Molecular Simulations of Pathways and Kinetics for Protein-protein Binding Processes

Protein-protein binding processes are crucial for biological functions and characterizing these processes fully has been a challenge in biophysics. In this work I use weighted ensemble path sampling method coupled with molecular simulations of varying levels of detail to answer long standing questions regarding protein-protein binding. In Chapter 3, I investigate the effects of preorganization on association between an intrinsically disordered peptide fragment of tumor suppressor p53 and the MDM2 protein using flexible residue level models. I simulated the binding process between p53 and MDM2 with varying degrees of preorganization in p53 and determined that the association rate constant of p53 peptide does not depend on the extent to which the peptide is preorganized for binding MDM2. In Chapter 4, I apply simulations with flexible molecular models to directly compute the “basal” kon for the association of the two proteins barnase and barstar, in the absence of electrostatics. I simulated the binding process between exact hydrophobic analogues barnase and barstar and determined the extent with which the electrostatics enhance the basal kon. Finally, in Chapter 5, I have generated binding pathways of barnase and barstar using all-atom simulations with explicit solvent. This study not only enabled a more detailed characterization of the binding mechanism but also provided an opportunity to determine the role of solvent in the binding process. Water molecules are proposed to play a crucial role in binding of barnase and barstar since water molecules can be found at the binding interface in the crystal structure and they increase the interfacial complementarity. Overall, the work presented here demonstrates the power of the weighted ensemble strategy in making it practical to characterize binding processes that are otherwise unfeasible for standard simulation