The primary goal of the proposed research is a detailed understanding of the relationships between structure and energy that determine the structure and functions of proteins. A major focus will be the development and application methods for the study of problems such as protein folding and protein stability and the prediction of binding free energies of substrates to proteins. In most applications, relative free energies in solution will be obtained by combining gas phase molecular mechanics calculation with the evaluation of solvation free energies. Electrostatic contributions to solvation are obtained from numerical solutions to the Poisson-Boltzmann (PB) equation. Non-polar contributions are calculated from free energy/surface area relationships. The methodological emphasis will involve the development of algorithms which yield rapid and accurate numerical solutions to the PB equation. The goal is to reach the point where a solvation free energy calculation on a molecule is as fast as a gas phase energy evaluation in molecular mechanics program. This will make it possible to include solvent directly in conformational search procedures, energy minimization and molecular dynamics. Preliminary estimates suggest that a combination of multigriding and adaptive gridding techniques make this a reasonable goal. Methods involving surface charges will also be developed to calculate forces on atoms directly from PB calculations. Proposed research on protein folding will focus on three areas. l) A recently developed method to calculate the pH dependence of protein stability will be used to study acid denaturation. The stability of the compact """"""""molten globule"""""""" states formed under acidic conditions will be considered in this context. 2) The factors that determine secondary structure stability, including individual amino acid propensities, will be elucidated with calculations of relative conformational energies. 3) A method to distinguish stable from unstable protein conformations will be developed. Calculations will be carried out of the relative binding energies of different inhibitors to the HIV protease. Studies will be made of compounds for which both structural and thermodynamic data are available. Similar studies will also be carried out on the binding of peptides to class I MHC proteins. High resolution structures are currently available for the binding of different peptides to the same class I protein and binding affinities have also been measured. An attempt will be made to understand the energetic principles and structural rules involved in antigen recognition. The health relatedness of the proposed research is in the general principles about protein structure and function that are being developed. In addition the results of the research will be applicable to structure- based drug design in general, with specific applications to the design of HIV protease inhibitors and to the design of antiviral drugs that act on MHC proteins.
Showing the most recent 10 out of 72 publications
