P106 INFLAMMATORY BOWEL DISEASE CHARACTERIZATION OF SHARED AND UNIQUE ALTERATIONS IN IMMUNE CELLS, MOLECULAR PATHWAYS, AND TRANSCRIPTS

Abstract

Inflammatory Bowel Disease (IBD), characterized by chronic inflammation of the intestine, is propelled by alterations in the immune profile and inflammatory mediators within affected tissue. However, as IBD pathobiology is heterogeneous in nature, a more comprehensive understanding of common and distinct variations within individuals is required for more precise diagnostic methods and tailored therapeutic regimens. Using publicly available IBD transcriptomes from American and European patient cohorts, we aimed to elucidate alterations, both similar and distinct, in the immune profile, molecular pathways, and transcripts in affected tissue. Whole transcriptomes from healthy controls, active or non-active IBD cohorts were utilized (~500 patient endoscopic biopsies, NCBI GEO). Immune profiles were assessed by the core LM22 signature (CIBERSORT) (p<0.05). Pathway analysis was performed using Ingenuity Pathway Analysis (IPA, Qiagen) (p<0.05). In healthy human colon, immune cell profiling revealed a consistent abundance of B cells (plasma), T cells (CD4 memory resting), mast cells (resting) and macrophages (M2). In active-IBD involved tissue, we observed substantial alterations in the immune landscape including increased neutrophils, T CD4 memory activated cells, active dendritic cells, M0/M1 macrophages, and B naïve cells, as well as noticeably reduced T CD8 cells, Tregs, B memory cells, dendritic resting cells, and M2 macrophages (Fig 1). This immune signature was similarly observed across IBD cohorts with the exception of eosinophils, T CD4 resting cells, and Tγδ cells. In uninvolved tissue from active-IBD, the immune cell composition was similar to healthy control with the exception of elevated T CD4 memory activated cells and reduced Tregs (Fig 1). Next, relative to healthy control, IPA of differentially expressed transcripts from IBD cohorts revealed similarly altered pathways linked to bacterial signaling (TLR, LPS/IL-1, fMLP), inflammation (Th1/2 responses, chemokine and cytokine), inflammasome activation, and activated cell signaling (NFkB, ERK/MAPK, p38, iNOS, PI3K). However, pathways that differed among IBD cohorts included select signaling linked to inflammation (IL17A, OX40), growth (TGFb, PTEN, p53), and metabolic (PPARa, leptin) functions. At the individual transcript level, we found several novel transcripts consistently altered in IBD affected tissues among cohorts including KYNU, LPCAT1, CLDN8 as well as those differentially altered including MASPIN, FGFR2, AGRP, INSR. We determined the global immune cell landscape and molecular pathways that were similarly and differentially altered, highlighting the complex, heterogeneic nature of IBD pathobiology. Further utilization of this approach may provide clues for development of precise diagnostic and personalized therapeutic intervention for IBD patients. Figure 1. Alterations in the immune cell landscape in healthy control, active-IBD-involved, and active-IBD-noninvolved tissue samples. Graphs represent changes in relative abundances of subsets of B cells, T cells, dendritic cells, macrophages, mast cells, and neutrophils (GSE38713, *,#p<0.05, *compared to healthy control, #compared to active-IBD-involved) (CIBERSORT, cibersort.stanford.edu).