The recent approval of the 36-amino-acid peptide enfuvirtide (T-20; Trimeris/Roche) heralds a new class of HIV-1 therapeutics: viral entry inhibitors. Before the launch of enfuvirtide in March 2003, all marketed HIV drugs targeted one of two viral enzymes — protease and reverse transcriptase. However, despite the significant success achieved with standard combination therapy, the lack of patient compliance due to adverse side effects and the emergence of resistant viral strains has resulted in many therapeutic failures and a significant unmet clinical need. Now, in the Proceedings of the National Academy of Sciences, a group of researchers at Bristol-Myers Squibb reveal details of a small molecule, BMS-378806, which targets the HIV-1 envelope protein and inhibits CD4 receptor binding, thereby preventing HIV-1 entry into the host T cell.

The viral entry process provides new anti-HIV-1 targets, and the potential for developing novel classes of anti-HIV drugs. The HIV-1 envelope consists of the exterior glycoprotein gp120 and the transmembrane domain gp41, both of which are processed from a gp160 precursor. After attachment of HIV-1 to a target T cell carrying the CD4 receptor, the interaction of gp120 with the CD4 receptor initiates a series of conformational changes in both gp120 and gp41 that allow the insertion of a region of gp41 into the host cell. Further changes in the conformation of gp41 bring the viral and cellular membranes close enough for membrane fusion, although the mechanism of fusion is still unresolved. Enfuvirtide binds to a region of gp41 that mediates part of the conformational change, whereas BMS-378806 binds to gp120 to inhibit CD4 binding.

A whole-cell infection screen against a strain of HIV-1 identified an indole analogue as an initial hit, which after optimization to improve potency, specificity and pharmacokinetic parameters resulted in the lead candidate BMS-378806, a 4-methoxy-7-azaindole derivative. The compound showed potent inhibitory activity against a panel of R5, X4, R5/X4 (viruses using the CCR5, CXCR4 or both chemokine receptors, respectively) HIV-1 laboratory and clinical isolates. BMS-378806 was also selective for HIV-1, and inactive against other viruses, including HIV-2 and simian immunodeficiency virus. Importantly, BMS-378806 is effective against HIV strains resistant to existing HIV-1 protease and reverse transcriptase inhibitors.

The authors carried out studies on the mechanism of action, and confirmed that BMS-378806 binds to gp120 and prevents further interaction of the viral protein with cellular CD4 receptors. Further confirmation was obtained by analysing the amino-acid substitution of resistant virus strains, isolated by passaging virus in the presence of the drug and selecting for variants resistant to BMS-378806. These experiments revealed that almost all the substitutions spanned the gp120 region, with a couple of changes residing in the gp41 region, and none outside the envelope. Significantly, the two key substitutions are situated in the CD4-binding pocket of gp120.

BMS-378806 showed good oral bioavailability and a clean safety profile in initial animal studies. Whereas other HIV entry inhibitors in early clinical trials block co-receptors and membrane fusion, BMS-378806 is the first small molecule to block the gp120–CD4 binding event.