Structural analysis of epidermal growth factor receptor and yeast signaling networks

Abstract

The recent progress in high-throughput technologies is allowing the reconstruction of large-scale models of signaling networks and development of corresponding analysis methods. Network properties of the most comprehensive published maps of the epidermal growth factor receptor (EGFR) signaling with related pathways and signal transduction mechanisms in yeast (Saccharomyces cerevisiae) were analyzed structurally using graph theoretic and pathway analysis techniques. The EGFR signaling pathway controls growth, survival, proliferation and differentiation of mammalian cells. It is one of the most investigated and best-understood signal transduction systems and now it is being used to understand the general mechanisms of signal transduction processes. The graph theoretic analyses of the EGFR and yeast signaling networks indicated that they have small-world topologies. The connectivity distributions of the signaling molecules in the networks have scale-free natures following power law model. The pathway analysis of EGFR signaling network indicated that as a result of signal transduction apoptosis may be the most frequent phenotype of the EGFR signaling having the smallest path lengths for nearly same number of linear paths for all phenotypes. The tumorigenesis and mitogenesis are found to be the least robust phenotypes having limited routes from the ligands that pass over the same six signaling molecules each time. Analysis via linear paths revealed that G1/S cell wall remodeling and mating are the most and least frequent phenotypes of yeast signaling network, respectively. The critical signaling molecules in the networks that participate in most of the linear paths from the ligands to the phenotypes were identified using pathway analysis. The cofactor GTP and the complex Gb/Gg are involved in more than 70 % of the linear paths through all phenotypes in the network indicating that the pathways going to all phenotypes of EGFR signaling network are crosstalking through these molecules. Similarly the proteins Rvs167 and Sla1 have roles in the linear paths going to all phenotypes in signal transduction network of yeast.