Metformin attenuates colitis via blocking STAT3 acetylation by reducing acetyl-CoA production

Abstract

Background and aims: While metformin has been shown to alleviate dextran sulfate sodium (DSS)-induced colitis in murine models, the mechanisms underlying its anti-inflammatory and barrier-restorative effects remain poorly defined. This study investigates the role of acetyl coenzyme A (acetyl-CoA)-dependent STAT3 acetylation in mediating metformin's therapeutic actions, with the goal of identifying novel molecular targets for ulcerative colitis (UC) treatment.

Methods: Acute colitis was induced in wild-type C57BL/6J mice via oral DSS administration, followed by daily intraperitoneal metformin treatment. Intestinal inflammation, barrier integrity, and STAT3 signaling were assessed using histopathology, western blotting, and transmission electron microscopy. To validate STAT3's critical role in colitis pathogenesis, intestinal epithelium-specific STAT3 knockout mice were employed, enabling targeted investigation of STAT3 acetylation and its regulation by metformin.

Results: Metformin attenuated DSS-induced colitis by suppressing pro-inflammatory cytokines (TNF-α, IL-6, IL-1β), reducing epithelial apoptosis, and restoring tight junction proteins (ZO-1, E-cadherin, Occludin). Mechanistically, metformin reduced acetyl-CoA levels, thereby inhibiting STAT3 acetylation and downstream pathway activation. The pivotal role of STAT3 in colitis progression was confirmed using STAT3 knockout mice, as the therapeutic effects of metformin were significantly diminished in the absence of STAT3-mediated inflammatory signaling.

Conclusion: This study identifies acetyl-CoA-dependent STAT3 acetylation as a novel mechanism through which metformin ameliorates intestinal inflammation and barrier dysfunction. These findings not only advance our understanding of metformin's immunomodulatory properties but also highlight the therapeutic potential of targeting acetyl-CoA metabolism in UC.

Keywords: Colitis; Metformin; STAT3; STAT3 acetylation; acetyl-CoA.

Graphical abstract

Fig. 1

Metformin alleviates the symptoms of DSS-induced colitis. (A) Schematic diagram of the experimental procedure: Wild-type (WT) mice were randomly assigned to six groups: Water, Met-low, Met-high, DSS, DSS + Met-low, and DSS + Met-high (n = 6). Mice were administered metformin (100 mg/kg or 200 mg/kg) or PBS intraperitoneally for 7 days, followed by water or 3 % DSS. (B) Changes in body weight of the mice starting from day 1. (C) Variations in the DAI. (D) Representative structure and dimensions of the colon across the different groups. (E) Comparison of colon length across the different groups. (F-G) Comparison of representative H&E staining morphology and pathological scores across the groups. (H) Comparison of mRNA levels of cytokines TNF-α, IL-6, and IL-1β in colon tissues of mice.

Fig. 2

Metformin improves impaired colonic barrier. (A) Representative images of AB-PAS and IHC staining for MUC-2. (B) Representative images of IF staining for ZO-1 in colon tissue, along with quantification of ZO-1 fluorescence intensity. (C) TUNEL staining of colonic tissue, followed by statistical evaluation of TUNEL-positive cells. (D) Electron microscope images illustrating mouse colonic tissue from different treatment groups, with a magnified view highlighting tight junctions. (E) Protein expression levels of E-cadherin, Occludin, Bcl-2, and Bax in colonic tissues from WT mice. (F) Western blot analysis of E-cadherin, Occludin, Bcl-2, and Bax protein expression in NCM460 cells. The cells were treated with metformin (100 μM or 200 μM for 24 h) followed by LPS treatment (1 μg/mL for 12 h). (G) Immunofluorescence analysis of ZO-1 expression in NCM460 cells. (H) Apoptosis rate assessment in NCM460 cells using flow cytometry. (I) Detection of cleaved caspase-3 protein in NCM460 cells by Western blot analysis.

Fig. 3

Metformin mitigated gut microbiota dysbiosis. (A) PCoA at the operational taxonomic unit (OTU) level demonstrated the differential distribution of microbial communities in the feces of WT mice (n = 6 per group). (B) Histogram illustrating the relative abundance of each group at the phylum level. (C) Community heatmap analysis at the genus level. (D) Linear discriminant analysis (LDA): Displays the most significantly abundant classifications of colon microbiota in each group (LDA > 2). (E) Relative abundance of Bacillota at the phylum level. (F) Relative abundance of Bacteroidota at the phylum level. (G) Ratio of the relative abundance of Bacillota to Bacteroidota. (H-J) Relative abundance of Duncaniella, Lactobacillus, and Acetivibrio at the genus level.

Fig. 4

Metformin attenuates colitis by suppressing the STAT3 signaling pathway. (A) Experimental flow chart: STAT3^fl/fl and STAT3^ΔIEC mice (n = 6) were administered metformin (200 mg/kg) or PBS intraperitoneally, followed by continuous delivery of 3 % DSS in their drinking water for 7 days. (B-C) Changes in body weight and DAI of mice were monitored from day 1. (D-E) A comparison of colon length and morphology in mice. (F) Representative images of H&E morphology, along with IHC analysis of AB-PAS and MUC-2. (G) Histological scores of colon tissue from mice in various treatment groups. (H-I) Representative immunofluorescence images of ZO-1 and TUNEL staining. (J) Scanning electron microscopy images of mouse colonic tissue sections. (K) WB analysis of colonic protein expression in mice, including STAT3, p-STAT3^Y705 , E-cadherin, Occludin, Bcl-2, and Bax. (L) IHC analysis of p-STAT3^Y705 expression in mouse colonic tissue. (M) WB analysis of protein expression in NCM460 cells. The cells were transfected with siSTAT3 or siNC using Lipofectamine 3000, followed by treatment with or without metformin (200 μM for 24 h) and stimulation with LPS (1 μg/ml for 12 h). (N-P) Representative IF images showing ZO-1, p-STAT3^Y705 , and STAT3 expression in cells. (Q) WB analysis of relevant protein levels in NCM460 cells. The cells were cocultured with LPS (1 μg/mL for 12 h) and transfected with a STAT3 overexpression plasmid (STAT3^WT) or negative control (NC), with or without metformin pretreatment (200 μM for 24 h). (R) WB assessment of relevant protein levels in NCM460 cells. The cells were treated with DMSO or Colivelin TFA (25 μg/ml for 13 h, Med Chem Express, #HY-P1061A) following treatment with LPS (1 μg/ml for 12 h) and metformin (200 μM for 24 h).

Fig. 5

Metformin supresses STAT3 acetylation to alleviate intestinal inflammation. (A-B) WB and IF analysis of protein expression in NCM460 cells treated with Ex527 (50 μM for 24 h) or DMSO, followed by LPS stimulation. (C-D) WB and IF analysis of protein expression in NCM460 cells treated with metformin (200 μM for 24 h) or SRT1720 (5 μM for 12 h, Med Chem Express, #HY-10532), followed by LPS stimulation. (E) Schematic of experimental design. STAT3^fl/fl and STAT3^ΔIEC mice (n = 5) were intraperitoneally administered Ex527 (10 mg/kg, 7 days) or DMSO, followed by metformin (200 mg/kg) and 3 % DSS treatment for 7 days. (F) Body weight changes and DAI scores. (G) Colon length and macroscopic morphology. (H) Representative H&E-stained colon sections and IHC analysis of AB-PAS and MUC-2. (I) Histopathological scoring of H&E-stained colon sections. (J) Statistical analysis of Western blot results. STAT3^fl/fl and STAT3^ΔIEC mice (n = 5) were intraperitoneally administered Ex527 (10 mg/kg, 7 days) or DMSO, followed by metformin (200 mg/kg) and 3 % DSS treatment for 7 days. (K) Statistical analysis of Western blot results. STAT3^WT - or STAT3^K685 -transfected NCM460 cells were treated with Ex527 (50 μM) or DMSO for 24 h, followed by LPS + Met. (L) Scanning electron microscopy (SEM) images of colonic epithelium. (M) WB analysis of colonic protein expression in mice. (N) IHC staining of p-STAT3^Y705 in STAT3^fl/fl mouse colons. (O) IF staining of ac-STAT3^K685 in STAT3^fl/fl mouse colons. (P-R) IF images of ZO-1, p-STAT3^Y705 , and ac-STAT3^K685 in NCM460 cells treated with Ex527 (50 μM) or DMSO for 24 h, followed by LPS + Met. (S) WB analysis of protein levels in STAT3^WT - or STAT3^K685 -transfected NCM460 cells treated with Ex527 (50 μM) or DMSO for 24 h, followed by LPS + Met.

Fig. 6

Elevated acetyl-CoA levels exacerbate UC by promoting STAT3 acetylation. (A) Schematic representation of the experimental model. WT mice (n = 6) were gavaged with acetate (500 mg/kg) or PBS for 3 weeks, followed by treatment with metformin (200 mg/kg) and 3 % DSS for 7 days. (B) Relative acetyl-CoA levels in mice. (C-D) Representative IF images of ac-STAT3^K685 and IHC images of p-STAT3^Y705 in mice. (E) Alterations in body weight and DAI. (F-G) Gross morphology and colon length measurements in mice. (H) Representative images of H&E staining for morphological analysis and IHC analysis of AB-PAS and MUC-2 expression. (I) Histological evaluation of colonic tissue based on morphological examination of colon sections. (J-K) Representative immunofluorescence images of ZO-1 and TUNEL staining, along with quantitative analysis. (L) Representative electron microscopy images of colonic tissue segments. (M) WB analysis of colonic protein expression in mice. (N) ELISA analysis of relative intracellular acetyl-CoA levels (n = 3). NCM460 cells were exposed to acetate (5 mM) or PBS for 24 h, followed by treatment with metformin and LPS. (O) WB analysis of relevant protein expression in cells treated with acetate or PBS. (P-R) Representative immunofluorescence images of ZO-1, p-STAT3Y⁷⁰⁵ , and ac-STAT3^K685 in cells.