Microbiota-Derived Indole Metabolites Promote Human and Murine Intestinal Homeostasis through Regulation of Interleukin-10 Receptor

Abstract

Interactions between the gut microbiota and the host are important for health, where dysbiosis has emerged as a likely component of mucosal disease. The specific constituents of the microbiota that contribute to mucosal disease are not well defined. The authors sought to define microbial components that regulate homeostasis within the intestinal mucosa. Using an unbiased, metabolomic profiling approach, a selective depletion of indole and indole-derived metabolites was identified in murine and human colitis. Indole-3-propionic acid (IPA) was selectively diminished in circulating serum from human subjects with active colitis, and IPA served as a biomarker of disease remission. Administration of indole metabolites showed prominent induction of IL-10R1 on cultured intestinal epithelia that was explained by activation of the aryl hydrocarbon receptor. Colonization of germ-free mice with wild-type Escherichia coli, but not E. coli mutants unable to generate indole, induced colonic epithelial IL-10R1. Moreover, oral administration of IPA significantly ameliorated disease in a chemically induced murine colitis model. This work defines a novel role of indole metabolites in anti-inflammatory pathways mediated by epithelial IL-10 signaling and identifies possible avenues for utilizing indoles as novel therapeutics in mucosal disease.

Figure 1

Tryptophan metabolism is altered in murine and human colitis. A: Relative levels of indole metabolites in serum and whole colon tissue of mice receiving either water or 3% dextran sodium sulfate (DSS) for 7 days. Metabolites were measured by liquid chromatography–mass spectrometry and gas chromatography–mass spectrometry analysis. B: Condensed pathway of tryptophan metabolism to indole and indole derivatives. C: EC-HPLC analysis of indole metabolites in serum of mice receiving either water or 2.5% DSS for 9 days. D: Concentrations of indole metabolites in serum of mice receiving either water or 2.5% DSS for 9 days. E: Indole-3-propionic acid (IPA) levels in serum samples from healthy controls (Con), subjects with active ulcerative colitis (UC), and subjects with UC in remission (Rem) profiled by EC-HPLC. Data are expressed as means ± SEM (A, D, and E). n = 5 (A, mice per treatment group, and D, water treatment); n = 10 (D, DSS treatment); n = 20 (E, control and remission); n = 15 (E, UC). ^∗P < 0.05, ^∗∗P < 0.01, t-test and one-way analysis of variance. IAld, indole-3-aldehyde.

Figure 2

Indole metabolites improve barrier formation and induce IL-10R1 on intestinal epithelia. A: Real-time quantitative PCR (qPCR) of IL-10R1 transcript levels in T84 cells treated with indole-3-propionic acid (IPA) or indole-3-aldehyde (IAld) at varying concentrations for 6 hours. B: Human intestinal organoids treated with 1 mmol/L IPA over 24 hours. C: Western blot analysis of IL-10R1 levels in T84 cells treated with IPA at varying concentrations for 6 hours or in response to 1 mmol/L IAld over 24 hours. D: Transepithelial electrical resistance of T84 cells treated with IAld (1 μmol/L), IL-10 (10 ng/mL), or a combination of both was measured over 72 hours. E: qPCR of SOCS3 transcript in T84 cells treated with IAld for 12 hours, followed by treatment with IL-10 for 6 hours. Data are expressed as the average TEER ± SEM (D) and as means ± SEM (A, B, and E). n = 3 (D, samples, and E, experiments). ^∗P < 0.05, t-test, ^∗∗P < 0.01 versus untreated cells, two-way analysis of variance, and ^∗∗∗P < 0.001, t-test. +ctrl, 6-formylindole[3,2-b] carbazole at 1 μmol/L; TEER, transepithelial electrical resistance; Veh, vehicle.

Figure 3

IL-10R1 expression is not induced by cells lacking aryl hydrocarbon receptor (AHR) binding partner aryl hydrocarbon receptor nuclear translocator (ARNT). A: Lentiviral shRNA-mediated knockdown of ARNT (shARNT) in T84 intestinal epithelial cells relative to T84 cells containing a nontemplate control (shNTC). B: Transcript levels of cytochrome P450 family 1 subfamily A member 1 (CYP1A1) and cytochrome P450 family 1 subfamily B member 1 (CYP1B1) by real-time quantitative PCR (qPCR) in shNTC and shARNT T84 cells following indole-3-aldehyde (IAld) treatment for 12 hours. C: qPCR analysis of IL-10R1 expression in shNTC and shARNT T84 cells treated with 1 mmol/L IAld for 24 hours. D: Western blot analysis of IL-10R1 levels in ARNT knockdown T84 cells. Confluent monolayers of T84 cells were treated with IAld for 24 hours. E and F: Protein expression was quantified by densitometry (E) and normalized to β-actin. (F) qPCR of IL-10R1 transcript levels in shNTC and shARNT T84 cells treated with AHR inhibitor (AHRi, 10 μmol/L), IAld (1 mmol/L), or both for 24 hours. Data are expressed as means ± SEM (A–C, E, and F). n = 3. ^∗P < 0.05, ^∗∗P < 0.01 versus untreated shNTC, t-test. tx, treatment; Veh, vehicle.

Figure 4

Bacterial indole production induces IL-10R1. A: Indole production test of cultures from E. coli K12 WT, tnaA, and tnaB mutants; plus sign indicates presence of indole. B: EC-HPLC chromatogram of supernatants collected from E. coli K12 WT and E. coli ΔtnaA. C: T84 cells were treated for 24 hours with serially diluted supernatants from E. coli WT or E. coli ΔtnaA cultures grown to stationary phase, followed by media replenishment. Germ-free mice were colonized with E. coli WT or E. coli ΔtnaA for 2 weeks followed by euthanasia. D and E: Indole concentrations in cecal contents were validated by HPLC in vehicle and colonized mice (D), and RNA was extracted from colons for real-time quantitative PCR analysis of Il10r1 expression (E). Data are expressed as means ± SEM (D and E). n = 5 (D and E, PBS and E. coli ΔtnaA); n = 4 (D and E, E. coli WT). ^∗P < 0.05, ^∗∗P < 0.01, t-test. + ctrl, interferon-γ at 10 ng/mL; M, molecular marker; PBS, phosphate-buffered saline; sup, supernatant; tx, treatment.

Figure 5

Indole-3-propionic acid (IPA) improves dextran sodium sulfate (DSS) colitis outcomes. Eight- to 10-week–old C57BL/6 mice were administered water or 2.5% DSS ± IPA ad libitum for 9 days. DSS was then removed and mice were allowed to recover for 2 days as predetermined before euthanasia. A: Concentration of IPA in colons of mice receiving DSS ± IPA for up to 9 days. B: Colon length measured at time of euthanasia. C: Representative hematoxylin and eosin–stained colonic sections isolated from water-treated, water/IPA-treated, DSS-treated, and DSS/IPA-treated mice. D: Histologic score of hematoxylin and eosin–stained colonic sections. E–G: Colon tissue was homogenized and cytokines were measured in protein lysates by Meso Scale Discovery analysis [interferon-γ (IFN-γ; E); tumor necrosis factor–α (TNF-α; F); and IL-1β (G)]. Data are expressed as means ± SEM (A, B, D, and E–G). ^∗P < 0.05, ^∗∗P < 0.01, and ^∗∗∗P < 0.001, two-way analysis of variance and t-test. Original magnification, ×20.