFSE−Vβ8.1/8.3+CD8+ T cells in MLN. (d) 3[H]-thymidine uptake by CD4+CD25− T cells cultured with LPS stimulated B cells and DCs from NOD mice in the absence or presence of 100 μM acetate or butyrate pulsed or not with insulin peptide ± IL-2 in vitro. B cells ## P=0.0015 (No SCFAs, insulin vs no peptide), **P=0.021, *P=0.001, (Insulin, acetate vs PBS; butyrate vs PBS) and DCs *P=0.0318 (Insulin, butyrate vs PBS). Panels b and c analyzed by one-way ANOVA and d by two-way ANOVA with Bonferroni’s multiple comparisons test. Data is representative of three reproducible experiments.
(a) Cumulative data showing frequency and numbers of CD4+FoxP3+CD103+ T cells from total CD4+ T cells isolated from the colon of 15 week-old female NOD mice fed HAMS, HAMSA or HAMSB diet. Data represent mean ± SD, n=5 mice. **P<0.01 (HAMSA vs HAMS; HAMSB vs HAMS). (b) Facs plots showing frequency of splenic, PLN and MLN CD4+FoxP3+ T cells. Data represent mean ± SEM, n=5-6 mice. (c) Gating strategy used for identification of CD4+ Foxp3+ T cells gated from total splenic CD4+ T cells. Data is representative from 10,000 events acquired from the total viable lymphocyte gate. (d) Flow cytometric analysis of Foxp3 protein expression on a per-cell basis from splenic and PLN Foxp3+ Treg cells from NOD-SCID recipient mice after 3 weeks post-transfer of CD4+Foxp3− T cells from NOD.FoxP3-GFP mice, fed with different diets as indicated. The data are shown as mean fluorescence intensity (MFI) mean ± SEM. Spleen **P=0.0023 (HAMSA vs HAMS); ****P<0.0001 (HAMSB vs HAMS). (e) Gating strategy used for single sorted CD45RBlowCD25+CD4+ T cells from PLN showing expression of Foxp3. (f) Flow cytometric analysis of Foxp3 protein expression on a per-cell basis from PLN Foxp3+ Treg cells from 15 week-old NOD/Lt mice fed with HAMS, HAMSA and HAMSB diets as indicated. *P<0.048 (HAMSB vs HAMS). Data analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test. Data are representative of at least three independent experiments.
(a) C57.Gpr43−/− mice were backcrossed 13 generations onto the NOD strain (NOD.Gpr43−/−). Once fully backcrossed, NOD.Gpr43−/− mice were genotyped at over 70,000 SNPs genome-wide. DNA from liver was purified and genotyped using the Mega-MUGA array (Geneseek, NB). Genotypes were compared to the reference (C57BL/6) alleles and to NOD alleles determined by the NOD genome sequence. The knockout was produced on the FVB background, so haplotypes are depicted as coming from the NOD genome (blue) or non-NOD genomes (red); grey indicates non-informative regions in which C57BL/6 and NOD have the same genotypes. This figure shows the strain of origin of haplotypes on each mouse chromosome, with the physical size of each chromosome shown on the X axis. This analysis demonstrated that all chromosomes were derived from the NOD strain, except for a region around the Gpr43 locus on chromosome 7. Thus, these mice harbor all NOD T1D susceptibility loci including the Idd7 and Idd27 loci mapped to ~19Mb and ~80-120 MB on chromosome 7, respectively. This enabled us to confirm that there have been no T1D susceptibility genes reported in the non-NOD interval. (b, c) T1D incidence in female NOD.Gpr43+/+ and NOD.Gpr43−/− littermates fed NP, HAMSA diet. *P=0.0392 (NOD.Gpr43+/+ mice vs NOD.Gpr43−/− fed HAMSA diet); *P=0.0067 (NOD.Gpr43−/− mice, HAMSB vs NP). Mantel-Cox log-rank test. Data shown is from two independent experiments (NOD.Gpr43−/− on HAMSA in b were from a different experiment to those shown in c). (d) Cumulative data showing absolute numbers of splenic and PLN CD4+FoxP3+ T cells and (e) IgM+B220+ B cells, n ≥ 5. Data were analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test. Data shown is from three independent experiments. (f) Frequency of PLN autoreactive IGRP tetramer+ CD8+ and IAg7/BDC2.5 tetramer+ CD4+ T cells from 15 week-old female NOD.Gpr43+/+ and NOD.Gpr43−/− mice fed NP or HAMSA diet. Data were analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test. IGRP+CD8+ T cells ***P<0.001 (NOD.Gpr43−/− vs NOD.Gpr43+/+ NP-fed; NOD.Gpr43−/− vs NOD.Gpr43+/+ HAMSA-fed; NS, NOD.Gpr43+/+ fed HAMSA vs NP); BDC2.5 CD4+ T cells (NOD.Gpr43+/+, HAMSA vs NP; NS, NOD.Gpr43−/− vs NOD.Gpr43+/+ NP-fed; NOD.Gpr43−/− vs NOD.Gpr43+/+ HAMSA-fed). Data represent mean ± SEM. Data shown is from three independent experiments.
(a) Microbial profile analysis of feces from NOD donors and GF re-colonized NOD mice fed HAMS, HAMSA and HAMSB diet by PCoA diagrams based on unweighted (left panel) and weighted (right panel) Unifrac distance metrics. The bacterial communities of different diets were significantly different based on both weighted and unweighted (P<1E−7) Unifrac, 1E7 permutations and Adonis permutational multivariate statistics. The legend represents the microbiota from donors NOD mice (circles) and GF NOD mice re-colonized by fecal transfer (FT) with HAMS, HAMSA and HAMSB modified microbiota in the same colour (squares). (b) Relative abundance of selected bacterial populations at (genus level) in donor NOD mice fed NP, HAMS, HAMSA or HAMSB diet and in GF NOD mice after fecal transfer. Data were analyzed by one-way ANOVA with Bonferroni’s multiple comparisons test. *P<0.05 (HAMSA vs HAMS; HAMSB vs HAMS; HAMSA.FT vs HAMS.FT); **P<0.01 (HAMS vs NP). Data represent mean ± SEM; each symbol represents individual mice. All data are representatives of two independent experiments. (c) Bar chart showing distribution of genera detected in feces from SPF donor NOD mice and GF NOD mice after fecal transfer (FT) for different diets, n=5-6 per group. Each genus is represented by a different colour and is proportional to the relative abundance in each sample. The legend shows the genera with relative abundance higher than 1%.
About this article
Cite this article
Mariño, E., Richards, J., McLeod, K. et al. Gut microbial metabolites limit the frequency of autoimmune T cells and protect against type 1 diabetes. Nat Immunol 18, 552–562 (2017). https://doi.org/10.1038/ni.3713
JCI Insight (2020)
Clinical Science (2020)
Molecular Carcinogenesis (2020)
Biochemical Society Transactions (2020)
Pretreatment with Yeast-Derived Complex Dietary Polysaccharides Suppresses Gut Inflammation, Alters the Microbiota Composition, and Increases Immune Regulatory Short-Chain Fatty Acid Production in C57BL/6 Mice
The Journal of Nutrition (2020)