Computational modelling of the cellular interplay in Rheumatoid Arthritis. Deciphering the role of innate and adaptive immunity in cartilage destruction and bone erosion

Immune dysregulation was first implicated in the pathogenesis of Rheumatoid Arthritis (RA) by the discovery of anti-immunoglobulin G (IgG) antibodies known as rheumatoid factors. However, concepts of how immune responses contribute to disease have evolved dramatically over the last 50 years. Many cells and their cytokines play critical roles in the development of RA. The synovial compartment is infiltrated by leukocytes and the synovial fluid is inundated with pro-inflammatory mediators that are produced to induce an inflammatory cascade, which is characterized by interactions of fibroblast-like synoviocytes with the cells of the innate immune system, including monocytes, macrophages, mast cells, dendritic cell as well as cells of adaptive immune system such as T cells and B cells. The fulminant stage contains hyperplastic synovium, cartilage damage, bone erosion, and systemic consequence.

The objective of my project is to construct a computational model able to decipher the interplay between cells of the innate and adaptive immunity in RA, that eventually leads to bone and cartilage breakdown.

To do so, we will start by creating separate maps for T cells, B cells, macrophages, fibroblasts, osteoblasts and osteoclasts. We will use different data mining tools, appropriate data bases such as KEGG (Kanehisa et al, 2000), REACTOME (Fabregat et al, 2018) as well as internal data generated within Sanofi. We will exploit the graph editor CellDesigner (Funahashi et al, 2003) and the platform Minerva (Gawron et al, 2016) for automatic annotation and reference of the cell specific maps. We will benefit greatly from a global, fully annotated RA specific map (Singh et al, 2018, Singh et al, 2020). This map features interactions implicated in RA coming from various cell types, but due to the extensive annotations the user can opt for cell specific interactions and extract the corresponding network. We are also going to use public datasets of expression data (microarrays, RNAseq, RNAseq single cell), data concerning metabolic pathways from MetaCyc and Sanofi’s proprietary datasets to enrich and expand existing pathway resources. These maps will be used to generate cell specific dynamic models using the tool CaSQ (Aghamiri et al, 2020). The next step is the creation of a multicellular model to understand how the different cells interact, contributing to the emergent behavior of the system. We will prioritize signature pathways for each cell type and combine them to build a logical model that will represent the intra- and intercellular relationships.