Allosterism is involved in almost all regulatory, signaling and control processes in cells. While there are many observations of the net effect of allosterism, a number of questions remain about the nature of this process at the molecular level. These include the role of induced fit vs. conformational selection in ligand binding and clarification of the actual mechanism of allosteric signal transmission, whether this involves a distinct pathway, a cooperative network of residues, correlated motional fluctuations or simply a perturbation of the energy landscape. Obtaining a deeper understanding of the dynamics of allostery has been especially challenging, since the signal transmission per se is not directly accessible to experiment. The idea behind our proposed project is that much can be learned about allosterism from the detailed study of a well-chosen prototype system by all-atom molecular dynamics (MD) simulations with analysis using Markov State Models (MSM), kinetic grouping analysis (KGA) and cFEP free energy profiles. The system chosen for study is the large, multi-domain protein MutS and its pathological mutants. MutS and eukaryotic homologs recognize errors in DNA replication and initiate mismatch repair. In Muts, DNA binding and ATP/ADP activities are allosterically coupled over a distance of ~ 70 Angstroms, but the mechanism of signal transmission over such a large distance has not been unequivocally determined. MutS and homologs are the focus of a biochemical research program in the laboratory of our collaborator here, Professor Manju Hingorani in the Department of Molecular Biology and Biochemistry. New in this proposal is the analysis of MD on MutS via a Markov State Model, which elucidates substate structure of a protein or DNA and enables the study of dynamical allosterism in terms of a proper Boltzmann ensemble. Specifically, we propose a) to carry out extended MD simulations on MutS and selected pathological mutants in aqueous solution at biological ionic strengths, considering all relevant combinations of DNA and ATP bound and unbound constructs; b) to cluster the MD trajectories on MutS and apply kinetic grouping analyses to obtain MSM to obtain substates of MutS, their structural properties, populations (thermodynamics), the rates of interconversions (kinetics) for the various MutS processes; and c) to analyze the MSM to elucidate the dynamics of allosteric signal transmission based on the calculated states and rates; apply this to search for new allosteric druggable sites.
A Molecular Dynamics Markov State Study of Allosterism in the DNA Repair Protein MutS David Beveridge, PhD 3-sentence summary: MutS and eukaryotic homologs recognize errors in DNA replication and initiate mismatch repair. In Muts, DNA binding and ATP/ADP activities are allosterically coupled over a distance of ~ 70 Angstroms, but the mechanism of signal transmission over such a large distance has not been unequivocally determined. Molecular dynamics (MD) computer simulations, Markov State Analysis and other computational modeling techniques will be used to investigate dynamical allosterism and mechanism of action of the mismatch repair protein MutS and homologs.