This project will consist of a series of molecular dynamics computer simulations aimed at understanding the thermodynamics and kinetics of the binding of gaseous ligands to heme proteins. Such events are of physiological importance for the oxygen storage and transport proteins myoglobin and hemoglobin, and are models of the ways in which internal cavities in proteins accomodate themselves to """"""""probe"""""""" molecules of varying size. Recent advances in computer simulation techniques, and in overall computer speed, make it feasible to study both equilibrium and kinetic rate constants using a microscopic model at an atomic level of resolution. Studies of the binding of these simple ligands also provides an important testing ground for these new approaches, and should point the way to future studies of more complex enzyme-substrate interactions. The proposed work is divided into three projects. In the first, free energy perturbation methods (based on molecular dynamics simulations) will be used to determine the equlibrium binding affinities for a series of inert gases (He, Ar, Xe, N2, CF4 and SF6) to cavities inside myoglobin that are known experimentally to bind Xe. Similar calculations will be carried out for the binding of dioxygen, carbon monoxide, and alkyl isocyanides (RNC, R=methyl, ethyl, n-propyl, n-butyl and tert-butyl) to the distal heme pocket. The second project will use two novel computer graphics techniques to search systematically for plausible binding pathways for entrance or exit from these cavities. The third project will apply modern theories of reaction rates in condensed phases to estimate rate constants for the """"""""gating"""""""" proceses that control access to these cavities from solution. This will allow comparisons to rate constants determined from laser photodissociation experiments, and will clarify the connection between elements of protein structure and the intermediates seen in kinetics experiments.
