Using systems biology to investigate how age-related changes in TGFβ signalling alter pro-inflammatory stimuli

Abstract

Osteoarthritis (OA) is a degenerative condition caused by dysregulation of multiple molecular signalling pathways. This dysregulation results in damage to cartilage, a smooth and protective tissue that enables low friction articulation of synovial joints. Matrix metalloproteinases (MMPs), especially MMP13, are key enzymes in the cleavage of type II collagen which is a vital component for cartilage integrity. Various stimuli have been identified as inducers of MMP expression such as excessive load, injury and inflammation. Although previously considered a non-inflammatory arthritis, recent research has shown that inflammation may play an important role in OA development. A novel meta-analysis of microarray data from OA patients was used to create a cytoscape network representative of human OA. This enabled the identification of key processes in OA development, of which inflammation was prominent. Examining various different signalling pathways highlighted a role for transforming growth factor beta (TGFβ) in protecting against pro-inflammatory cytokine-mediated MMP expression. Indeed, TGFβ plays key roles in all facets of cartilage biology including development and maintenance of cartilage integrity. With age there is a change in the ratio of two TGFβ type I receptors (ALK1/ALK5), a shift that results in TGFβ losing its protective role in cartilage homeostasis. Instead, TGFβ promotes cartilage degradation and this correlates with the spontaneous development of OA in murine models. However, the mechanism by which TGFβ protects against pro-inflammatory responses and how this changes with age has not been extensively studied. Mathematical modelling has previously revealed how stochastic changes in TGFβ signalling during ageing led to the upregulation of MMPs. I have expanded the TGFβ section of this model to incorporate the pro-inflammatory stimulus interleukin-1 (IL-1) + oncostatin M (OSM) in order to investigate how TGFβ mediates MMP repression, specifically MMP-13. TGFβ signalling appears to interact with the activator protein 1 (AP-1) complex, which has an important role in MMP upregulation. However, the model indicates this interaction alone is insufficient to mediate the full effect of TGFβ, predicting it may also reduce MMP-13 mRNA stability. Furthermore, the model enabled me to predict how age alters these interactions; it suggested TGFβ would provide limited repression with a prolonged inflammatory response. Combining the modelled genes with the microarray network provided a global overview of how alterations in one pathway can affect others and lead to OA development. This study therefore demonstrates the power of combining computational biology with experimentally-derived data to provide insight into the importance of TGFβ signalling, and how age-related changes can lead to cartilage damage and OA development.