A gene regulatory network model evaluating the impact of individual factors in the hypertrophic switch

Chondrocytes undergoing hypertrophy show a major switch in phenotype underlied by a change in expression from the chondrocyte master gene, Sox9, to the osteoblastic one, Runx2. Strategies to stimulate or inhibit this switch are of use in bone and cartilage tissue engineering respectively, as well as in the prevention of ectopic hypertrophy in osteoarthritis.
We have constructed a literature based network comprised of 46 nodes and 161 interactions shown to play a part in chondrocyte hypertrophy. Network dynamics are simulated in discrete time through random updating by the use of additive functions to determine each node’s value. 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 states and the influence of individual nodes in this transition as a second measure of stability.
Our results show that the hypertrophic state, marked by Runx2 expression, has a larger attractor basin and is more stable to perturbation than the proliferative state characterized by Sox9. The added time resolution seems to favour the Runx2 phenotype. The results for single nodes in overexpression or knockout simulations show a certain asymmetry, indicating that factors that are necessary for maintaining a certain phenotype are not necessarily useful in inducing it.