20 April 2007
Ravi Iyengar
Department of Pharmacology and Systems Therapeutics
Mount Sinai School of Medicine
Mammalian cells have a large intracellular regulatory network that communicates information from receptors to cellular machines to evoke cellular responses and phenotypic behavior. This network is made of interacting signaling components such as heterotrimeric and small G proteins, second messenger producing enzymes such as adenylyl cyclases, protein kinases and phosphatases, as well as anchors and scaffolds. The regulatory network interfaces with components in cellular machines to coordinately regulate their function. The coordinated operation of the cellular machines in the different cell types gives rise to their distinctive phenotypic behavior. As information flows through the regulatory network, it is processed and this processing of information forms the basis for cellular decisions regarding state change and maintaining homeostatic behavior. Information processing occurs due to presence of regulatory motifs such as feedback, feedforward and bifan motifs within the network. Each of these motifs has distinct information processing capability. Recent computational and experimental studies in our laboratory show that a feedforward motif can provide prolonged signal output and that nested feedforward motifs from the β-adrenergic-receptor to the transcription factor CREB can drive differentiation of kidney podocytes. Computational analyses of bifan motifs made of protein kinases regulating transcription factors indicate that bifans can serve as noise filters. Positive feedback loops function as bistable switches. Stacking of a positive feedback, bifan and negative feedback loop yields tight regulation of transcription. The organization of motifs within large intracellular regulatory networks has been studied using graph-theory approaches. We find that regulatory networks have few long paths and long feedback loops and are enriched for small loops indicating that embedding of small loops within long paths may be favored over large loops. Information about the higher-level organization of signaling networks, along with the properties and functional location of the motifs allows us to develop hypothesis of how cellular regulatory networks are organized for flexible decision making within the cell.