Background: Intestinal mucus is a highly glycosylated gel-like substance that forms a protective barrier between epithelial cells and the contents of the gut lumen. MUC2 is the main component of intestinal mucus in both mice and humans. Humans also secrete low amounts of MUC5B. Interestingly, the non-intestinal mucin, MUC5AC, has recently been identified in mice infected with the nematode Trichuris trichiura and during DSS induced colitis. MUC5AC is canonically expressed in the respiratory tract, stomach, and eyes. Clostridioides difficile infection decreases colonic MUC2 expression, but whether C. difficile infection causes aberrant expression of MUC5AC in the intestinal mucosa to promote biofilm formation is unknown. We hypothesized that mice and humans infected with C. difficile would display altered mucin production including secretion of MUC5AC and shifted mucosal glycosylation. Methods & Results: Analysis of publicly available single cell RNA sequencing data demonstrated that MUC2 was the predominant intestinal mucin and is highly expressed in goblet cells. MUC5B was also expressed by goblet cells but to a lesser extent than MUC2. The expression of MUC5AC was extremely limited or absent. To identify if C. difficile altered the levels of MUC5AC, we analyzed human colonic tissue samples from 9 healthy individuals and 9 patients with recurrent C. difficile. Staining revealed crypt hyperplasia and immune cell infiltration indicative of inflammation in C. difficile infected tissues. Immunostaining of control tissue revealed high levels of MUC2 throughout the colonic crypts and moderate levels of MUC5B restricted to the base of colonic crypts. In C. difficile infected colon samples, we observed increased MUC5B and secretion of MUC5B from the crypts. Additionally, in 7 out of 9 C. difficile infected samples, we found multiple MUC5AC positive goblet cells. To assess alterations in N-glycosylation during infection, we performed imaging mass spectrometry on human tissue. C. difficile infected patients exhibited significantly altered mucosal N-glycan composition. The mucosa of C. difficile infected patients contained decreased high mannose glycans and increased branched and pauci-mannose glycans. There was a significant increase in glycans containing fucose and sialic acid residues and a concurrent decrease in afucosylated, asialylated glycans in the mucosa of C. difficile patients compared to controls. To confirm the ability of C. difficile to drive colonic MUC5AC expression we infected mice with C. difficile and analyzed MUC5AC levels (n=5/group). Similar to our patient samples, we observed MUC5AC expression in C. difficile infected mouse colon. Conclusions: These findings indicate that goblet cells can produce MUC5AC in response to bacterial-driven inflammation and that intestinal mucins and glycan profiles are altered during C. difficile infection.

Amy Engevik

Vanderbilt University Medical Center

Appearances