A Gene Regulatory Network Model to Assess the Stability of the Cartilage Phenotype

Introduction
Chondrocyte hypertrophy entails the switching of a genetic program driven by Sox9 to one under control of the osteoblast master regulator Runx2. The switch is a prerequisite step in the bone forming process (endochondral ossification) during development and in postnatal fracture repair of larger bone defects. However, this switch can also be detrimental in tissue engineered cartilage constructs and in osteoarthritis development [Saito, 2010]. Therefore, a detailed model of the pathways that can facilitate, or inhibit, this phenotypic switch will lead to a more profound understanding of these processes and provide hints as to how to manipulate them. Methods
The model formalism accommodates the qualitative information that is typically available in developmental studies. The literature based network comprises 46 nodes and 161 interactions, shown to be important in endochondral ossification. To simulate network dynamics in discrete time the normalized value of each gene is determined by additive functions where all interactions are assumed to be equally powerful. Furthermore, each species is represented by a fast variable (activity level, as determined by post translation modifications) which is assumed to be in equilibrium with a slow variable (mRNA) at all times.
Through a Monte Carlo approach the importance of each node in the stability of chondrocytic phenotypes (proliferating, hypertrophic) is assessed in random initial conditions. A perturbation analysis of the stable states is used to determine the transition likelihood between them as a second measure of stability. Results
Both measures of stability indicate that the hypertrophic (Runx2 driven) state is more stable than the proliferating one driven by Sox9. The results for the second measure are given in Fig.1. This higher stability seems to be partly conferred by faster reactions that favour the hypertrophic phenotype. In addition, the results point out that some transcription factors are necessary for the induction of a certain phenotype, whereas other transcription factors are required to maintain the phenotype, but are not necessary capable of inducing it. Discussion
These results may relate to the difficulty experienced by researchers in maintaining a stable cartilage phenotype in culture and the occurrence of ectopic hypertrophy in osteoarthritis. By analysing the effect of changes to individual nodes, strategies to stabilise the proliferating phenotype can be developed. Overall, the model allows the importance of several important factors in the fate decision of mesenchymal cells to be quantitatively assessed based mainly on topological information.