Abstract:Our understanding of relatively simple protein interaction networks – comprised of only two or three components – can nevertheless still be hindered by the complexity of potential individual interactions or topological network structures. As examples, there are ~80 distinct molecular species that arise in the binding of one trivalent protein to another, and there are ~16,000 ways to wire a three-component signaling motif. In this talk, I will discuss our work to gain new mechanistic insights into: 1) the binding kinetics of multivalent protein-protein interactions; 2) the robustness of minimal protein network topologies in achieving switch-like responses; and 3) the energy landscape of a gene regulatory network. First, I will discuss our efforts to understand changes in the binding dynamics of bivalent and trivalent proteins as a function of the binding affinity of individual monomer units, the linker length/structure between the monomers, and the overall valency of each multivalent protein. The model provides new mechanistic insights for several noncanonical features that arise in multivalent binding kinetics and facilitates rational design of multivalent proteins with desired binding properties. In the second vignette, I will describe a computational approach for identifying two- and three-component network topology structures through which cells can robustly achieve all-or-none responses, a hallmark of binary cellular processes such as differentiation and proliferation. This analysis has provided insights into natural cell signaling and we have also used the emergent design principles to engineer synthetic switch-like responses in cells. Finally, I will present a computational method that we have developed to quantify the energy landscape for a network topology of interest. Applying this approach to a genetic toggle switch, we elucidate the roles of various intrinsic and extrinsic cues in shaping the landscape and in modulating the dynamics of cell transitions between energetic minima.