Design and Discovery of HIV-1 IN Inhibitors with Novel Mechanism of Action
Tino Sanchez, University of Southern California, Los Angeles
Advisor: Nouri Neamati
Training in Basic Biomedical Sciences
Dissertation Award
2009
The long term goal of this study is to identify novel lead molecules as potential clinical candidates for the inhibition of HIV-1 integrase (IN) and the prevention of viral replication. Cellular cofactor LEDGF/p75 plays an essential role in IN-led incorporation of viral DNA into the human genome. Thus, designing compounds to disrupt IN-LEDGF/p75 complexes serves as a novel mechanistic approach different from current antiviral therapies. Recently, MK-0518, a diketo-enol derivative, was approved by the FDA for IN inhibition and the treatment of AIDS. Mechanistic studies show MK-0518 and the majority of IN inhibitors effectively chelate a magnesium ion within IN active site. I hypothesize potent IN inhibitors identified from IN-LEDGF/p75 models will have a mechanism of action different from MK-0518 and other studied IN inhibitors. I further hypothesize that IN inhibitors identified here that show positive physiochemical and synergistic effects can serve as a focusing tool for the rational design of new lead candidates. A solved co-crystal structure of IN and LEDGF/p75 in a protein-protein complex provides three-dimensional coordinates to create shape and structure-based models for the discovery of site-specific IN inhibition. Key amino acids that stabilize the complex act as a template to design eight independent computer-aided pharmacophore models to search compound databases, and establish training sets to create future activity-based models. Biological testing, synergy studies, and physiochemical ADMET (absorption, distribution, metabolism, excretion, and toxicity) simulations in silica will allow me to build predictive models to inhibit IN in vivo. In collaboration with Professor Debyser, we will test the identified lead molecules against virally infected cells and drug-resistant cell lines and validate the proposed models.
