Clinical, virological and biochemical evidence supporting the association of HIV-1 reverse transcriptase polymorphism R284K and thymidine analogue resistance mutations M41L, L210W and T215Y in patients failing tenofovir/emtricitabine therapy

Abstract

Background: Thymidine analogue resistance mutations (TAMs) selected under treatment with nucleoside analogues generate two distinct genotypic profiles in the HIV-1 reverse transcriptase (RT): (i) TAM1: M41L, L210W and T215Y, and (ii) TAM2: D67N, K70R and K219E/Q, and sometimes T215F. Secondary mutations, including thumb subdomain polymorphisms (e.g. R284K) have been identified in association with TAMs. We have identified mutational clusters associated with virological failure during salvage therapy with tenofovir/emtricitabine-based regimens. In this context, we have studied the role of R284K as a secondary mutation associated with mutations of the TAM1 complex.

Results: The cross-sectional study carried out with > 200 HIV-1 genotypes showed that virological failure to tenofovir/emtricitabine was strongly associated with the presence of M184V (P < 10-10) and TAMs (P < 10-3), while K65R was relatively uncommon in previously-treated patients failing antiretroviral therapy. Clusters of mutations were identified, and among them, the TAM1 complex showed the highest correlation coefficients. Covariation of TAM1 mutations and V118I, V179I, M184V and R284K was observed. Virological studies showed that the combination of R284K with TAM1 mutations confers a fitness advantage in the presence of zidovudine or tenofovir. Studies with recombinant HIV-1 RTs showed that when associated with TAM1 mutations, R284K had a minimal impact on zidovudine or tenofovir inhibition, and in their ability to excise the inhibitors from blocked DNA primers. However, the mutant RT M41L/L210W/T215Y/R284K showed an increased catalytic rate for nucleotide incorporation and a higher RNase H activity in comparison with WT and mutant M41L/L210W/T215Y RTs. These effects were consistent with its enhanced chain-terminated primer rescue on DNA/DNA template-primers, but not on RNA/DNA complexes, and can explain the higher fitness of HIV-1 having TAM1/R284K mutations.

Conclusions: Our study shows the association of R284K and TAM1 mutations in individuals failing therapy with tenofovir/emtricitabine, and unveils a novel mechanism by which secondary mutations are selected in the context of drug-resistance mutations.

Figure 1

Principal axis factoring analysis of correlations between mutations associated with tenofovir/emtricitabine therapy failure. A factor scores plot (in rotated factor space) is shown. Amino acid changes with high coefficients of covariation are close together, while large distances separate those substitutions that show low or negative coefficients of association. Major mutations of the TAM1 and TAM2 complexes are indicated in red and blue, respectively.

Figure 2

Replication kinetics of WT and mutant RTs in the absence and presence of AZT and tenofovir disoproxil fumarate (Tenofovir-DF). In each case, histograms show the relative replication capacity (%), compared to the WT virus in the absence of drug, based on the slopes of p24 antigen production of each recombinant virus after infection of stimulated PBMCs. The significance of the difference between slopes was calculated using the GraphPrism v. 4 software and significant p values are represented above the bars. Statistical analyses were performed by using a Student t test.

Figure 3

Rescue DNA polymerization initiated from AZTMP-, d4TMP-, and tenofovir-terminated primers annealed to a DNA template. Reactions were carried out with D38/25PGA or D38T/25PGA complexes (sequences given above). The 25-nucleotide primer (lane P) is first blocked with the nucleotide analogue (lane B). The excision of the inhibitor, followed by extension of the primer is achieved after addition of a mixture containing 3.2 mM ATP and the four dNTPs. A fully extended 38-nucleotide product is formed. The gel on the right shows a representative time course experiment of a primer rescue reaction. Lanes 1 to 9 correspond to aliquots removed 2, 4, 6, 8, 10, 12, 15, 20, and 30 minutes after the addition of 3.2 mM ATP. Graphs of time course experiments of primer rescue reactions initiated from inhibitor-terminated primers are given below. All dNTPs in the assays were supplied at 100 μM, except for dATP or dTTP (depending on the reaction) whose concentration was 1 μM. Template-primer and active RT concentrations in these assays were 30 and 24 nM, respectively. The values (averaged ± standard deviations [error bars]) were obtained from three independent experiments.

Figure 4

Rescue DNA polymerization initiated from AZTMP-, d4TMP-, and tenofovir-terminated primers annealed to an RNA template. Time course experiments of excision reactions were carried out in the presence of 3.2 mM ATP. The nucleotide sequences of template-primers used are given above their corresponding graphs. All dNTPs in the assays were supplied at 200 μM, except for dATP or dTTP (depending on the reaction) whose concentration was 2 μM. Template-primer and active RT concentrations in these assays were 30 and 24 nM, respectively. The values (averages standard deviations [error bars]) were obtained from three independent experiments.

Figure 5

RNase H activity of wild-type and mutants RTs M41L/L210W/T215Y and M41L/L210W/T215Y/R284K. [³²P]RNA/DNA substrates (50 nM) were cleaved at 37°C in the presence of the corresponding RT at 50 nM concentration. Template-primer sequences are shown below. Arrows in the template sequences indicate the cleavage sites. For D38Trna/25PGA, the time points were taken after incubating the samples for 20 s, 40 s, and 1, 2 and 4 minutes. Catalytic rate constants for the cleavage of D38Trna were 0.34 ± 0.15 min^-1, 0.39 ± 0.18 min^-1 and 1.16 ± 0.54 min^-1 for WT, and mutant RTs M41L/L210W/T215Y and M41L/L210W/T215Y/R284K, respectively. For 31Trna/21P, the time points were drawn after 20 s, 40 s, and 1, 2, 3 and 4 minutes. The catalytic rate constants with this substrate were 0.33 ± 0.05 min^-1 for WT RT, and 0.35 ± 0.03 min^-1 and 0.87 ± 0.12 min^-1 for mutants M41L/L210W/T215Y and M41L/L210W/T215Y/R284K, respectively. Kinetic data were averages of three independent experiments.

Figure 6

Kinetics of the ATP-dependent excision of AZTMP and d4TMP from DNA/DNA template-primers. Time course experiments for the excision reaction of AZTMP- and d4TMP-terminated primers (26-mers) annealed to their corresponding 38-nucleotide DNA templates (30 nM) were determined in the presence of 3.2 mM ATP. The excision reaction was catalyzed by WT and mutant RTs (210 nM). The calculated k_obs values for the AZTMP excision reaction were 2.82 x 10^-3 ± 1.48 x10^-4 min^-1 for WT RT, 2.42 x 10^-3 ± 1.22 x 10^-4 min^-1 for mutant R284K RT, 4.69 x 10^-2 ± 1.69 x 10^-3 min^-1 for M41L/L210W/T215Y RT, and 3.64 x10^-2 ± 3.08 x 10^-3 min^-1 for M41L/L210W/T215Y/R284K RT. For the excision of d4TMP, the k_obs values for WT and mutants R284K, M41L/L210W/T215Y, and M41L/L210W/T215Y/R284K were 3.37 x 10^-3 ± 1.93 x 10^-4 min^-1, 2.62 x 10^-3 ± 1.42 x 10^-4 min^-1, 3.72 x 10^-2 ± 1.58 x 10^-3 min^-1, and 3.70 x 10^-2 ± 2.29 x 10^-3 min^-1, respectively.