P032 Exploring the genetic sequence of action across the natural history of IBD

Abstract

Listen

Translate article

Abstract Background We have been very successful at identifying genetic variants associated with inflammatory bowel disease (IBD). From a single locus by the early 2000s, we now know about over 300 regions involved in IBD susceptibility. However, we face two main gaps of knowledge. First, we ignore the mechanisms by which they lead to disease. Second, we ignore their role through the natural history of the disease (e.g., prognosis). To address these gaps, we must integrate inferences from other omic techniques and characterize the genetic sequence of action that takes place across the distinct stages of the disease. Methods We have processed publicly available important studies of the genomics and transcriptomics of IBD to make inferences about the sequence of action of genes involved. In total, we processed the large IBDGC GWAS, five transcriptomics and two proteomic datasets (Table 1). We studied different phases of the disease: preclinical, debut and remission, both in Crohn’s disease and ulcerative colitis. We selected sets of genes associated with stages of IBD by the mentioned omics and looked for the overlaps. Last, we performed transcriptome-wide association studies (TWAS) to study the gene expression levels determined genetically in each individual. Results We report three main observations from this omic-wide exploration on the genetic sequence of action in IBD. First, we observed a lack of overlap in the gene lists that are relevant in each stage (Figure 1). For instance, just 6.4% of genes discovered by GWAS are differentially expressed at the time of diagnosis and/or remission e.g., NOD2, C2, CCL2, IL27, CSF2, CCL11 and BACH2. In turn, gene signatures discovered through transcriptomics are not particularly enriched for GWAS signals. Second, the genetic risk of each individual, measured through polygenic risk scores, does not correlate with the clustering of patients detected molecularly using gene expression. Third, TWAS analysis reveals an intriguing twist, since we observe large differences between the genetically determined component and the actual gene expression observed in each stage. Conclusion Although the genetic makeup of an individual influences expression of key genes involved in IBD, many other genes not detected in genetic studies (GWAS) conform the backbone of molecular alterations present at the time of diagnosis. This global observation challenges the conventional assumption, implying that different actors play a leading role in each stage. Moreover, genetic predisposition is not a strong determinant of the molecular heterogeneity observed in patients across stages. We conclude that the gene regulation dynamics across the natural history of IBD are more intricate than usually assumed.