192 We then investigated the genes differentially expressed between inflamed colonic tissue from patients with CD and UC. In total, 1,466 genes were differentially abundant, of which 733 (50%) were overrepresented in CD and 733 (50%) in UC (FDR<0.05) (Supplementary Table S2). Pathway enrichment analysis showed the Notch-1 signaling pathway (e.g. HDAC9 and HEY1, Figure 2D) to be highly upregulated in CD compared to UC, whereas vitamin, cofactor and lipoprotein metabolism pathways (e.g. VNN1 and APOB, Figure 2D) were more pronounced in UC (Figure 2E), which corroborates previous findings.17-20 Cell type–deconvolution revealed that plasma cells, endothelial cells and Th2-lymphocytes were significantly increased in UC compared with CD (FDR<0.05, Supplementary Table S3), suggesting that distinct immunological mechanisms are involved in CD and UC. Mucosal microbiota composition is highly personalized The most common bacterial phylum observed across all tissue samples was Bacteroidetes (CD: 58%, UC: 58%, controls: 66%), followed by Firmicutes (CD: 27%, UC: 33%, controls: 23%) and Proteobacteria (CD: 14%, UC: 8%, controls: 9%). Interestingly, the overall mucosa-attached microbial composition was similar between colonic and ileal biopsies and independent of inflammation (Extended Data Figure S2). Only seven bacterial taxa were differentially abundant between patients and controls (Supplementary Tables S4-5), consistent with previous findings.13,21,22 Shannon diversity was significantly lower in samples from patients with CD compared to UC and non-IBD controls (P=2.75x10-16 and P=0.03, respectively, Figure 3A). This difference was still present when comparing only colonic biopsies from patients with CD to those from UC, indicating that this difference was not solely attributable to ileal CD (Extended Data Figure S3). Differences in microbial communities between tissue samples were evaluated by quantifying the Aitchison’s distance (Figure 3B-C). We obtained comparable findings when we externally validated our results using data derived from the HMP2 cohort (Extended Data Figure S4).13 Figure 3 | Overall characterization of mucosa-attached microbiota in patients with IBD and controls. a, Microbial alpha-diversity (Shannon index) was lowest in patients with CD (n=351) compared to patients with UC (n=289) and non-IBD controls (n=56). b, PCA plot based on Aitchison’s distances demonstrates the microbial dissimilarity of the mucosa-attachedmicrobiota (colors as in a). c, Microbial dissimilarity (Aitchison’s distances) comparison between non-IBD control, CD and UC. Microbial dissimilarity is highest in biopsies from patients with CD, followed by patients with UC and non-IBD controls. d, Microbial dissimilarity is higher in samples from different individuals when compared to paired samples from the same individual, which includes paired inflamed–non-inflamed tissue from ileum and colon (left panel, inter-colon: n=11,430, interileum: n=7,377, intra: n=203), paired colonic tissue samples from inflamed and non-inflamed areas (middle panel, inter-inflamed: n=7,372, inter-non-inflamed: n=8,369, intra: n=166) and paired ileal tissue samples from inflamed and non-inflamed areas (right panel, inter-inflamed: n=1,590, inter-non-inflamed: n=1,592, intra: n=73). e, Hierarchical analysis performed using an end-to-end statistical algorithm (HAllA) indicates the main phenotypic factors that correlate with intestinal mucosal microbiota composition. Heatmap color palette indicates normalized mutual information. Numbers or dots in cells identify significant pairs of features (phenotypic factors vs. bacterial taxa) in patients with IBD and controls. CD, Crohn’s disease. PCA, principal coordinate analysis. UC, ulcerative colitis. Chapter 6
RkJQdWJsaXNoZXIy MjY0ODMw