P020 Mapping the changing intercellular communication and its downstream effect in Ulcerative Colitis

Potari-gul, L.(1);Modos, D.(1,2);Turei, D.(3);Valdeolivas, A.(3);Madgwick, M.(1,2);Saez-Rodriguez, J.(3,4);Korcsmaros, T.(1,2);

(1)Earlham Institute- Norwich Research Park, Organisms and Ecosystems, Norwich, United Kingdom;(2)Quadram Institute Bioscience- Norwich Research Park, Gut Microbes and Health Programme, Norwich, United Kingdom;(3)Heidelberg University, Faculty of Medicine and Heidelberg University Hospital- Institute of Computational Biomedicine- Bioquant, Heidelberg, Germany;(4)RWTH Aachen University, Faculty of Medicine- Joint Research Centre for Computational Biomedicine JRC-COMBINE, Aachen, Germany


Intercellular communication is essential for growing and differentiating in multicellular organisms by transducing the signal from cell to cell. Despite its importance, the molecular background is less discovered due to the lack of data. This gap has started to be addressed with the appearance of single-cell omics approaches providing an insight among others into the gene expression of individual cells.


We have developed a method to predict and compare cell-cell signalling interactions using single-cell RNAseq data from colon biopsies. Transcriptomic data alone is not capable of connecting the cells, a reliable network resource is needed to mediate the signal via protein-protein interactions between the source and target cells. Here we used OmniPath - a resource providing not only intra- and intercellular interactions but also annotations of proteins involved in the interplay of cells - to reconstruct signalling networks. We examined intercellular communication among five cell-types (regulatory T cell, macrophage, dendritic cell, goblet cell and myofibroblast) in healthy colon and during Ulcerative Colitis.


Our analysis shows that there are significant differences in the type of cell-cell communication (ligand-receptor connections, adherens junctions, etc.) between the healthy and Ulcerative Colitis (UC) conditions, and these differences lead to altered downstream effects in the signal receiving cell. In both conditions, the ligand-receptor and adhesion connections were overrepresented, however cell junctions were less abundant in UC. Regarding the communication among the five cell-types, in healthy condition, cells are tightly connected to dendritic cells while in diseased condition to regulatory T cells. Focusing on ligand-receptor interactions between myofibroblasts and regulatory T cells, our pipeline identified the MAPK, Toll-like receptor (TLR) 2/6 and TLR 7/8 pathways enriched downstream in healthy conditions. In contrast, TLR3 and TLR4 pathways were affected by the myofibroblast in Ulcerative Colitis.


We found key intercellular mechanisms leading to well-defined differential pathway activation profiles. We showed that in uninflamed UC condition myofibroblasts disrupt the anti-inflammatory effect of regulatory T cells. Our pipeline is able to predict and analyse cell-cell interactions and their downstream effects and to highlight the differences in healthy and diseased states.