Systematic profiling of substrate binding response to multidrug-resistant mutations in HIV-1 protease: Implication for combating drug resistance

Abstract

Human immunodeficiency virus 1 (HIV-1) protease (PR) represents one of the primary targets for developing antiviral agents for the treatment of HIV-infected patients. However, a number of multidrug-resistant mutations in the enzyme have been observed over the past decades, largely limiting the application of PR inhibitors in antiviral therapy. A systematic investigation of the intermolecular interaction between the multidrug-resistant mutants of HIV-1 PR and its substrates would help to establish a complete profile of substrate response to PR mutations and to design new antiviral agents combating drug resistance. Here, we describe an integrative method to profile 6 clinical multidrug-resistant PR mutants against a panel of 16 substrate octapeptides that flank 12 distinct PR cleavage sites in viral precursor polyproteins. It is found that most multidrug-resistant mutations have only a modest or moderate effect on substrate peptide binding, although these mutations would cause a large free energy loss in PR inhibitor binding. Structural analysis reveals that the substrate peptides are loosely bound within PR active pocket to form a wide contact interface between them, and thus mutation of just single or few residues seems not to influence PR–substrate binding considerably. In addition, peptides derived from variable cleavage sites are generally more sensitive to the mutations as compared to those derived from conserved sites, supporting the co-evaluation mechanism of HIV-1 PR and its substrates under drug suppression. We also identify 12 functionally conserved key residues around the enzyme’s active site, which play crucial role in substrate recognition. In vitro fluorescence anisotropy assays confirm that wild-type PR can bind substrate peptides ARVL/AEAM and NLAF/PQGE with a moderately high affinity (KD=2 and 16μM, respectively). In contrast, the key residue mutations N25D/D29N can completely eliminate (KD=n.d.) or largely reduce (KD=32 and 120μM, respectively) the binding capability of the two peptides, suggesting that these PR residues could be the potential target sites for developing resistance-free anti-HIV agents.