MUSIC Simulations on HIV-1 Integrase to Develop a Dynamic Pharmacophore Model for Anti-AIDS Drug Design

摘要

Three enzymes encoded by the HIV virus are the reverse transcriptase, protease and integrase. Some drugs have been approved by the FDA that target the transcriptase or protease. However, no drugs exist for the HIV integrase. The ability of HIV to rapidly evolve drug resistance, together with toxicity problems, requires the development of an additional class of antiviral drugs. Integrase is an attractive target for antivirals because it is essential for HIV replication and no counter part is known in the human genome. HIV integrase is a 32-kDa enzyme that carries out DNA integration via a two-step mechanism - 3′-processing to remove two nucleotides from each 3′-end of the viral DNA, and DNA strand transfer to integrate the ends of the viral DNA into the host genome. Integrase is composed of three structurally and functionally distinct domains, N-terminal domain, catalytic domain, and C-terminal domain. The catalytic domain contains the typical D,D(35)-E motif, which is necessary for catalysis. The latest crystal structure of the integrase core domain complexed with an inhibitor (1QS4) was resolved in 1999. Integrase is a difficult system for drug development because it exists as a dimer, tetramer or higher order mulitimer and domain-domain interactions and enzyme-DNA complexation have not been well described. In this work, we provide an approach to develop receptor based phamacaphore models via MUSIC simulations that use structures obtained from our previous molecular dynamics studies performed on the HIV-1 integrase for lns. MUSIC is a multi-unit searching algorithm for interacting conformers carried out by Monte Carlo simulations, available in the BOSS program. Different types of small molecules representing important functional groups are chosen as probes to dock against the protein. The snapshots of protein structures are obtained from molecular dynamics simulations, taken at 100ps intervals. The binding sites for functional groups that complement the active site are determined through the MUSIC calculations. All protein conformations, together with docked probes, are overlaid and the conserved binding regions for the probes are identified. Those conserved binding sites define a "dynamic pharmacophore model". A similar calculation carried out against a single protein structure results in a "static phamacophore model". The pharmacophore models have been tested by verifying their ability to identify known inhibitors in small molecule databases. Finally, the dvnamic and static pharmacophore models will be used to carry out 3D database searches of the Available Chemicals Directory (ACD) to identify potential new inhibitors that will then be acquired and submitted to activity and toxicity assays.