RNF43 is a gatekeeper for colitis-associated cancer

Abstract

Somatic mutations in the tumor suppressor Ring finger protein 43 (RNF43) were frequently found in colitis-associated cancer (CAC) and related to the duration of chronic inflammation, but their significance in inflammation and inflammation-associated carcinogenesis remained elusive. We assessed the onset of RNF43 mutations at different stages of human CAC development by exome sequencing, and comprehensively characterized RNF43 loss-of-function-driven malignant transformation in mice by RNA sequencing, flow cytometry, immunohistochemistry, computational transcriptome-microbiome associations, and determined the underlying mechanisms by performing functional stem-cell derived organoid studies and fecal microbiota transfers. Mutations in RNF43 were frequent (12.9 %) in precancerous lesions of ulcerative colitis (UC) patients and eventually detectable in 24.4 % of CAC patients. In a bacterial-induced colitis mouse model, Rnf43 mutations caused invasive colorectal carcinomas by aggravating and perpetuating inflammation due to impaired epithelial barrier integrity and pathogen control. We could demonstrate that Rnf43 loss-of-function-mutations were even sufficient to cause spontaneous intestinal inflammation, resulting in UC-typical pathological features and subsequent invasive carcinoma development. In detail, mutant Rnf43 impaired intestinal epithelial and particularly goblet cell homeostasis in a cell-intrinsic manner, and caused dysbiosis. The altered microbiota composition induced epithelial DNA damage and spontaneous mucosal inflammation characterized by TGF-{beta}-activating dendritic cells and pro-inflammatory (IL-17+, IL-22+, TNF+) T cells. Over time, the continuous epithelial and goblet cell dysfunction, combined with pro-tumorigenic and pro-inflammatory microbiota, resulted in accumulated epithelial damage with transformation into inflammation-associated cancer in the presence of constitutive WNT signaling activation. We identified mutant RNF43 as susceptibility gene for UC and bona fide driver of CAC.