A type 2 immune circuit in the stomach controls mammalian adaptation to dietary chitin

Abstract

Dietary fiber improves metabolic health, but host-encoded mechanisms for digesting fibrous polysaccharides are unclear. In this work, we describe a mammalian adaptation to dietary chitin that is coordinated by gastric innate immune activation and acidic mammalian chitinase (AMCase). Chitin consumption causes gastric distension and cytokine production by stomach tuft cells and group 2 innate lymphoid cells (ILC2s) in mice, which drives the expansion of AMCase-expressing zymogenic chief cells that facilitate chitin digestion. Although chitin influences gut microbial composition, ILC2-mediated tissue adaptation and gastrointestinal responses are preserved in germ-free mice. In the absence of AMCase, sustained chitin intake leads to heightened basal type 2 immunity, reduced adiposity, and resistance to obesity. These data define an endogenous metabolic circuit that enables nutrient extraction from an insoluble dietary constituent by enhancing digestive function.

Fig. 1.. Innate type 2 immune responses are triggered by gastric distension and dietary chitin.

(A) Dietary responses in wild-type (WT) or TKO mice on triple-reporter (YRS) backgrounds. (B) Representative stomach image (scale bar: 1 cm), stomach size, and luminal content, (C) relative Il25 and Il33 expression in stomach tuft (CD45⁻EpCAM⁺SiglecF⁺) and epithelial cells (CD45⁻EpCAM⁺), and (D) expression of R5 (Il5) and S13 (Il13) reporter alleles among stomach ILC2s (Yarg⁺, pre-gated on CD45⁺Lin⁻Thy1.2⁺) in WT and TKO mice on indicated diet for 24 hours. (E) Relative stomach gene expression in WT or TKO mice after Yoda1 administration or gastric distension. (F) R5 and S13 expression in stomach ILC2 after GsMTx4 administration on indicated diet for 12 hours. (G) R5 or (H) S13 expression in stomach and small intestine ILC2s after vehicle, Yoda1, and/or NmU administration. Data represent individual biological replicates except in (C), which are pooled from 2-3 mice and are presented as means±SD comprising two or more independent experiments (n≥3 mice per group). P-values were calculated by unpaired t test (B, F, and G), one-way ANOVA (E and H), or two-way ANOVA with Tukey’s multiple comparisons test (C and D). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS: not significant.

Fig. 2.. Sustained chitin intake promotes gastrointestinal remodeling and adipose ILC2 responses.

(A) Representative stomach histology (H&E) (scale bar: 100 μm), (B) Ki67-expressing stomach epithelial cells, (C) stomach tuft cells, and (D) representative image of small intestine and small intestinal length in indicated mice on diet for 2 weeks. Scale bar: 1 cm. (E) Eosinophils, total ILC2s, and R5-expressing ILC2s per gram of epididymal white adipose tissue (eWAT) in WT and TKO mice on diet for 2 weeks. Data represent individual biological replicates and are presented as means±SD comprising two or more independent experiments (n≥3 mice per group). P-values were calculated by unpaired t test (A, C, and D), one-way ANOVA (B), or two-way ANOVA with Tukey’s multiple comparisons test (E). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS: not significant.

Fig. 3.. AMCase is required for dietary chitin digestion.

(A) Immunostaining of AMCase-expressing chief cells in glandular stomach. Magenta: AMCase; blue: DAPI; green: autofluorescence. Scale bar: 50 μm. (B) Stomach ChiaRed⁺ (CR: AMCase reporter; CD45⁻EpCAM⁺CR⁺) or total (CD45⁻EpCAM⁺CR⁻) epithelial cells were sorted and analyzed for (C) relative expression of Gif, Clps, and Pgc by qPCR. (D) Analysis of chitin binding, digestion of soluble chitooligomer and insoluble crystalline chitin substrates, and production of GlcNAc reaction products by chitooligomer oxidase (ChitO) assay in stomach lavage. (E) Immunoblot of chitin-bound AMCase proteins. (+) indicates recombinant AMCase postive control. (F) Chitinase activity with soluble substrate in stomach lavage samples, with and without predepletion of AMCase using insoluble chitin. (G) Gastric fluid pH in mice fasted overnight after 2 weeks on indicated diet. (H) Digestion of insoluble colloidal chitin after 96-hour incubation with inactivated (heat-treated) or fresh stomach lavage from WT or CC mice. Scale bar: 500 μm. (I) ChitO assay for soluble GlcNAc reaction products in supernatants from insoluble particle digestion in (H). Data points represent individual biological replicates except in (C), which represent samples pooled from 3-4 mice. Data represent two or more independent experiments (n≥3 mice per group) and are presented as means±SD. *P<0.05, ***P<0.001, ****P<0.0001 (unpaired t test).

Fig. 4.. Stomach adaptation to dietary chitin is controlled by a type 2 immune circuit.

(A) Relative Chia1 stomach expression in WT and (B) indicated mouse strains (2 weeks on diet). (C) Chia1 expression in stomach tissue from WT and ILC2 deleter mice after IL-25 administration. (D) Breeding scheme and (E) relative Il4ra expression in ChiaRed⁺ or total stomach epithelial cells (EpCAM⁺) from ChiaRed and CR-Il4ra^fl/fl mice. (F) Percentage of ChiaReD⁺ (CR) chief cells out of total stomach epithelial cells in ChiaRed, CR-Il4ra^fl/fl, and CR-Stat6-KO mice on diet for 2 weeks. (G) WT and CC stomach size on diet for 2 weeks. (H) S13 (IL-13)-expressing ILC2s and tuft cells in stomach, (I) SI tuft cells, SI and adipose eosinophils, ILC2s, and S13⁺ ILC2s, (J) eWAT to body weight ratio, and (K) body weights of WT and CC mice on control or chitin diets as indicated. Data represent individual biological replicates except in (A, E, H, and I), which are pooled from 3-8 mice and are presented as means±SD comprising two or more independent experiments (n≥3 mice per group). P-values were calculated by unpaired t test (A to C and E to G) or two-way ANOVA with Tukey’s multiple comparisons test (H to K). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, NS: not significant.

Fig. 5.. Dietary chitin improves metabolic health in high fat diet-induced obesity.

(A) Experimental design. WT or CC mice were fed a high-fat diet (HFD) or chitin-containing HFD (CHFD) followed by metabolic cage analyses (CLAMS), glucose tolerance (GTT) and insulin tolerance tests (ITT). Part of the illustration was created with Biorender.com. (B) Body weights, (C) adiposity, and (D) food intake (24-hour average) at 8 weeks on diet. (E) Eosinophils and ILC2s in adipose tissue, and (F) tuft cells and ILC2s in SI and stomach tissues at 13 weeks on diet. (G) GTT and (H) ITT curves at 10 and 12 weeks, respectively. (I) Heat curve and averages over 24 hours in CLAMS cages at 8 weeks. Data represent individual biological replicates except in (B, E, and F), where each data point represents 8 pooled mice and are presented as means±SD (E, F) or means±SEM (B to D and G to I), comprising two or more independent experiments (n=8 mice per group). P-values were calculated by one-way ANOVA (C), two-way ANOVA (E, F), or two-way ANOVA with repeated measures (B, D, and G to I) with Tukey’s multiple comparisons test. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.