Vancomycin eliminates gut deoxycholic acid, restoring ER proteostasis in ILC2s and relieving colitis
Qiuheng Tian, Han Liu, Xiang Gu, Jing Shen, Xi Yuan, Mengqi Zheng, Yunjiao Zhai, Yatai Chen, Penghu Han, Yangchun Ma, Wei Xin, Hongyue Ma, Yu Li, Sihan Wang, Lei Guo, Detian Yuan, Yanbo Yu, Shiyang Li
Qiuheng Tian, Han Liu, Xiang Gu, Jing Shen, Xi Yuan, Mengqi Zheng, Yunjiao Zhai, Yatai Chen, Penghu Han, Yangchun Ma, Wei Xin, Hongyue Ma, Yu Li, Sihan Wang, Lei Guo, Detian Yuan, Yanbo Yu, Shiyang Li
View: Text | PDF
Research Article Gastroenterology Immunology

Vancomycin eliminates gut deoxycholic acid, restoring ER proteostasis in ILC2s and relieving colitis

Abstract

Ulcerative colitis (UC) remission is marked by gut microbiota restructuring, but how microbial metabolites influence immune-mediated tissue repair is unclear. Here, we demonstrate that oral vancomycin alleviates colitis symptoms in murine models, mirroring its clinical efficacy in inducing remission in patients with UC. Mechanistically, vancomycin’s therapeutic effect is achieved by reducing deoxycholic acid (DCA). We reveal that DCA impairs mucosal repair driven by group 2 innate lymphoid cells (ILC2s) by inducing ER stress through direct binding to thioredoxin-related transmembrane protein 2 (TMX2). This interaction disrupts TMX2’s role in protein folding, triggering unresolved unfolded protein response via hyperactivation of PERK/eIF2α signaling, which suppresses the production of pro-healing molecules by ILC2s. Pharmacological inhibition of PERK phosphorylation restores ILC2 function and accelerates colitis resolution. Our work uncovers a pathogenic microbiota/DCA/ILC2 axis that obstructs mucosal healing and positions vancomycin as a targeted strategy to eliminate DCA, thereby promoting UC remission.

Authors

Qiuheng Tian, Han Liu, Xiang Gu, Jing Shen, Xi Yuan, Mengqi Zheng, Yunjiao Zhai, Yatai Chen, Penghu Han, Yangchun Ma, Wei Xin, Hongyue Ma, Yu Li, Sihan Wang, Lei Guo, Detian Yuan, Yanbo Yu, Shiyang Li

×

Figure 5

DCA disrupts protein folding in ILC2s by binding to TMX2 in the ER.

Options: View larger image (or click on image) Download as PowerPoint
DCA disrupts protein folding in ILC2s by binding to TMX2 in the ER. (A) ...
(A) ILC2s were treated with DCA-iF647 (1 μM, red) for 15 minutes, followed by staining with ER-Tracker (green) or Mito-Tracker (green), and Hoechst 33342 (blue). Representative confocal laser scanning microscopy (LSCM) images. (B and C) Protein misfolding was quantified using the molecular rotor Proteostat with affinity for aggregated proteins. Increased protein aggregation causes the dye to stop spinning and emit fluorescence. We analyzed misfolding in ILC2s treated with DCA (100 μM) for 24 hours. (B) Representative confocal LSCM images. (C) Relative mean intensity per cell was plotted and calculated after blinding. n = 36 for DMSO-treated ILC2s, n = 34 for ILC2s treated with 100 μM DCA. The experiment was repeated 3 times. (D) Schematic of DCA-protein pulldown assays in cell lysates in combination with mass spectrometry analysis. (E) Representative immunoblot of TMX2 in proteins pulled down from ILC2 lysates by biotin-DCA (100 μM). (F) Heatmap of TMX family genes by performing RNA-Seq analysis with purified ILC2s. (G and H) Molecular docking showing the binding of DCA to TMX2. (I and J) Surface plasmon resonance assay for the affinity between different concentrations of DCA and purified TMX2 protein. RU, resonance unit. (K and L) Reductase activity of TMX2. (K) Experimental strategy. (L) Purified TMX2 protein with or without DCA (200 μM) was incubated with insulin, and the ability to reduce insulin disulfide bonds was measured. The absorbance at 595 nm was monitored every 2.5 minutes. n = 3 wells per group. Data shown as mean ± SD. Statistical analysis was performed using unpaired 2-tailed t test. ****P < 0.0001.