Molecular modeling of biomolecular interactions: Natural products, peptides and proteins

The works in this thesis are mainly dealing with computational molecular modeling applied to biomolecular systems. First of all I present the works of predicting inhibiting mechanisms of HIV-1 entry by a natural product OLE (Olive Leaf Extract). We systematically studied the main effective polyphenolic compound Oleuropein and three of its metabolites. Our modeling works provide the detailed mechanism of OLE mediated HIV-1 entry inhibition at the atomic level, and latter experiments provided adequate proofs for the accuracy of our predictions. This computational study complements the corresponding experimental investigation and helps establish a good starting point for further refinement of OLE-based gp41 inhibitors.;In the second case, molecular dynamics (MD) simulation and MM-PBSA were used to understand how the covalent modification of the natural Bak sequence affects the binding to Bcl-xL at molecular levels. The present MD result shows that the helicities of HBS peptides are increased and the presence of the N-terminal HBS macrocycle impacts residues at the C-terminus of the helix. Our analysis also indicates that substitution of an aspartic acid residue---a helix breaker---with a hydrophobic residue not only enhances the helicity of the peptide but also stabilizes the structure of the binding complex. The present computational result is consistent with the experimental observation and provides explanations for the altered binding properties of the artificial Bak alpha-helix.;However, there are certain limits for current molecular mechanics and force field. In the third case, we studied the temperature-dependent conformation distributions of a short alanine-rich peptide XAO by an improved AMBER force field ff03 with replica exchange molecular dynamics (REMD). The results are largely deviated from the experiments, both in explicit and implicit solvent models, where experiments determined that the poly-L-proline (PPII) conformation is dominant at the low temperatures while simulations give the prediction that alpha-helical conformation is dominant at the same temperatures. Further ab initio calculations suggest that this deviation arises from lacking of polarization effects of classical force fields. These results reflect the importance of correct inclusion of polarization effects for protein folding simulations.