The M50I polymorphic substitution in association with the R263K mutation in HIV-1 subtype B integrase increases drug resistance but does not restore viral replicative fitness

Abstract

Background: First-generation integrase strand-transfer inhibitors (INSTIs), such as raltegravir (RAL) and elvitegravir (EVG), have been clinically proven to be effective antiretrovirals for the treatment of HIV-positive patients. However, their relatively low genetic barrier for resistance makes them susceptible to the emergence of drug resistance mutations. In contrast, dolutegravir (DTG) is a newer INSTI that appears to have a high genetic barrier to resistance in vivo. However, the emergence of the resistance mutation R263K followed by the polymorphic substitution M50I has been observed in cell culture. The M50I polymorphism is also observed in 10-25% of INSTI-naïve patients and has been reported in combination with R263K in a patient failing treatment with RAL.

Results: Using biochemical cell-free strand-transfer assays and resistance assays in TZM-bl cells, we demonstrate that the M50I polymorphism in combination with R263K increases resistance to DTG in tissue culture and in biochemical assays but does not restore the viral fitness cost associated with the R263K mutation.

Conclusions: Since the combination of the R263K mutation and the M50I polymorphism results in a virus with decreased viral fitness and limited cross-resistance, the R263K resistance pathway may represent an evolutionary dead-end. Although this hypothesis has not yet been proven, it may be more advantageous to treat HIV-positive individuals with DTG in first-line than in second or third-line therapy.

Figure 1

Occurrence of the M50I polymorphism in treatment-naïve individuals living with HIV-1 subtype B. Sequence analysis of subtype B integrase of 2,253 clinical isolates from treatment-naïve patients from the Stanford HIV Drug Resistance Database for the following polymorphisms: M50M, M50I, M50T, M50L and M50R.

Figure 2

Addition of M50I to R263K does not increase integrase strand-transfer activity. Strand-transfer activity and V_max/_1/2MaxProt of wild-type and mutant enzymes. A) Relative strand-transfer activity when varying protein concentration. B)V_max values. C)_1/2MaxProt values. D) Enzyme efficiency as determined by the division of V_max by _1/2MaxProt. Error bars represent the standard errors of the means (SEM).

Figure 3

The effect of M50I with R263K on enzyme activity. Strand-transfer activity and enzyme efficiency of wild-type and mutant enzymes. A) Relative strand-transfer activity when varying the concentration of biotinylated target DNA. B)V_max values. C)K_m values. D) Enzyme efficiency as determined by the division of V_max by K_m. Error bars represent the standard errors of the means (SEM).

Figure 4

M50I does not compensate for the reduction in HIV replication associated with R263K. Effects of the M50I and R263K mutations on HIV infectivity in TZM-bl cells (A) and replication capacity in PM1 cells (B).