Impaired cellular RNA editing as a cause of inflammation in inflammatory bowel disease

PROJECT SUMMARY AND ABSTRACT: The etiology of inflammatory bowel disease (IBD), defined as Crohn's disease (CD) or ulcerative colitis (UC), remains elusive. Although genetic susceptibility plays a vital role in IBD, with over 250 GWAS risk loci, the mechanisms underlying their genetic impact on IBD are still poorly understood. Our surprising discovery reveals

that 78 of the IBD susceptibility loci are positioned adjacent to long double-stranded RNAs (dsRNAs), collectively accounting for about 20% of genetic heritability. This insight is founded on a comprehensive understanding of dsRNA editing and sensing mechanisms, supported by human genetic analyses. To prevent undesired immune

responses triggered by abundant cellular long dsRNAs, the ADAR1 protein performs adenosine-to-inosine (A- to-I) RNA editing on cellular dsRNAs, marking them as "self" and preventing dsRNA-induced interferon (IFN) responses. Notably, we observed that mice lacking ADAR1 RNA editing cannot survive, but exhibit full lifespan

upon eliminating the dsRNA sensor MDA5. Our recent work indicates that insufficient dsRNA editing is associ- ated with presumed dsRNA-mediated inflammation in various autoimmune and inflammatory diseases, including IBD. We discovered that genetic regulatory variants of RNA editing, referred to as edQTLs, are highly enriched

in the GWAS susceptibility loci of common immune-related diseases, including IBD. Further, GWAS-defined risk variants are linked to reduced levels of adjacent long dsRNAs, increasing their susceptibility to MDA5 activation. In summary, these findings highlight a crucial role of dsRNA editing and sensing in IBD etiology.

Our research objective is to develop effective approaches to systematically evaluate the genetic effects of IBD risk variants on dsRNA editing, allowing construction and validation of a predictive polygenic model of dsRNA-mediated disease risk, known as "dsRNA burden", in both colon enteroids and in vivo animal models.

First, we will leverage genetic data from IBD and edQTLs to prioritize dsRNA loci for IBD susceptibility. Subse- quently, we will systematically assess how risk and protective alleles of IBD GWAS variants affect cis-linked dsRNA editing, achieving direct functional validation of causality. Second, we will develop and validate a predic-

tive model to assess an individual's dsRNA burden, encapsulating dsRNA-associated IBD in patient-derived enteroid lines. This model will aid in stratifying IBD patients, particularly those with impaired dsRNA editing and heightened dsRNA sensing, contributing to MDA5-dependent chronic inflammation. Finally, we will recapitulate

a dsRNA burden system in an in vivo animal model to replicate inflammation and pathology in a native environ- ment. In summary, this work bridges a significant knowledge gap in understanding the etiology of IBD and pro- vide essential mechanistic insights and therapeutic potential for individuals with genetically dysregulated dsRNA

editing, leading to chronic IFN responses in IBD.