In vitro activity of dolutegravir against wild-type and integrase inhibitor-resistant HIV-2

Dolutegravir recently became the third integrase strand transfer inhibitor (INSTI) approved for use in HIV-1–infected individuals. In contrast to the extensive dataset for HIV-1, in vitro studies and clinical reports of dolutegravir for HIV-2 are limited. To evaluate the potential role of dolutegravir in HIV-2 treatment, we compared the susceptibilities of wild-type and INSTI-resistant HIV-1 and HIV-2 strains to the drug using single-cycle assays, spreading infections of immortalized T cells, and site-directed mutagenesis. HIV-2 group A, HIV-2 group B, and HIV-1 isolates from INSTI-naïve individuals were comparably sensitive to dolutegravir in the single-cycle assay (mean EC50 values = 1.9, 2.6, and 1.3 nM, respectively). Integrase substitutions E92Q, Y143C, E92Q + Y143C, and Q148R conferred relatively low levels of resistance to dolutegravir in HIV-2ROD9 (2- to 6-fold), but Q148K, E92Q + N155H, T97A + N155H and G140S + Q148R resulted in moderate resistance (10- to 46-fold), and the combination of T97A + Y143C in HIV-2ROD9 conferred high-level resistance (>5000-fold). In contrast, HIV-1NL4-3 mutants E92Q + N155H, G140S + Q148R, and T97A + Y143C showed 2-fold, 4-fold, and no increase in EC50, respectively, relative to the parental strain. The resistance phenotypes for E92Q + N155H, and G140S + Q148R HIV-2ROD9 were also confirmed in spreading infections of CEM-ss cells. Our data support the use of dolutegravir in INSTI-naïve HIV-2 patients but suggest that, relative to HIV-1, a broader array of replacements in HIV-2 integrase may enable cross-resistance between dolutegravir and other INSTI. Clinical studies are needed to evaluate the efficacy of dolutegravir in HIV-2–infected individuals, including patients previously treated with raltegravir or elvitegravir.

Conclusions: Our data support the use of dolutegravir in INSTI-naïve HIV-2 patients but suggest that, relative to HIV-1, a broader array of replacements in HIV-2 integrase may enable cross-resistance between dolutegravir and other INSTI. Clinical studies are needed to evaluate the efficacy of dolutegravir in HIV-2-infected individuals, including patients previously treated with raltegravir or elvitegravir.
In the present study, we examined the activity of dolutegravir against wild-type and INSTI-resistant HIV-2 strains using an indicator cell assay that restricts viral replication to a single cycle [15]. This methodology enables a direct comparison of HIV-1 and HIV-2 drug susceptibility while avoiding potential confounders such as differences in replication rates, infectivity, cytopathic potential and cell-to-cell spread.
Altogether, nine of the 13 HIV-2 integrase mutants tested were resistant to dolutegravir in the single-cycle assay (Figure 2A).
Taken together, our results indicate that prototypic HIV-1 and HIV-2 strains, as well as HIV-1 and HIV-2 isolates from INSTI-naïve individuals, are comparably sensitive to dolutegravir in a single cycle of viral replication in MAGIC-5A indicator cells (Figure 1). These findings complement previous data from spreading infections of PBMC [51]-using a different methodology and target cell type-and suggest that dolutegravir would be an appropriate treatment choice for INSTI-naïve HIV-2 patients when combined with other HIV-2-active ARV. We also report the effects of raltegravir-associated mutations on dolutegravir susceptibility using site-directed mutagenesis of genetically-defined HIV-1 and HIV-2 molecular clones (pNL4-3 and pROD9, respectively).
Our analysis shows that equivalent amino acid changes in the integrase proteins of HIV-1 and HIV-2 can have differing effects on dolutegravir susceptibility ( Figure 2D) and that, in HIV-2 ROD9 , integrase changes Q148K, T97A + Y143C, E92Q + N155H, T97A + N155H, and G140S + Q148R confer moderate to high levels of dolutegravir resistance (≥10-fold; Figure 2A-C and Table 1). We cannot exclude the possibility that the resistance levels observed in our site-directed HIV-2 mutants are specific to the ROD9 molecular clone, as the genetic context within integrase can have a substantial impact on the phenotypic expression of INSTI resistance [71,72]. For example, in the aforementioned study by Charpentier et al. Curve fits were generated as described in Figure 1A. (D) Comparative analysis of equivalent site-directed mutants of HIV-1 and HIV-2 integrase. Each point is the result of a single dose-response assay performed as described in Figure 1A. Horizontal bars indicate the mean EC 50 values for each strain. Red arrows and text indicate fold increases in the mean EC 50 values for HIV-2 relative to HIV-1. ANOVA results for these data are described in the main text. In all panels, error bars represent standard deviations.

Competing interests
The authors declare that they have no competing interests.
Authors' contributions RS and GG conceived the study, designed the experiments and prepared the final version of the manuscript. RS performed the experiments, analyzed the data, and drafted the manuscript. DR and CP helped conduct the virologic assays. GG, PS, MS and JM provided intellectual input throughout the study and helped interpret the data. All authors read and approved the final manuscript.