DOP04 Single cell RNA sequencing identifies distinct intestinal inflammation patterns between primary sclerosing cholangitis-associated colitis and Ulcerative Colitis

Uniken Venema, W.(1);Bangma, A.(1);Van der Wijst, M.(2);Kats-Ugurlu, G.(1);Bjork, J.(1);Vich Vila, A.(1);Weersma, R.(1);Festen, E.(1)

(1)University Medical Center Groningen, Department of Gastroenterology and Hepatology, Groningen, The Netherlands;(2)University Medical Center Groningen, Department of Clinical Genetics, Groningen, The Netherlands

Background

Primary sclerosing cholangitis (PSC) is an inflammatory disorder of the bile ducts. The etiology of PSC is unknown, but it is hypothesized that intestinal barrier dysfunction, as seen in inflammatory bowel disease (IBD), plays a role. Roughly 75% of PSC patients have concomitant IBD (PSC-IBD). PSC-IBD is phenotypically different from ulcerative colitis (UC) with predominantly right-sided disease and a higher risk for colorectal cancer. In this study we aim aim to find probable distinct pathomechanisms for PSC-IBD, by comparing gut mucosal biopsies between PSC-IBD and UC using single-cell RNA sequencing.

Methods

47 gut mucosal samples from the colon of subjects with either PSC-IBD (n=24), UC (n=18) or non-IBD control (n=5) were collected, from which 28 were paired inflamed and non-inflamed. Whole biopsies were cryopreserved and dissociated into single cells using collagenase digestion. Library preparation was done using the 10x Genomics system and subsequent sequencing was performed on an MGI2000 sequencer. The ‘Seurat’ R package was used for analysis.

Results

A total of 75.078 high quality cells identified 38 distinct cell types. No differences in cell type composition were observed between PSC-IBD and UC. We did see different cell type composition and gene expression between inflamed and non-inflamed samples. For example, in PSC-IBD specifically, an enterocyte subtype defined by DUOX2-expression showed inflammatory pathways upon inflammation. UC inflammation, on the other hand, was characterized by involvement of BEST4+ enterocytes and inflammatory fibroblasts. In addition, activated B cells and IgA plasma cells expressed stress-related genes in PSC, but not in UC inflammation. 

Conclusion

We show that intestinal inflammation in PSC-IBD is characterized by distinct, cell-specific gene expression patterns as compared to UC. This highlights differential cell types mediating inflammation between these IBDs. Our study provides insight in cellular mechanisms underlying intestinal disease in PSC, and may serve as a starting point for further studies, for example on the functions of DUOX2+ enterocytes.