Abstract Body

Background: Protease inhibitor (PI)- based regimens are the mainstay of antiretroviral therapy for HIV-2. However, HIV-2 exhibits some degree of intrinsic resistance to the majority of FDA-approved PI, only retaining clinically-useful susceptibility to lopinavir, darunavir, and saquinavir. The mechanisms for this resistance remain largely unknown; although HIV-1 and HIV-2 proteases share only 40-50% sequence identity, structural studies indicate that the ligand-binding sites differ by just four amino acids. In the current study, we examined the contributions of these four residues to intrinsic PI resistance in HIV-2.

Methods: We used site-directed mutagenesis to construct an HIV-2ROD9 molecular clone in which protease codons 32, 47, 76, and 82 were substituted to encode the amino acids found in wild-type HIV-1 (clone PRΔ4: I32V+V47I+M76L+I82V), as well clones containing each single substitution. We used single-cycle assays to quantify the phenotypic sensitivity of the five mutant clones, as well as wild-type (WT) HIV-1NL4-3and HIV-2ROD9, to nine FDA-approved PI. EC50 values were calculated from dose-response data using sigmoidal regression and tested for statistically significant differences by ANOVA.

Results: Relative to WT HIV-2, HIV-2 clones containing single amino acid substitutions I32V, V47I, M76L, or I82V conferred effects ranging from no change to a 13-fold decrease in EC50, depending on the PI tested. Clone PRΔ4 displayed significant reductions in EC50 (3.6 to 60-fold) to all PI except saquinavir. EC50 values for PRΔ4 vs. WT were as follows: saquinavir 12 vs. 31 nM, ritonavir 160 vs. 580 nM, lopinavir 21 vs. 105 nM, tipranavir 120 vs. >1000 nM, indinavir 16 vs. 150 nM, nelfinavir 47 vs. 490 nM, darunavir 2.2 vs. 58 nM, amprenavir 23 vs. >1000 nM, and atazanavir 1.1 vs. 66 nM. EC50 values for PRΔ4 were lower than, or equivalent to, those for WT HIV-1.

Conclusions: Taken together, our data show that four amino acid changes in HIV-2 protease are sufficient to confer a pattern of PI susceptibility comparable to that of HIV-1. These findings enhance our overall understanding of the genetic basis of PI susceptibility and show that active site residues in protease are the primary determinants of intrinsic PI resistance in HIV-2.