Computational Advances in Structure Based Drug Design with Applications to HIV-1 Reverse Transcriptase

Computational Advances in Structure Based Drug Design with Applications to HIV-1 Reverse Transcriptase Robert Christopher Rizzo 2001 Computational advances in structure based drug design are presented which emphasize the development of protocols and methodology used in force-field parameterization, scoring function development, structure prediction and validation, and docking. Force-field parameters have been developed for amines primarily by fitting to experimental data for pure liquids and to hydrogen−bond strengths from gas-phase ab initio calculations. The parameters were used to compute relative free energies of hydration using free energy perturbation calculations in Monte Carlo simulations (MC/FEP). The results are in excellent agreement with experimental data, in contrast to numerous prior computational reports. MC simulations for the pure liquids of thirteen additional amines demonstrated the transferability of the force field. The interactions and energetics associated with the binding of 20 HEPT and 20 nevirapine non-nucleoside inhibitors of HIV-1 reverse transcriptase (RT) have been explored in an effort to establish simulation protocols and methods that can be used in the development of more effective anti-HIV drugs. Each inhibitor was modeled in the bound and unbound states via MC statistical mechanics methods. A viable regression equation was obtained using only four descriptors to correlate the 40 experimental activities with a r of 0.75 and cross-validated q of 0.69. The MC results revealed three physically reasonable parameters that control the binding affinities. Molecular docking and simulation methods have been used to generate a model of the FDA-approved inhibitor Sustiva bound to HIVRT. The docking protocol was validated with known NNRTI complexes. MC/FEP simulations confirmed that the predicted structures yield correct results for the effects of the Y181C and V106A mutations on the activity of Sustiva, nevirapine, MKC-442, and 9-Cl TIBO. A subsequently reported crystallographic complex of Sustiva with HIVRT fully confirmed the prediction. Docking studies that include cluster analysis are presented in an effort to reduce the number of candidate conformers that need to be docked for very flexible ligands. Despite a limited conformational search, clustering based on a rmsd value of 2.5 A dramatically reduced the total number of clusters yet still retained at least one cluster representative with a conformation similar to the experimental bound-like conformation for the majority of systems tested. Computational Advances in Structure Based Drug Design with Applications to HIV-1 Reverse Transcriptase A Dissertation Presented to the Faculty of the Graduate School of Yale University in Candidacy for the Degree of Doctor of Philosophy by Robert Christopher Rizzo Dissertation Director: William L. Jorgensen