Succinate drives gut inflammation by promoting FOXP3 degradation through a molecular switch

Abstract

Succinate levels are increased in inflammatory bowel disease (IBD), but its role in disease pathogenicity remains unknown. Here we showed that succinate promoted colitis in mice by reducing the expression of FOXP3 and increasing the expression of interleukin-17 in regulatory T (T_reg) cells. Succinate selectively reduced the expression of 2-oxoglutarate dehydrogenase complex (OGDHc), the enzyme for succinyl-CoA synthesis, which in turn reduced FOXP3 succinylation and made FOXP3 lysine residues available for ubiquitination and FOXP3 protein degradation. Genetic deletion of Dlst, a member of OGDHc, in T_reg cells led to reduced expression of FOXP3, impaired T_reg cells function and severe gut inflammation. Restoring FOXP3 expression fully rescued the immune suppressive functions of Dlst-deficient T_reg cells. In individuals with IBD, FOXP3 and OGDHc levels were reduced in T_reg cells and negatively correlated with succinate levels and inflammation severity. This study identifies succinate as a pathogenic factor in IBD, uncovering a succinate-driven molecular switch that regulates FOXP3 stability and T_reg cells function during inflammation.

Extended Data Fig.1|. Analysis of succinate on DSS-induced colitis.

a, The absolute number of CD45⁺ immune cells in the colon at day 7 in C57BL/6J mice treated with 2.5% DSS (hereafter 2.5% DSS, n = 8) or 3% DSS (3% DSS, n = 5) in drinking water for 7 days, or with 3% succinate in drinking water for 3 days before treatment with 2.5% DSS and 3% succinate for 7 days (2.5% DSS+3% succinate, n = 8) or none of the above (control, n = 4). b, Representative flow cytometry of CD4⁺ and CD8⁺ T cells in CD45⁺ cells in the inflamed colon at day 7 in mice as in a. c, The percentage and absolute numberof CD4⁺CD45⁺ T cells in the inflamed colon at day 7 in mice as in a. d, The percentage and absolute numberof CD8⁺CD45⁺ T cells in the inflamed colon at day 7 in mice as in a. e-f, Representative flow cytometry (e) of and absolute number (f) of CD45⁺CD4⁺FoxP3⁻ conventional T cells in the inflamed colon at day 7 in mice as in a. Data are shown as means ± s.d., and analyzed by one-way ANOVA with Tukey’s post-hoc test. Numbers in the bars represent exact P values. Data are representative of two independent experiments.

Extended Data Fig.2|. Analysis of ROR-γt, IL-2, IL-4, and IFN-γ expression in iT_reg cells by succinic acid treatment.

a, Representative flow cytometry of the percentage of IL-17⁺FoxP3⁺ iT_reg cells in C57BL/6J murine splenic CD4⁺ T cells cultured under iTreg cell polarizing conditions (IL-2+TGF-β) in the presence of 0 mM (control), 1.0 mM, or 5.0 mM SA for 5 days. b-c, Representative flow cytometry (b) and quantification (c) of the percentage of ROR-γt⁺FoxP3⁺ iT_reg cells in CD4⁺T cells as in a. d-e, Representative flow cytometry (d) and quantification (e) of the percentage of IL-2⁺FoxP3⁺ iT_reg cells in CD4⁺T cells as in a. f-g, Representative flow cytometry (f) and quantification (g) of the percentage of IL-4⁺FoxP3⁺ iT_reg cells in CD4⁺T cells as in a. h-i, Representative flow cytometry (h) and quantification (i) of the percentage of IFN-γ⁺FoxP3⁺ iT_reg cells in CD4⁺T cells as in a. j-k, Representative flow cytometry (j) and quantification (k) of the percentage of CD25⁺FoxP3⁺ iT_reg cells and FoxP3 mean fluorescent intensity (MFI) in C57BL/6J murine splenic CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of 0 mM (control), 0.1 mM, 0.25 mM, 0.5 mM, 1.0 mM, or 5.0 mM succinic acid (SA) for 5 days. l-m, Representative flow cytometry (l) and quantification (m) of FoxP3 MFI CD25⁺FoxP3⁺ iT_reg cells in C57BL/6J murine splenic CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of SA (0.1 mM) or IFN-γ (50 ng/mL) or both for 5 days. Data are shown as means ± s.d., and analyzed by one-way ANOVA with Tukey’s post-hoc test. Numbers in the bars represent exact P values. Data are representative of at least three independent experiments.

Extended Data Fig.3|. Analysis of the GPR91 signaling and HIF-1α activation on succinate-mediated FoxP3 downregulation.

a-b, Immunoblot (a) and quantification (b) of FoxP3 protein levels in human T_reg-like MT-2 cells cultured with succinic acid (SA, 5 mM) or the SUCNR1-specific antagonist NF-56-EJ40 (1 μM) or both, for 2 days. c-d, Representative flow cytometry (c) and quantification (d) of the percentage of CD25⁺FoxP3⁺ iT_reg cells in human naïve CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of SA (1 mM) and/or NF-56-EJ40 (1 μM) for 5 days. e-f, Representative flow cytometry (e) and quantification (f) of the percentage of CD25⁺FoxP3⁺ iT_reg cells in C57BL/6J murine splenic CD4⁺ T cells cultured under iTreg cells polarizing conditions (IL-2+TGF-β) in the presence of in the presence of SA (1 mM) and/or Bay 87–2243 (20 nM), a HIF-1α inhibitor, for 5 days. g, RT-PCR expression of Suclg1, Suclg2, and Sucla2 in splenic C57BL/6J murine CD4⁺ T cells cultivated under T_H0 (IL-2), T_H1 (IL-2+IL-12), T_H2 (IL-2+IL-4), T_H17 (IL-6+TGF-β) and iT_reg cells (IL-2+TGF-β) polarization conditions for 3 days. Data are shown as means ± s.d., and analyzed by one-way ANOVA with Tukey’s post-hoc test (b,d,f) or unpaired 2-tailed Student’s t test (g). Numbers in the bars represent exact P values. Data are representative of at least three independent experiments.

Extended Data Fig.4|. Immune profiling in Dlst^Treg-WT and Dlst^Treg-KO mice.

a-b, Representative images of H&E staining (a) and histopathological scores (b) from the liver, lung, and kidney of 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 5, per group. Scale bar, 100 μm. c-e, Representative flow cytometry (c) and quantification of the percentage (d) and absolute numbers (e) of CD4⁺CD8⁻, CD4⁻CD8⁺, CD4⁺CD8⁺, and CD4⁻CD8⁻T cells in the thymus from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 7, per group. f-h, Representative flow cytometry (f) and quantification of the percentage (g) and absolute numbers (h) of CD4⁺CD8⁻ and CD4⁻CD8⁺ T cells in the spleen from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 7, per group. i-k, Representative flow cytometry (i) and quantification of the percentage (j) and absolute numbers (k) of CD4⁺CD8⁻ and CD4⁻CD8⁺ T cells in the lymph nodes from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 7, per group. Data are shown as means ± s.d. (d,e,g,h,j,k) or sem (b), and analyzed by unpaired 2-tailed Student’s t test. Numbers in the bars represent exact P values.

Extended Data Fig.5|. Immune profiling in Dlst^Treg-WT and Dlst^Treg-KO mice.

a-c, Representative flow cytometry (a) and quantification of the percentage of CD62L^hiCD44^low naïve and CD62L^lowCD44^hi effector CD4⁺ T cells in the spleen (b) and lymph node (c) from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 7, per group. d-f, Representative flow cytometry (d) and quantification of the percentage of CD62L^hiCD44^low naïve and CD62L^lowCD44^hi effector CD8⁺ T cells in the spleen (e) and lymph nodes (f) from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 7, per group. g-h, Representative flow cytometry (g) and quantification (h) of the percentage of IL-2⁺CD4⁺FoxP3⁻, IL-4⁺CD4⁺FoxP3⁻, IL-17⁺CD4⁺FoxP3⁻, and IFN-γ⁺CD4⁺FoxP3⁻ T cells in the spleen from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 6, per group. i-n, Histogram and quantification of specified cell surface proteins in CD4⁺CD25⁺FoxP3⁺ T_reg cells in the thymus (i,j), spleen (k,l), and lymph node (m,n) from 3-week-old Dlst^Treg-WT and Dlst^Treg-KO mice. n = 6, per group. Numbers in quadrants indicate the value of mean fluorescent intensity (MFI) (i,k,m). Data are shown as means ± s.d. (b,c,e,f,h) or sem (j,l,n), and analyzed by unpaired 2-tailed Student’s t test.

Extended Data Fig.6|. Immune profiling in Dlst^Treg-iWT and Dlst^Treg-iKO mice.

6–9-week-old Dlst^Treg-iWT and Dlst^Treg-iKO mice were treated with tamoxifen for 6 days, and sacrificed at day 8. a, Spleen weight in Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. b, Total cellularity of thymus, spleen, and lymph node in Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. c-e, Representative flow cytometry (c) and quantification of the percentage (d) and absolute numbers (e) of CD4⁺CD8⁻, CD4⁻CD8⁺, CD4⁺CD8⁺, and CD4⁻CD8⁻T cells in the thymus of Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. f-h, Representative flow cytometry (f) and quantification of the percentage (g) and absolute numbers (h) of CD4⁺CD8⁻, CD4⁻CD8⁺ T cells in the spleen of Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. i-k, Representative flow cytometry (i) and quantification of the percentage (j) and absolute numbers (k) of CD4⁺CD8⁻, CD4⁻CD8⁺ T cells in the lymph node of Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. l-q, Histogram and quantification of specified cell surface proteins in CD4⁺CD25⁺FoxP3⁺ T_reg cells in the thymus (l,m), spleen (n,o), and lymph node (p,q) from Dlst^Treg-iWT and Dlst^Treg-iKO mice. n = 5, per group. Numbers in quadrants indicate the value of mean fluorescent intensity (MFI). (l,n,p) Data are shown as means ± s.d. (a,b,c,d,e,g,h,j,k) or sem (m,o,q), and analyzed by unpaired 2-tailed Student’s t test.

Extended Data Fig.7|. Metabolite alterations in DLST-deficient T_reg cells.

a-c, Representative flow cytometry (a) and the quantification of the percentage of CD25⁺FoxP3⁺ iT_reg cells (b) and FoxP3 MFI in CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of 4-hydroxytamoxifen (1 μM) for 5 days from the spleen of Dlst^Treg-iWT and Dlst^Treg-iKO mice. d-g, Heatmap showing the altered metabolites in iT_reg cells polarized from Dlst^Treg-iWT and Dlst^Treg-iKO mice, quantified by LC–MS (d), the pathway enrichment for the upregulated (e) and downregulated (f) metabolites in iT_reg cells polarized from Dlst^Treg-iKO mice, and TCA cycle-associated metabolites in iT_reg cells polarized from Dlst^Treg-iWT and Dlst^Treg-iKO mice, quantified by LC–MS (g). Data are shown as means ± s.d., and analyzed by unpaired 2-tailed Student’s t test. Data are representative of three independent experiments.

Extended Data Fig.8|. Analysis of FoxP3 mRNA levels in iT_regs and total protein succinylation in MT-2 cells.

a, RT-PCR expression of FoxP3 in splenic C57BL/6J mice CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of 0 mM (control) or 1 mM succinic acid (SA) for 5 days. b, RT-PCR expression of FoxP3 in splenic Dlst^Treg-iWT and Dlst^Treg-iKO mice CD4⁺ T cells cultured under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of SA (1 mM) and 4-hydroxytamoxifen (1 μM) for 5 days. c, Immunoblot of FoxP3, DLST, succinyllysine, and GAPDH in the cell lysates of human T_reg-like MT-2 cells after 24 hour-treatment with SA (5 mM) for 24 hours. d, Immunoblot of succinyllysine, FoxP3, and GAPDH in the cell lysates of MT-2 cells after 24 hour-treatment with 0 μM, 1 μM, or 5 μM succinyl-CoA. e, Immunoblot of succinyllysine, FoxP3, and GAPDH in the cell lysates of MT-2 cells after 24 hour-treatment with 0 μM, 1 μM, or 5 μM succinyl-CoA. Data are shown as means ± s.d., and analyzed by unpaired 2-tailed Student’s t test. Data are representative of at least two independent experiments.

Extended Data Fig. 9|. Analysis of Succinyl-CoA effects on the suppressive functions of Dlst^Treg-iKO cells in vitro.

a-b, Histogram (a) and quantification (b) of the in vitro suppressive activity of iT_reg cells polarized from splenic CD4⁺T cells of Dlst^Treg-iWT and Dlst^Treg-iKO mice assessed by the division of CD8⁺ T cells in the presence or absence of succinyl-CoA (5 μM). 1:1, 1:2, and 1:4 indicated the ratio of iT_reg cells to CD8⁺T cells. See method for details. Splenic CD4⁺ T cells from Dlst^Treg-iWT and Dlst^Treg-iKO mice were cultivated under iT_reg cells polarizing conditions (IL-2+TGF-β) in the presence of 4-hydroxytamoxifen (1 μM) for 5 days, and further treated with IL-2 (5 ng/mL), with/without inflammatory cytokines IFN-γ (100 ng/mL), IL-4 (10 ng/mL), IL-6 (50 ng/mL), and IL-12 (10 ng/mL) for another 3 days. c-d, Representative flow cytometry (c) and quantification of the MFI of FoxP3 (d) in CD4⁺CD25⁺FoxP3⁺ iT_reg cells from Dlst^Treg-iWT and Dlst^Treg-iKO mice. e, The degradation rate of FoxP3 MFI in CD4⁺CD25⁺FoxP3⁺ iT_reg cells in various inflammatory contexts, normalized to no treatment (control).Data are shown as means ± s.d., and analyzed by unpaired 2-tailed Student’s t test. Data are representative of three independent experiments.

Fig. 1.. Succinate downregulates FOXP3 and OGDH expression in T_reg cells.

a, Schematic of the experimental setup showing C57BL/6J mice were treated with 2 or not (control, n = 5) with 2.5% DSS (hereafter 2.5% DSS, n = 8) or 3% DSS (3% DSS, n = 7) for 7 days, or with 3% succinate for 3 days followed by 3% succinate + 2.5% DSS, both in drinking water, for 7 days (2.5% DSS + 3% succinate, n = 8). b-e, Body weight changes relative to starting weight (b), colon length (c), representative H&E staining in the colon (d) and histological score in the colon (e) in control mice, and mice treated with 2.5% DSS, 2.5% DSS + 3% succinate or 3% DSS, as in a. Scale bar, 100 μm. f,g, Representative flow cytometry (f) and quantification of the percentage of FOXP3⁺ cells and FOXP3 MFI (g) in CD4⁺FOXP3⁺ T_reg cells from the colons of mice as in a. h–j, Representative flow cytometry (h), frequency of CD25⁺FOXP3⁺ T_reg cells and FOXP3 MFI (i) and quantification of IL-17⁺FOXP3⁺ cells (j) in C57BL/6J mouse splenic CD4⁺ T cells cultured in iT_reg cell-polarizing condition (IL-2 + TGFβ) in the presence of 0 mM (control), 1.0 mM or 5.0 Mm succinic acid (SA) for 5 days. k,l, Representative flow cytometry (k) and quantification of the percentage of CD25⁺FOXP3⁺ T_reg cells and FOXP3 MFI (l) in splenic iT_reg cells as in h–j in the presence of with 0 μM (control), 10 μM, 25 μM or 50 μM cell-permeable succinate (NV-118) for 5 days. m,n, Representative flow cytometry (m), quantification of the percentage of CD25⁺FOXP3⁺ T_reg cells and FOXP3 MFI (n) in human CD4⁺CD25⁺FoOXP3⁺ iT_reg cells derived from CD4⁺ T_N cells isolated from peripheral blood and cultured with IL-2 + TGFβ in the presence or not (control) of 1 mM succinic acid for 5 days. o,p, RT–PCR expression of Ogdh, Dlst and Dld (o) and Suclg1, Suclg2 and Sucla2 (p) in mouse splenic iT_reg cells polarized as in h,i in the presence or not (control) of 1 mM succinic acid for 3 days. q, RT–PCR expression of Ogdh, Dlst and Dld in C57BL/6J mouse spleen CD4⁺ T cells cultured under T_H0 (IL-2), T_H1 (IL-2 + IL-12), T_H2 (IL-2 + IL-4), T_H17 (IL-6 + TGFβ) and iT_reg (IL-2 + TGFβ) cell polarization conditions for 3 days. Data are shown as mean ± s.d. and analyzed by two-way analysis of variance (ANOVA) (b), one-way ANOVA (c,e,g,i,j,l) with Tukey’s post hoc test or two-tailed unpaired Student’s t-test (n–q). Numbers in the bars represent exact P values.

Fig. 2.. DLST is required for T_reg cell restriction of lethal inflammation.

a, Survival of Dlst^Treg-WT and Dlst^Treg-KO mice (P < 0.0001 using one-sided log-rank test). b,c, Representative images of H&E staining (b) and histopathological scores (c) from the colon of Dlst^Treg-WT or Dlst^Treg-KO mice (n = 5 mice per group). Scale bars, 100 μm. d, Spleen weight in Dlst^Treg-WT or Dlst^Treg-KO mice (n = 7 mice per group). e, Total cell numbers in thymus, spleen and lymph nodes in Dlst^Treg-WT or Dlst^Treg-KO mice (n = 7 mice per group). f–i, Representative flow cytometry showing percentage (upper numbers) and FOXP3 MFI (lower numbers) (f), frequency among CD4⁺ T cells (g), absolute numbers (h) and FOXP3 MFI (i) of CD4⁺CD25⁺FOXP3⁺ T_reg cells in the thymus, spleen and lymph nodes of Dlst^Treg-WT or Dlst^Treg-KO mice (n = 7 mice per group). j,k, Representative flow cytometry (j) and quantification (k) of intracellular staining of IL-2, IL-4, IL-17 and IFNγ in splenic CD4⁺FOXP3⁺ T_reg cells from Dlst^Treg-WT or Dlst^Treg-KO mice (n = 6 mice per group). Data are shown as mean ± s.d. and analyzed by two-tailed unpaired Student’s t-test.

Fig. 3.. DLST is required for the suppressive function of Treg cells.

a, Immunoblot analysis of DLST protein in CD4⁺CD25⁺FOXP3-GFP⁺tdTomato⁺ T_reg cells from the spleen of 6–9-week-old Dlst^wt/wtFoxp3^GFP-creERT2ROSA26Sor^CAG-tdTomato (Dlst^Treg-iWT) or Dlst^fl/flFoxp3^GFP-creERT2ROSA26Sor^CAG-tdTomato (Dlst^Treg-iKO) mice at day 8 post-initiation of tamoxifen treatment for 6 days. b-g, Flow cytometry analysis in the thymus (b), spleen (c) and lymph node (d), frequency among CD4⁺T cells (e), absolute cell numbers (f) and FOXP3 MFI (g) of CD4⁺CD25⁺FOXP3⁺ T_reg cells in Dlst^Treg-iWT or Dlst^Treg-iKO mice (n = 5 mice per group). h,i, Intracellular staining and quantification of IL-2 (h) and IL-17 (i) in splenic CD4⁺CD25⁺FOXP3⁺ T_reg cells in Dlst^Treg-iWT or Dlst^Treg-iKO mice (n = 4 mice per group). j, Intracellular staining and quantification of IL-17 in iTreg cells derived from splenic CD4⁺ T cells from Dlst^Treg-iWT or Dlst^Treg-iKO mice (n = 4). k, Histograms and quantification of the division of CTV⁺CD8⁺ T cells cultured with Dlst^Treg-iWT or Dlst^Treg-iKO iT_reg cells for 3 days. CTV, Cell Trace Violet. l,m, Changes in body weight relative to baseline at week 1 to week 8 (W1–W8) (l) and H&E staining and histological score of colon tissues at week 8 (m) in Rag1^−/− mice that received splenic CD4⁺ TN cells from CD45.1 mice either alone (no Treg cells, n = 4) or with Dlst^Treg-iWT iT_reg cells (n = 5) or Dlst^Treg-iKO iT_reg cells (n = 5). Scale bar, 100 μm. n–p, Representative flow cytometry of colon tissue (n), absolute number of CD45.1⁺, IFNγ⁺CD45.1⁺ and IL-17⁺CD45.1⁺ cells (o) and absolute number of CD45.2⁺, IFNγ⁺CD45.2⁺ and IL-17⁺CD45.2⁺ iT_reg cells (p) in Rag1^−/− mice that received splenic CD4⁺T_N cells from CD45.1 mice either alone (no Treg cells, n = 4) or with Dlst^Treg-iWT iT_reg cells (n = 5) or Dlst^Treg-iKO iT_reg cells (n = 5). Data are shown as mean ± s.d. and analyzed by one-way ANOVA with Tukey’s post hoc test (l,m,o) or two-tailed unpaired Student’s t-test (e–k,p).

Fig. 4.. Succinate promotes T_reg cell production of IL-17 independent of FoxP3.

a,b, Representative flow cytometry analysis (a) and quantification (b) of colonic tdTomato⁺FOXP3–eGFP⁺CD4⁺ T_reg cells and tdTomato⁺FOXP3–eGFP⁻CD4⁺ exT_reg cells in total CD4⁺ T cells and total tdTomato⁺ cells in 2.5% DSS (n = 4) or 2.5% DSS + 3% succinate (n = 5) Foxp3^GFP-creERT2ROSA26Sor^CAG-tdTomato mice at day 7. c,d, Representative flow cytometry analysis (c) and quantification (d) of colonic tdTomato⁺FOXP3⁺ T_reg cells and tdTomato⁺FOXP3⁻ exT_reg cells in total CD4⁺ T cells and total tdTomato⁺ cells in mice as in a,b. e–g, Representative flow cytometry analysis (e,f) and quantification (g) of IL-17⁺ cells among tdTomato⁺FOXP3⁺ T_reg cells (e) and tdTomato⁺FOXP3⁻ exT_reg cells (f) in mice as in a,b. h,i, Representative flow cytometry analysis (h) and quantification (i) of splenic tdTomato⁺FOXP3–eGFP⁺CD4⁺ T_reg cells and tdTomato⁺FOXP3–eGFP⁻CD4⁺ exT_reg cells in total CD4⁺T cells and total tdTomato⁺ cells in Dlst^Treg-iWT and Dlst^Treg-iKO mice (n = 4 mice per group). j,k, Representative flow cytometry analysis (j) and quantification (k) of splenic tdTomato⁺FOXP3⁺ T_reg cells and tdTomato⁺FOXP3⁻ exT_reg cells in total CD4⁺ T cells and total tdTomato⁺ cells in Dlst^Treg-iWT or Dlst^Treg-iKO mice as in h,i. l–n, Representative flow cytometry analysis (l,m) and quantification (n) of IL-17⁺ cells among tdTomato⁺FOXP3⁺ T_reg cells (l) and tdTomato⁺FOXP3⁻ exT_reg cells (m) in Dlst^Treg-iWT or Dlst^Treg-iKO mice as in j,k. Data are shown as mean ± s.d. and analyzed by two-tailed unpaired Student’s t-test.

Fig. 5.. Succinate inhibits succinyl-CoA levels in T_reg cells.

a, Heatmap showing the differentially expressed genes in sorted CD4⁺CD25⁺tdTomato⁺FOXP3–GFP⁺ iT_reg cells at day 2 of culture of splenic CD4⁺ T cells from Dlst^Treg-iWT or Dlst^Treg-iKO mice in the presence of 4-hydroxytamoxifen, IL-2 and TGFβ (n = 3, per group). b, Functional enrichment pathways for the genes upregulated in Dlst^Treg-iKO iT_reg cells. c, Volcano plot showing metabolites alterations between Dlst^Treg-iKO iT_reg cells and Dls^tTreg-iWT iT_reg cells at day 5 of culture, quantified by liquid chromatography–mass spectrometry (LC–MS) (n = 3, per group). d, TCA cycle schematic (top) and α-KG and succinic acid abundance (bottom) in Dlst^Treg-iKO iT_reg cells and Dlst^Treg-iWT iT_reg cells, quantified by LC–MS (n = 3, per group). e, OCR plot and quantification of Dlst^Treg-iWT iT_reg cells (n = 10) and Dls^tTreg-iKO iT_reg cells (n = 11) after 5-day culture. f, Expression of succinyl-CoA (n = 6, per group) and acetyl-CoA (n = 3, per group) in Dlst^Treg-iKO iT_reg cells and Dlst^Treg-iWT iT_reg cells, quantified by ELISA or LC–MS, respectively. g, Immunoblot of succinyllysine after immunoprecipitation with FOXP3 antibodies in human Treg-like MT-2 cells. h, Immunoblot of FOXP3 after immunoprecipitation with succinyllysine (Ksucc) antibodies in MT-2 cells. i, Immunoblot of succinyllysine after immunoprecipitation with FOXP3 antibodies in DlstT^reg-iKO iT_reg cells and Dlst^Treg-iWT iT_reg cells at day 5 in culture. j, Immunoblot of succinyllysine after immunoprecipitation with FOXP3 antibodies in MT-2 cells 24 h post-treatment with 5 μM succinyl-CoA. k,l, Representative flow cytometric analysis (k), frequency of CD4⁺CD25⁺FOXP3⁺ iT_reg cells and FOXP3 MFI (l) among CD4⁺ T cells treated with succinic acid (1 mM) or succinyl-CoA (5 μM) for 5 days. m,n, Immunoblot of succinyllysine after immunoprecipitation with FOXP3 antibodies (m) and quantification of succinylated FOXP3 (su-FOXP3) levels (n) in WT and Suclg2^−/− MT-2 cells. o, Expression of succinyl-CoA in WT and Suclg2^−/− MT-2 cells, quantified by ELISA. Data are shown as mean ± s.d. and analyzed by two-tailed unpaired Student’s t-test (d–f,i,j,n,o) or one-way ANOVA (l) with Tukey's post-hoc test.

Fig. 6.. Succinate promotes the ubiquitination and degradation of FoxP3.

a, Immunoblot of HA on MYC immunoprecipitants in HEK293T cells transfected with MYC-FOXP3 and HA-ubiquitin plasmids at 24 h post-treatment with 5 μM succinyl-CoA. b, Immunoblot analysis of FOXP3 protein in cell lysates as in a. c,d, Representative flow cytometry (c) and quantification (d) of FOXP3-GFP expression in HEK293T cells transfected with a FOXP3-GFP fusion-expressing plasmid at 24 h post-treatment with 5 μM succinyl-CoA. e, Representative immunoblot and quantification (n = 3) of FOXP3 protein amount in MT-2 cells treated with 5 μM succinyl-CoA for 24 h followed by treatment with cycloheximide (CHX, 20 μg ml⁻¹) for 0 h, 1 h, 2 h and 4 h. f,g, Representative flow cytometry (f) and quantification (n = 4) (g) of FOXP3 in iT_reg cells derived from mouse C57BL/6J CD4⁺T cells and treated with 5 μM succinyl-CoA and CHX (20 μg ml⁻¹) as in e. h, Immunoblot of Ub-FOXP3 after immunoprecipitation with FOXP3 antibodies in Dlst^Treg-iKO and Dlst^Treg-iWT T_reg cells. i,j, Representative flow cytometry (i) and quantification (n = 3) (j) of FOXP3 protein in Dlst^Treg-iWT and Dlst^Treg-iKO iT_reg cells treated with CHX (20 μg/mL) for 0 h, 4 h, 8 h and 12 h. k,l, Immunoblot of FOXP3 ubiquitination (k) and cell lysates analyzed with HA antibodies, MYC antibodies and GAPDH as loading controls (l) in HEK293T cells transfected with MYC-tagged FOXP3 and HA-tagged WT, K48 or K63 mutant ubiquitin for 24 h, then treated with 5 mM succinic acid for another 24 h. m–o, Representative immunoblot (m) and semi-quantification of FOXP3 (n) and STUB1 (o) protein levels in WT and Stub1^−/− MT-2 cells treated with or without 5 mM succinic acid for 24 h. p, Representative immunoblot of cell lysates from MT-2 cells treated with or without 5 mM succinic acid for 24 h. Data are shown as mean ± s.d. and analyzed by one-way ANOVA with Tukey’s post hoc test (d,e,g,j,n) or two-tailed unpaired Student’s t-test (b,o).

Fig. 7.. Succinylation protects FOXP3 from proteasomal degradation.

a, Mass spectra showing ubiquitination to succinylation on K8 and K263 site in MT-2 cells treated with succinyl-CoA (5 μM) for 24 h followed by FOXP3 protein immunoprecipitation and MS. b, Representative immunoblot and quantification (n = 3) of FOXP3 succinylation in HEK293T cells transfected with FOXP3^WT, FOXP3^K8R, FOXP3^K263R and FOXP3^KDR for 24 h and immunoprecipitated with an HA antibody. c, Immunoblot of FOXP3 ubiquitination in HEK293T cells transfected with FOXP3^WT and FOXP3^KDR and Flag-ubiquitin plasmids and immunoprecipitated with an HA antibody. d, Representative immunoblot and quantification (n = 3) of FOXP3 protein in HEK293T cells transfected with FOXP3^WT and FOXP3^KDR followed by treatment with CHX (20 μg ml⁻¹) for 1 h, 2 h or 4 h. e, Representative immunoblot and quantification (n = 3) of FOXP3 protein amounts in HEK293T cells transfected with FOXP3WT and FOXP3KDR for 24 h, followed by treatment with 5 mM succinic acid for another 24 h. f, Representative flow images and quantification (n = 3) of CTV⁺ CD8⁺ T cells division in Dlst^Treg-iWT and Dlst^Treg-iKO iT_reg cells transfected with FOXP3^KDR or control plasmids (EV). g, Representative flow cytometry analysis of FOXP3 expression in Dlst^Treg-iWT and Dlst^Treg-iKO iT_reg cells transduced with a Thy1.1-FOXP3^KDR retrovirus and cultured for 4 days. h–j, Body weight changes (h), H&E staining (i) and histological analysis (j) of colon tissues in Rag1^−/− mice transferred with naive CD4⁺ cells from CD45.1 WT mice (no iTreg cells, n = 4) or with FOXP3^KDR-Thy1.1⁺ Dlst^Treg-iWT iT_reg cells (n = 3) or FOXP3^KDR-Thy1.1⁺ Dlst^Treg-iKO iT_reg cells (n = 4). Scale bar, 100 μm. Data are shown as mean ± s.d. and analyzed by one-way ANOVA with Tukey’s post hoc test.

Fig. 8.. Elevated succinate associate with T_reg cell dysfunction in IBD patients.

a–d, Representative flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ (a) and CD4⁺CD25⁺CD127^lo (d) T_reg cells and percentage (b) and FOXP3 MFI (c) of CD4⁺CD25⁺FOXP3⁺ T_reg cells in peripheral blood mononuclear cells (PBMCs) from patents with active Crohn’s disease (A-CD, n = 67) or during recovery phase of Crohn’s disease (R-CD, n = 23) and healthy control (HCs; n = 37). e, ELISA of plasma succinate level in patients as in a–d. f, Correlation of FOXP3 MFI with plasma succinate levels in all individuals with Crohn’s disease. g,h, Succinate level in the plasma of healthy controls (n = 37), active ulcerative colitis (A-UC, n = 25) and recovery ulcerative colitis (R-UC, n = 10) patients (g) or in the colon tissues of healthy controls (n = 4), individuals with A-UC (n = 4) and individuals with R-UC (n = 10) (h). i–l, Representative flow cytometry analysis of CD4⁺CD25⁺FOXP3⁺ (i) and CD4⁺CD25⁺CD127^lo (l) T_reg cells and percentage (j) and FOXP3 MFI (k) of CD4⁺CD25⁺FOXP3⁺ T_reg cells in PBMCs from individuals with ulcerative colitis. m, Correlative analysis of succinate level in plasma and colon tissue in individuals with ulcerative colitis. n, Correlative analysis of plasma succinate level and FOXP3 MFI in individuals with ulcerative colitis. o, RT–PCR of Ogdh, Dlst and Dld expression in CD4⁺CD25⁺CD127^lo T_reg cells sorted from the PBMCs of healthy controls (n = 33), individuals with A-CD (n = 67) and individuals with R-CD (n = 23). p–r, Correlation of Ogdh (p), Dlst (q) and Dld (r) expression with plasma succinate level in all individuals with Crohn’s disease. s–u, Correlative analysis of Ogdh (s), Dlst (t) and Dld (u) expression and FOXP3 MFI in all individuals with Crohn’s disease. Data are shown as mean ± s.d. and analyzed by one-way ANOVA with Tukey’s post hoc test (b,c,e,g,h,j,k,o) or Pearson’s correlation analysis (f,m,n,p–u).