P040 Prediction of endoscopic features of disease from single-cell RNA sequencing data in sigmoid colon of patients with ulcerative colitis

Abstract

Abstract Background Patients suffering from ulcerative colitis (UC) have reduced life qualities and an increased risk of developing cancer. Current treatment options are limited and have a ceiling effect with remission rates of ~35%. A possible reason is the multi-factorial aetiology, which are associated with an overactivation of the immune system and a dysbiosis of the gut microbiome. An important step towards a better understanding is the characterization of epithelial lesions. Their severity is classified by the Mayo score system based on their macroscopic appearance. While previous studies analysed cellular changes within the colon in UC patients, a detailed characterization of different UC-associated lesions are still missing, which is the aim of our study in collaboration with Sanofi Pharmaceuticals. Methods We recruit patients with active disease at our endoscopy unit and collect different biopsies from multiple regions within these patients to compare the reproducibility of single cell RNA sequencing (scRNAseq) results of similar appearing lesions within and between patients. For the sample preparation, we include a cellular enrichment step for immune, epithelial and stromal cells before performing 10x scRNAseq to obtain equal numbers for each cell fraction from every sampled area. Results Our cellular isolation strategy allows successful recovery of fibroblast, epithelial and immune cell fractions from each lesion. All expected epithelial cell types as previously described are detected in our dataset as well. Moreover, our experimental setup to obtain comparable numbers of fibroblasts, epithelial and immune cells enables us now to investigate cell-cell communication differences within different Mayo score lesions. For example, CXCL signal increased with the severity of macroscopic lesions and involves multiple cell types. Conclusion Our preliminary results promise exciting insights into cell-cell communications at an unprecedented depth, which identifies aberrant signalling pathway that may contribute to treatment failure in the clinics.