Computational studies of protein-peptide interactions.

The interactions between proteins and peptides in aqueous solution have been investigated using a classical molecular dynamics procedure with a molecular mechanical representation of the potential energy surface. During post-processing of the trajectory an implicit solvation method has been applied in order to calculate the free energy of each of the complex, protein and peptide structures in solution, allowing the binding free energy of the protein-peptide complex to be evaluated. Entropic contributions have been estimated using classical ideal gas thermodynamics. A program has also been developed that systematically mutates each of the peptide residues to alanine and determines the effect on the binding free energy. The method has been applied to the interaction between the oncoprotein Mdm2 and the tumour suppressor peptide p53 and reasonable agreement has been found with previous theoretical and experimental studies. The method has also been extended to the interactions between IQN17, an engineered protein that represents a potential drug target in the HIV-1 gp41-mediated cellular fusion process, and several peptides that have been shown to inhibit cellular fusion. The key residues in the binding of each protein-peptide system have been identified and quantitative information regarding the factors influencing binding obtained and compared with available experimental data, demonstrating encour aging agreement with analogous alanine scanning experiments