Background:  In HIV-1-infected patients, the Q151M mutation confers resistance to almost all the nucleoside reverse transcriptase inhibitors (NRTI). In HIV-2 infection, many studies have reported a high frequency of selection of this mutation but its influence on phenotypic susceptibility of HIV-2 isolates remains unclear.

Methods:  We selected 4 HIV-2-infected patients from the French ANRS HIV-2 cohort, with evidence of Q151 mutation in both plasma and available peripheral blood mononuclear cells (PBMC) co-cultivated supernatants. Phenotypic susceptibilities to different NRTI—zidovudine (AZT), didanosine (ddI), stavudine (d4T), abacavir (ABC), lamivudine (3TC), and tenofovir (TDF)—were evaluated using a modified version of the ANRS PBMC method. The 50% tissue culture infective dose (TCID₅₀) was assessed by measuring the number of RNA HIV-2 copies in the supernatant using a real-time quantitative polymerase chain reaction (RT-PCR) assay and the drug concentration inhibiting 50% and 90% (IC₅₀ and IC₉₀, respectively) of the replication of 100 TCID₅₀ was calculated. After isolation IC₅₀ values were compared either before (T0) and after the selection of the Q151M (T1) or to wild type reference isolates when NRTI mutations were present at T0. 

Results:  At T0, mean IC₅₀ of the 2 HIV-2 wild type isolates were:  0.01 µM for AZT, 0.07 µM for 3TC, 0.04 µM for d4T, 0.70 µM for ddI, 0.10 µM for ABC, and 0.30 µM for TDF. The 2 other isolates had M184V mutation at T0 and showed a >60-fold increase in IC₅₀ for 3TC, 4-fold for ddI and >7-fold for ABC as compared with ROD wild type reference strain. At T1, RT bulk sequencing showed the selection of Q151M alone (n = 1) or associated with other mutations (n = 3). When Q151M mutation was alone, there was no decrease in susceptibility to AZT, 3TC, ddI, and TDF but an increase in IC₅₀ for d4T and ABC (22- and 37-fold, respectively) was observed. When Q151M was selected with M184V (n = 1), an increase in IC₅₀ (>10-fold) was observed for all NRTI except TDF. When Q151M was selected with V111I (n = 2) an increase in IC₅₀ (4 to >10) was observed for all NRTI, except for ddI when K65R/N69S/V111I/Q151M (n = 1) were co-selected.

Conclusions:  In HIV-2 isolates, Q151M mutation alone influences only the phenotypic susceptibility to d4T and ABC. A decrease in susceptibility to all NRTI was observed when Q151M was selected with V111I, mutation of unknown effect on HIV-1 resistance. Clinical relevance of these phenotypic susceptibility results needs to be evaluated in HIV-2⁺ treated patients.