BACKGROUND AND AIMS: The trigger hypothesis opens the possibility of anti-flare-initiation therapies by stating that ulcerative colitis (UC) flares originates from inadequate responses to acute mucosal injuries. However, experimental evidence is restricted by a limited use of suitable human models. We thus aimed to investigate the acute mucosal barrier injury responses in humans with and without UC using an experimental injury model.

METHODS: A standardized mucosal break was inflicted in the sigmoid colon of 19 patients with UC in endoscopic and histological remission and 20 control subjects. Post-injury responses were assessed repeatedly by high resolution imaging and sampling to perform Geboes scoring, RNA sequencing, and injury niche microbiota 16S rRNA gene sequencing.

RESULTS: UC patients had more severe endoscopic post-injury inflammation than controls (p<0.01), an elevated modified Geboes score (p<0.05), a rapid induction of innate response gene sets (p<0.05) and anti-microbial peptides (p<0.01), and engagement of neutrophils (p<0.01). Innate lymphoid cells type 3 (ILC3s) markers were increased pre-injury (p<0.01) and ILC3 activating cytokines were highly induced post-injury resulting in an increase in ILC3 type cytokine IL-17A. Across groups, the post-injury mucosal microbiome had higher bacterial load (p<0.0001) and lower α-diversity (p<0.05).

CONCLUSIONS: UC patients in remission respond to mucosal breaks by an innate hyper-response engaging resident regulatory ILC3s and a subsequent adaptive activation. The post-injury IBD-like microbiota diversity decrease is irrespective of diagnosis suggesting that the dysbiosis is secondary to host injury responses. We provide a model for the study of flare initiation in the search for anti-trigger directed therapies.