Protein-protein interactions (PPIs) underlie numerous molecular processes in living organisms, such as signal transduction, cell cycle regulation, and cell metabolism, and have increasingly been targeted in drug development. Most PPIs are dynamic processes involving the interplay of molecular interactions and protein conformational changes to bind. This project will study the time-dependent process of molecular recognition and binding that give rise to specific contacts. In a collaboration between Andrei Tokmakoff and Stephen Kent at the University of Chicago and Vijay Pande at Stanford University, we propose to merge our unique technologies into an integrated method that will reveal the molecular dynamics of PPIs. Particular emphasis is placed on describing protein conformational ensembles and disorder in the mechanism of recognition and binding. Our investigation will be directed at the association of insulin monomers into a dimer. Insulin monomer, the active form of the hormone, extends the C-terminal residues of its B-chain in the process of binding insulin receptor. This chain is thought to be disordered in the monomer; however, it folds into a compact intermolecular ? sheet in the dimer or hexamer form it adopts in pancreatic ?-cell secretory granules. Our studies will characterize the structure and disorder of insulin monomer and the hinging and dynamic interactions of its B chain, factors that critically influence glucose regulation. Furthermore, we will directly visualize the fleeting encounter species that arise durin recognition and binding, to help test whether insulin dimer association follows a fly-casting mechanism. The approach will use two-dimensional infrared spectroscopy (2D IR) of protein amide I vibrations in concert with advanced synthesis and simulation methods. 2D IR can be used to characterize protein structures with ultrafast time resolution. Using a laser temperature-jump (T-jump), we generate the transient encounter complexes that exist at the transition state for recognition and binding, and use 2D IR to time-resolve their conformational changes over nanosecond to millisecond time scales. Site-specific information on protein conformation is achieved with the help of isotope labels inserted during total chemical synthesis of insulin by Kent. Atomistic molecular modeling using computational IR spectroscopy in conjunction with Pande's large- scale molecular dynamics (MD) simulations and Markov state models (MSMs) allows us to test and refine structural and dynamical ensembles observed in experiment.
Protein-protein interactions underlie the molecular processes of life and are targets for drug discovery. We will develop experiments to visualize the molecular dynamics of insulin dimer association, a process related to how it binds its receptor, and an important factor in glucose regulation for diabetics.