Intestinal microbiota-derived short-chain fatty acids regulation of immune cell IL-22 production and gut immunity

Abstract

Innate lymphoid cells (ILCs) and CD4⁺ T cells produce IL-22, which is critical for intestinal immunity. The microbiota is central to IL-22 production in the intestines; however, the factors that regulate IL-22 production by CD4⁺ T cells and ILCs are not clear. Here, we show that microbiota-derived short-chain fatty acids (SCFAs) promote IL-22 production by CD4⁺ T cells and ILCs through G-protein receptor 41 (GPR41) and inhibiting histone deacetylase (HDAC). SCFAs upregulate IL-22 production by promoting aryl hydrocarbon receptor (AhR) and hypoxia-inducible factor 1α (HIF1α) expression, which are differentially regulated by mTOR and Stat3. HIF1α binds directly to the Il22 promoter, and SCFAs increase HIF1α binding to the Il22 promoter through histone modification. SCFA supplementation enhances IL-22 production, which protects intestines from inflammation. SCFAs promote human CD4⁺ T cell IL-22 production. These findings establish the roles of SCFAs in inducing IL-22 production in CD4⁺ T cells and ILCs to maintain intestinal homeostasis.

Fig. 1. SCFAs promote IL-22 production in CD4⁺ T cells and ILCs in vitro.

a WT splenic CD4⁺ T cells were activated with anti-CD3/CD28 mAbs ± butyrate (0.5 mM) for 2 days (n = 3 biologically independent samples per group). RNA sequencing was performed. Il10, Ifng, and Il22 expressions were shown in heatmap. b, c CBir1 Tg CD4⁺ T cells were cultured with APCs and CBir1 peptide ± acetate (10 mM), propionate (0.5 mM), or butyrate (0.5 mM) for 2 days (n = 3/group). Il22 expression was analyzed by qRT-PCR (b), and IL-22 in supernatants was assessed by ELISA (c). d CBir1 Tg CD4⁺ T cells were cultured with APCs and CBir1 peptide ± butyrate (0.5 mM) for 2 days (n = 3/group) under neutral, Th1, Th17, or Treg conditions. Il22 was analyzed by qRT-PCR. e CD4⁺ T cells were activated with anti-CD3/CD28 mAbs ± butyrate (0.5 mM) for 2 days (n = 3 biologically independent samples per group) under Th1 conditions. RNA sequencing was performed. Expression of Il10, Ifng, and Il22 was shown in heatmap. f–h CBir1 Tg CD4⁺ T cells were activated with APCs and CBir1 peptide ± butyrate (0.5 mM) under Th1 conditions (n = 3/group). IL-22 was analyzed by qRT-PCR at different time point (f), and ELISA at 60 h (g), and IL-22 and IL-17 were measured flow cytometry on day 5 (h). i CD4⁺ T cell-depleted splenic cells were treated with IL-23 (20 ng/ml) ± acetate (10 mM), propionate (0.5 mM), or butyrate (0.5 mM) for 16 h (n = 3/groups). IL-22 and IL-17 production in ILCs were analyzed by flow cytometry. One representative of three independent experiments was shown (b–d, f–i). Data were expressed as mean ± SD. Statistical significance was tested by two-tailed one-way ANOVA (b, c, i) or two-tailed unpaired Student t-test (d, g, h). b **p = 0.0014 (acetate vs control) and 0.0028 (propionate vs control), ****p < 0.0001; c *p = 0.0212 (acetate vs control) and 0.0139 (propionate vs control), ***p < 0.0004; d **p = 0.0029 (neutral), 0.0019 (Th1), and 0.0026 (Th17), ***p = 0.0003; g **p = 0.0016; h ***p = 0.0002, ns, no significance; i **p = 0.0033, ****p < 0.0001, ns no significance.

Fig. 2. Butyrate promotes intestinal CD4⁺ T cell and ILC production of IL-22.

WT mice were treated with or without 200 mM butyrate in drinking water for 3 weeks (n = 4 mice/group). a Mice were weighed daily. b Fecal pellets were collected prior and after 3-week treatment of butyrate, and butyrate levels were measured by LC–MS. Mice were killed on day 21, and IL-22 production in serum (c) and colonic organ cultures (d) were measured by ELISA. IL-22 production in CD4⁺ T cells (e) and ILCs (g) were analyzed in the spleen, MLN, and intestinal LP by flow cytometry. IL-22 levels in Th1, Th17, Treg cells (f), and ILCs (h) were measured in intestinal LP by flow cytometry. One representative of three independent experiments was shown. Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test. b ***p = 0.0009; c *p = 0.0145; d *p = 0.0393; e middle panel: *p = 0.0158 (SP), 0.0151 (MLN), and 0.0022 (LP); right panel: *p = 0.0359 (SP), 0.0377 (MLN), and 0.0481 (LP); f middle panel: *p = 0.0356 (Th1) and 0.0375 (Th17); right panel: *p = 0.0432 (Th1) and 0.0158 (Th17); g middle panel: *p = 0.0149 (SP), 0.0227 (MLN), and 0.0232 (LP); right panel: *p = 0.0126 (SP), 0.0448 (MLN), and 0.0462 (LP); h middle panel: *p = 0.0458; right panel: *p = 0.0436.

Fig. 3. Butyrate promotes IL-22 production through GPR41 and HDAC inhibition.

a, b CBir1 Tg CD4⁺ T cells were cultured with APCs and Cbir1 peptide with or without butyrate (0.5 mM) ± AR420626 (5 µM) or/and TSA (10 mM) under Th1 conditions (n = 3/group). IL-22 mRNA (a) and protein (b) were measured by qRT-PCR and ELISA at 60 h. IL-22 production was measured by flow cytometry on day 5 (c). d CD4⁺ T cells were cultured with anti-CD3/CD28 mAbs under Th1 conditions with or without butyrate (0.5 mM) or TSA (10 mM) (n = 3/group). Cells were collected at 24 h for analysis of HDAC activity at fluorescence intensity at excitation/emission (490/525 nm) by using the HDAC Activity Assay Kit. One representative of three independent experiments was shown. Data were expressed as mean ± SD. Statistical significance was tested by two-tailed one-way ANOVA. a ****p < 0.0001, **p = 0.0016, *p = 0.0282; b ****p < 0.0001; c ****p < 0.0001, ***p = 0.0002, **p = 0.0054; d *p = 0.0144, ***p = 0.0004.

Fig. 4. HIF1α and AhR mediate butyrate induction of IL-22 in CD4⁺ T cells.

a WT CD4⁺ T cells were activated with anti-CD3/CD28 mAbs under Th1 conditions ± butyrate (0.5 mM) for 2 days (n = 3 biologically independent samples per group). RNA sequencing was performed. Hif1α, Ahr, and Prdm1 were shown in heatmap. b–f CD4⁺ T cells were activated with anti-CD3/CD28 mAbs ± butyrate (0.5 mM) under Th1 conditions (n = 3/group). Hif1a (b) and Ahr (c) were analyzed by qRT-PCR. HIF1α (d) and AhR (e) protein was analyzed by western blot on day 2. HIF1α activity was measured using HIF1α Transcription Factor Assay Kit (f). g Raw 264.7 cells were transduced with XRE/AhR Luciferase Reporter Gene Lentivirus, and treated ± butyrate (0.5 mM) 3 days post transduction. AhR activity was assessed by luciferase. h–j Cbir1 Tg CD4⁺ T cells were activated with APCs and Cbir1 peptide under Th1 conditions with butyrate (0.5 mM) ± YC-1 (5 µM) or/and CH-223191 (3 µM) for 60 h (n = 3/group). IL-22 mRNA (h) and protein (i) were measured by qRT-PCR and ELISA. j IL-22 was measured by flow cytometry on day 5. k WT and HIF1α^−/− CD4⁺ T cells were activated with anti-CD3/CD28 mAbs ± butyrate (0.5 mM) for 5 days (n = 3/group). IL-22 was assessed by flow cytometry. l, m CD4⁺ T cells were activated with anti-CD3/CD28 mAbs under Th1 conditions with or without butyrate (0.5 mM), AR420626 (5 µM), or TSA (10 nM) for 60 h (n = 3/group). Hif1a (l) and Ahr (m) were measured by qRT-PCR. One representative of three independent experiments was shown (b–m). Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test (b–g) or two-tailed one-way ANOVA (h–m). b **p = 0.0033 (24 h), ***p = 0.0002 (36 h), **p = 0.0032 (48 h), *p = 0.0310 (60 h); c *p = 0.0338 (24 h), **p = 0.0054 (36 h), ***p = 0.0003 (48 h), ***p = 0.0007 (60 h); d *p = 0.0178; e *p = 0.0325; f *p = 0.0273; g *p = 0.0435; h ****p < 0.0001; i ****p < 0.0001, ***p = 0.0002 (butyrate + YC-1 vs butyrate), ***p = 0.0006; j ****p < 0.0001; k **p = 0.0015, *p = 0.0325; l **p = 0.0014 (butyrate vs control) and 0.0036 (AR429626 vs control); m ****p < 0.0001, ***p = 0.0009.

Fig. 5. Stat3 and mTOR regulate IL-22 production by CD4⁺ T cells.

a–d WT CD4⁺ T cells were activated with anti-CD3/CD28 mAbs under Th1 conditions with or without butyrate (0.5 mM) (n = 3/group). Phosphorylated Stat3 (6 h) (a, b) and phosphorylated mTOR (24 h) (c, d) were assessed by western blot and flow cytometry. Phosphorylated S6K was analyzed by flow cytometry (e). f–i CBir1 Tg CD4⁺ T cells were activated with APCs and CBir1 peptide under Th1 conditions with butyrate (0.5 mM) ± rapamycin (1 µM) or HJC0152 (1 µM). IL-22 mRNA (f) and protein (g) were assessed by qRT-PCR and ELISA at 60 h (n = 3/group). Expression of Hif1a (h) and Ahr (i) was analyzed at 48 h by qRT-PCR. j WT and Stat3^−/− CD4⁺ T cells were treated with or without butyrate (0.5 mM) for 5 days (n = 3/group). IL-22 production was measured by flow cytometry. One representative of three independent experiments was shown. Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test (a–e, j) or two-tailed one-way ANOVA (f–i). a *p = 0.0134; b ***p = 0.0002; c **p = 0.0059; d ***p = 0.0002; e **p = 0.0010; f, ****p < 0.0001; g **p = 0.0019 (butyrate vs control) and 0.0069 (butyrate + rapamycin vs butyrate), *p = 0.0141; h ***p = 0.0004, **p = 0.0012; i, ***p = 0.0004 (butyrate vs control) and 0.009 (butyrate + HJC0152 vs butyrate), **p = 0.0030; j **p = 0.0017, *p = 0.0338.

Fig. 6. Butyrate promotes HIF1α binding to Il22 promoter in CD4⁺ T cells.

a Schematic diagram of HIF1α binding to Il22 promoter. b WT CD4⁺ T cells were activated with anti-CD3/CD28 mAbs under Th1 conditions for 2 days (n = 3/group). HIF1α binding to Il22 promoter was analyzed by CHIP assay. c–e WT CD4⁺ T cells were cultured under Th1 conditions with or without butyrate (0.5 mM) for 2 days (n = 3/group). HIF1α binding to Il22 promoter was analyzed by CHIP assay (c). The H3K9 acetylation (d) and trimethylation (e) levels in HIF1α-binding site on Il22 promoter were assessed by CHIP assay. One representative of three independent experiments (b, c), or two independent experiments (d, e) was shown. Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test. b **p = 0.0047; c *p = 0.0278; d *p = 0.0105; e **p = 0.0094.

Fig. 7. Butyrate protects the intestines from Citrobacter rodentium infection.

a–f WT mice (n = 4 mice/group) were orally infected with Citrobacter rodentium (C. rodentium, 5 × 10⁸ CFU/mice), and treated with or without butyrate (200 mM) in drinking water for 10 days. Mice were weighed daily (a), and killed on day 10. Colonic histopathology (b), LP IL-22⁺, IL-10⁺, IFN-γ⁺, and IL-17⁺ CD4⁺ T cells (c), and IL-22⁺ and IL-17⁺ ILCs (d) were measured. CFU in feces (e) and liver (f) were measured. g, h WT mice (n = 4 mice/group) were orally infected with C. rodentium (5 × 10⁸ CFU/mice) on day 0, and with or without butyrate (200 mM) in drinking water for 10 days. Mice were administrated with anti-IgG antibody (25 mg/kg), anti-CD4 antibody (25 mg/kg), or anti-Thy1 (25 mg/kg) i.p. every other day. Mice were killed on day 10. Colonic histopathology was assessed (g), and CFU in feces was measured (h). One representative of three independent experiments (a–f) or two independent experiments (g, h) was shown. Scale bar, 300 µm (b, g). Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test (a, c–f, h), the non-parametric two-tailed Mann–Whitney U test (b, g). a *p = 0.0477, **p = 0.0041; b *p = 0.0286; c right: *p = 0.0169 (IL-22), 0.0107 (IL-10), and 0.0387 (IFN-γ); lower left: *p = 0.0247 (IL-22), 0.0224 (IL-10), and 0.0060 (IFN-γ); d middle: *p = 0.0308; right: *p = 0.0188; e *p = 0.0157; f **p = 0.0073; g **p = 0.0079, *p = 0.0159 (anti-CD4 + butyrate vs anti-CD4) and 0.0317(anti-CD4 vs anti-Thy1); h, ***p = 0.0002 (anti-IgG ⁺ butyrate vs anti-IgG) and 0.0005 (anti-CD4 vs anti-IgG), *p = 0.0214 (anti-CD4 + butyrate vs anti-CD4) and 0.0263 (anti-Thy1 vs anti-CD4), **p = 0.0020.

Fig. 8. Butyrate inhibits intestinal infection by promoting IL-22 production.

WT and IL-22^−/− mice (n = 4 mice/group) were orally infected with C. rodentium (5 × 10⁸ CFU/mice) on day 0, and treated with or without butyrate (200 mM) in drinking water for 10 days. Mice were weighed daily (a). At day 10, colonic histopathology was analyzed (b), and colonic IL-6 (c) and TNF (d) production in colonic tissue was determined by ELISA. CFU in feces (e) and liver (f) were measured. One representative of three independent experiments was shown. Scale bar, 300 µm (b). Data were expressed as mean ± SD. Statistical significance was tested by two-tailed unpaired Student t-test (a, c–f) or the nonparametric two-tailed Mann–Whitney U test (b). a *p = 0.0379; b *p = 0.0286; c *p = 0.0254 (WT butyrate vs WT control) and 0.0156 (IL-22^−/− control vs WT control); d *p = 0.00105 (WT butyrate vs WT control) and 0.0452 (IL-22^−/− control vs WT control); e *p = 0.0155, **p = 0.0086; f, *p = 0.0163, ***p = 0.0001.

Fig. 9. Butyrate induces human CD4⁺ T cell IL-22 production.

a–e Peripheral blood CD4⁺ T cells were isolated from healthy controls (HC, n = 8 biologically independent samples), patients with active Crohn’s colitis (CD, n = 10 biologically independent samples) and ulcerative colitis (UC, n = 7 biologically independent samples), and activated with anti-CD3/CD28 mAbs with or without butyrate (0.5 mM). Il22 expression was assessed at day 3 by qRT-PCR (a), IL-22⁺ cells were measured by flow cytometry at day 5 (b), and IL-22 production in supernatants was measured at day 3 by ELISA (c). Hif1a (d) and Ahr (e) expression in CD4⁺ T cells were analyzed by qRT-PCR at day 3. f Peripheral blood CD4⁺ T cells from healthy controls, CD, and UC patients were treated with or without butyrate (0.5 mM) ± YC-1 (20 µM) or CH-223191 (5 µM) for 5 days (n = 3/group). IL-22 production was analyzed by flow cytometry. One representative of three independent experiments was shown. Scale bar, 300 µm. Data were expressed as mean ± SD. Statistical significance was tested by two-tailed paired Student t-test (a–e), or two-tailed one-way ANOVA (f). a **p = 0.0024, ***p = 0.0008 (CD), and 0.0003 (UC); b ***p = 0.0001, ****p < 0.0001, **p < 0.0011; c *p = 0.0185 (HC) and 0.0130 (CD), **p = 0.0046; d **p = 0.0023 (HC), 0.0024 (CD), and 0.0094 (UC); e **p = 0.0040 (HC), 0.0014 (CD), and 0.0023 (UC); f ****p < 0.0001, ***p = 0.003 (CD butyrate + YC-1 vs CD butyrate) and 0.004 (CD butyrate + CH-223191 vs CD butyrate).

Fig. 10. SCFAs induction of IL-22 in CD4⁺ T cells and ILCs.

Gut microbiota-derived SCFAs promote IL-22 production in CD4⁺ T cells and ILCs. Mechanically, butyrate promotes IL-22 production through GPR41 and HDAC inhibition. Furthermore, butyrate upregulates HIF1α and AhR, which is differentially regulated by mTOR and Stat3. HIF1α directly binds to the Il22 promoter, and butyrate increases HIF1α binding to the Il22 promoter through histone modifications.