Overcoming the emergence of drug resistance in HIV is a major challenge to the scientific community. We use the established computational method of classical molecular dynamics to investigate the molecular basis of resistance in HIV-1 protease to the inhibitor saquinavir, using the wildtype and the G48V, L90M and G48V7L90M mutant HIV-1 proteases throughout this thesis. Firstly we reveal insights into a G48V mutation-assisted lateral drug escape mechanism from the protease active site. Such a mechanism allows drug escape without the full opening of the flaps of the protease. Furthermore, the mechanism is facilitated by differential drug-protease interactions, induced by mutations that take advantage of the conformational flexibility of the inhibitor. Secondly, we investigate the thermodynamic basis of binding of this set of mutants, using established 'approximate' free energy methods. The absolute and relative free energies of saquinavir binding to this set of proteases are successfully determined using our simulation and free energy analysis protocol and exhibit excellent correlation with experiment. This study is thus a template for an extended study on a larger range of HIV-1 protease-drug combinations. We describe a tool, the 'Binding Affinity Calculator', which has been designed to automate this protocol and which can be routinely applied, using high performance computing and grid technology, to meet the intensive computational demands of such an investigation. The free energy of binding of the NC-pl natural substrate cleaved by the protease is also deter mined. The enhanced flexibility of the substrate over the drug precludes the guarantee of a converged free energy result, even from the 10 ns duration of each simulation. However, qualitative insight into the thermodynamic basis of binding is gleaned as well as the effect of these mutations on the catalytic efficiency of the protease. Furthermore, we combine drug and substrate binding free energies to develop a metric for evaluating the approximate enzymatic fitness of a given mutant protease, computable directly from molecular simulation.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:719096 |
| Date | January 2008 |
| Creators | Sadiq, S. K. S'ad |
| Publisher | University College London (University of London) |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://discovery.ucl.ac.uk/1445831/ |
Page generated in 0.0194 seconds