Network modeling and analysis of MAP kinase pathway to assess role of genes in tumor development

Despite decades of research, cancer remains one of the biggest health challenges. Due to the intricate interplay between multiple factors and different cancer types, it is still impossible to pinpoint a common cause for all forms of cancer. Computational modeling can be helpful in integrating scattered information to derive comprehensive information about malignancy. We describe a discrete dynamic network model of a mitogen-activated protein kinase pathway consisting of 66 nodes and 95 edges. The network consists of five input signals (Fas ligand, DNA damage, insulin, tumor necrosis factor alpha and transforming growth factor beta) and three output nodes (proliferation, apoptosis and growth arrest). Using a random asynchronous update method andin siliconode perturbations, the accuracy of the model is ensured. The results of simulations and perturbations were in agreement with the gene knockout and constitutive expression studies reported in the literature, underscoring the high precision of the deduced comprehensive network. The fidelity of our model makes it useful to understand the etiology of malignancy. Both anti-cancer and pro-cancer roles have been attributed to DUSP1 in different forms of cancers and, in our model, DUSP1 knockout under insulin and DNA damage signaling was found to universally enhance the proportion of cells undergoing apoptosis (i.e. a pro-cancerous role), thus highlighting its potential in designing novel therapeutic interventions. Moreover, although MYC is a well-known oncogene, we found that MYC's overexpression can activate p53, a prominent anti-growth agent, through the p14 and MDM2 pathways.Implications:Our findings suggest a novel role of the DUSP1 and MYC genes in regulating cell proliferation.