Structure-based drug design of non-nucleoside inhibitors for wild-type anddrug-resistant: HIV reverse transcriptase

The generation of anti-HIV agents using structure-based drug design methods has yielded a number of promising non-nucleoside inhibitors (NNIs) of HIV reverse transcriptase (RT). Recent successes in identifying potent NNIs are reviewed with an emphasis on the recent trend of utilizing a computer mode l of HIV RT to identify space in the NNI binding pocket that can be exploit ed by carefully chosen functional groups predicted to interact favorably wi th binding pocket residues. The NNI binding pocket model was used to design potent NNIs against both wild-type RT and drug resistant RT mutants. Molec ular modeling and score functions were used to analyze how drug-resistant m utations would change the RT binding pocket shape, volume, and chemical mak e up, and how these changes could affect inhibitor binding. Modeling studie s revealed that for an NNI of HIV RT to be active against RT mutants such a s the especially problematic Y181C RT mutant, the following features are re quired: (a) the inhibitor should be highly potent against wild-type RT and therefore capable of tolerating a considerable activity loss against RT mut ants (i.e. a picomolar-level inhibitor against wild-type RT may still be ef fective against RT mutants at nanomolar concentrations), (b) the inhibitor should maximize the occupancy in the Wing 2 region of the NNI binding site of RT, and (c) the inhibitor should contain functional groups that provide favorable chemical interactions with Wing 2 residues of wild-type as well a s mutant RT. Our rationally designed NNI compounds HI-236, HI-240, HI-244, H-253, HI-443, and HI-445 combine these three features and outperform other anti-HIV agents examined. (C) 2000 Elsevier Science Inc.