BROWNIAN DYNAMICS SIMULATION FOR PROTEIN FOLDING AND BINDING

A protein with a certain amino acid sequence folds into its unique three-dimensional shape spontaneously and binds to its ligand or partner proteins to play its own role in a living system. This relationship between a protein sequence and its three-dimensional structure constitutes the second part of the genetic code that links DNA sequence information of a gene with the function of its product, which remains an enigma in biology. Computer simulation is a powerful tool that can analyze and calculate energies of various conformations of proteins at atomic resolution. However, when the protein folding and binding are pursued using this method, we face to two barriers: efficient sampling of protein conformations and improvement of force field. To break these barriers, I have developed an atomistic Brownian dynamics (BD) simulation method that uses physics-based energy terms and force field. The BD method does not treat water molecules explicitly and can adopt a long time step, resulting reduction of the computation time greatly. Some peptides folded into their native structures from extended conformations without statistical information obtained from databases of proteins using the BD method. I have also developed an umbrella sampling method combined with the BD to calculate absolute binding affinity of protein-protein interaction. By using the BD/umbrella sampling method, binding free energy of WW-domain/Pro-rich peptide could be estimated quantitatively. In these respects, the BD method would be effective for analysis of protein folding and binding.