Hypersusceptibility to substrate analogs conferred by mutations in human immunodeficiency virus type 1 reverse transcriptase

Abstract

Human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) contains four structural motifs (A, B, C, and D) that are conserved in polymerases from diverse organisms. Motif B interacts with the incoming nucleotide, the template strand, and key active-site residues from other motifs, suggesting that motif B is an important determinant of substrate specificity. To examine the functional role of this region, we performed "random scanning mutagenesis" of 11 motif B residues and screened replication-competent mutants for altered substrate analog sensitivity in culture. Single amino acid replacements throughout the targeted region conferred resistance to lamivudine and/or hypersusceptibility to zidovudine (AZT). Substitutions at residue Q151 increased the sensitivity of HIV-1 to multiple nucleoside analogs, and a subset of these Q151 variants was also hypersusceptible to the pyrophosphate analog phosphonoformic acid (PFA). Other AZT-hypersusceptible mutants were resistant to PFA and are therefore phenotypically similar to PFA-resistant variants selected in vitro and in infected patients. Collectively, these data show that specific amino acid replacements in motif B confer broad-spectrum hypersusceptibility to substrate analog inhibitors. Our results suggest that motif B influences RT-deoxynucleoside triphosphate interactions at multiple steps in the catalytic cycle of polymerization.

FIG. 1.

Relationship of motif B to other structural components in the ternary complex of HIV-1 RT (Protein Data Bank entry 1RTD) (24). (A) Surface representation of conserved structural motifs in the polymerase domain. Amino acids 107 to 118 (motif A), 147 to 169 (motif B), and 180 to 189 (motif C) are shown, as assigned by Poch et al. (53). Residues 102 to 106 (motif A), 143 to 146 (motif B), and 190 (motif C) are omitted for clarity. Boundaries for motif F (residues 64 to 75) are based on recent alignments of viral RNA-dependent RNA polymerases (4, 76). The region designated “premotif A” (residues 75 to 91) was originally identified in an alignment of HIV-1 RT with negative-stranded RNA virus polymerases (45). (B) Location of several motif B amino acids and other residues discussed in the text. Motif F and premotif A have been removed for clarity. Amino acid substitutions at residues Q151 and P157 (labeled in red) are known to confer resistance to nucleoside analogs (25, 36, 61). Residues Y115 and M184 are key determinants of dNTP substrate selectivity in motifs A and C, respectively (40, 64). Mg²⁺ ions coordinated at the active site are shown as small black spheres. These views were produced using MacPyMOL version 0.95 (http://pymol.sourceforge.org).

FIG. 2.

Replication capacities of RT motif B mutants. 293T cells were transfected with the wild-type (WT) HIV-1 pR9 clone or with clones containing specific substitutions in motif B. Titers in the resulting cultures were measured using HeLa-P4 cells and normalized to the concentration of HIV-1 capsid p24 in the supernatant to determine the infectivity of each mutant relative to that of the wild-type virus. ND, not determined; Q151M+4, multinucleoside-resistant mutant Q151M/A62V/V75I/F77L/F116Y (25). *, statistically different from the wild type by one-way analysis of variance (P < 0.05).

FIG. 3.

Representative dose-response data for the pyrimidine analogs 3TC (A) and AZT (B). Profiles for wild-type HIV-1 (dotted lines) and several motif B variants (solid lines) are shown. Analog sensitivities were measured by quantitating the dose-dependent reduction of Lac⁺ foci in HeLa-P4 cells. The percentages of solvent-only control foci are plotted as a function of nucleoside analog concentration. Curves were generated using a sigmoidal regression equation (GraphPad Prism 4 software package [44]). The results for each strain are from a single assay, with two determinations of focus formation per drug concentration. These data are representative of the responses observed for each mutant in multiple independent experiments. •, P157S; ○, W153F; ▪, wild type; □, Q151A; ▿, Q151V; ▴, Q151M/A62V/V75I/F77L/F116Y; ▵, Q161G; ♦, Q151M; ⋄, G155Q.

FIG. 4.

Susceptibilities of motif B mutants to 3TC (A) and AZT (B). Concentrations of nucleoside analog required to inhibit Lac⁺ focus formation in HeLa-P4 cells by 50% (EC₅₀) were calculated by regression analyses of dose-response data (Fig. 3), as described in Materials and Methods. The abscissa is set at the EC₅₀ for wild-type (WT) virus; bars above and below the abscissa represent analog resistance and hypersusceptibility, respectively. Each bar represents the mean ± standard error from at least three independent experiments. *, significantly different from the wild type by one-way analysis of variance (P < 0.05); **, significantly different from the wild type by a paired t test (P < 0.05). ND, not determined. Variants Q151M+4 (Q151M/A62V/V75I/F77L/F116Y) and M184V, which exhibit high-level resistance to AZT and 3TC, respectively, served as positive controls in these assays (25, 55, 69).

FIG. 5.

Sensitivities of AZT-hypersusceptible mutants to PFA. EC₅₀s for PFA (A) and AZT (B) were measured and graphed as described in Materials and Methods and the legend to Fig. 3. Each bar represents the mean ± standard error from at least three independent experiments. *, significantly different from the wild type (WT) by one-way analysis of variance (P < 0.05).

FIG. 6.

Hypersusceptibility of Q151 mutants to substrate analog inhibitors. Data from Fig. 4 and 5 and Table 2 are plotted to illustrate the change (n-fold) in EC₅₀ for each Q151 mutant relative to that of wild-type HIV-1. Values above and below the x axis are resistant and hypersusceptible to drug, respectively. Horizontal bars indicate the median change (n-fold) in EC₅₀ for each analog for the entire set of Q151 mutants. ▪, Q151G; □, Q151M; ▾, Q151V; ▴, Q151A; ♦, Q151I; •, Q151C.