Received 18 September 2008/Accepted 25 January 2009

Transcription

1 JOURNAL OF VIROLOGY, Apr. 2009, p Vol. 83, No X/09/$ doi: /jvi Copyright 2009, American Society for Microbiology. All Rights Reserved. The Human Immunodeficiency Virus Type 1 Nonnucleoside Reverse Transcriptase Inhibitor Resistance Mutation I132M Confers Hypersensitivity to Nucleoside Analogs Zandrea Ambrose, 1 * Brian D. Herman, 1 Chih-Wei Sheen, 1 Shannon Zelina, 1 Katie L. Moore, 4 Gilda Tachedjian, 4,5 Dwight V. Nissley, 2,3 and Nicolas Sluis-Cremer 1 Division of Infectious Diseases, Department of Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania ; Basic Science Program, SAIC-Frederick, Frederick, Maryland ; Gene Regulation & Chromosome Biology Laboratory, NCI-Frederick, Frederick, Maryland ; Molecular Interactions Group, Centre for Virology, Macfarlane Burnet Institute for Medical Research and Public Health, Melbourne, Victoria 3004, Australia 4 ; and Department of Microbiology, Monash University, Clayton, Victoria 3168, Australia 5 Received 18 September 2008/Accepted 25 January 2009 We previously identified a rare mutation in human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT), I132M, which confers high-level resistance to the nonnucleoside RT inhibitors (NNRTIs) nevirapine and delavirdine. In this study, we have further characterized the role of this mutation in viral replication capacity and in resistance to other RT inhibitors. Surprisingly, our data show that I132M confers marked hypersusceptibility to the nucleoside analogs lamivudine (3TC) and tenofovir at both the virus and enzyme levels. Subunit-selective mutagenesis studies revealed that the mutation in the p51 subunit of RT was responsible for the increased sensitivity to the drugs, and transient kinetic analyses showed that this hypersusceptibility was due to I132M decreasing the enzyme s affinity for the natural dctp substrate but increasing its affinity for 3TC-triphosphate. Furthermore, the replication capacity of HIV-1 containing I132M is severely impaired. This decrease in viral replication capacity could be partially or completely compensated for by the A62V or L214I mutation, respectively. Taken together, these results help to explain the infrequent selection of I132M in patients for whom NNRTI regimens are failing and furthermore demonstrate that a single mutation outside of the polymerase active site and inside of the p51 subunit of RT can significantly influence nucleotide selectivity. Downloaded from Human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) is a key target for antiretroviral drug development. To date, 12 RT inhibitors have been approved for the treatment of HIV-1 infection and can be classified into two distinct therapeutic groups. These include the nucleoside/nucleotide RT inhibitors (NRTIs) that block HIV-1 replication by acting as chain terminators in DNA synthesis and the nonnucleoside RT inhibitors (NNRTIs) that are allosteric inhibitors of HIV-1 RT DNA polymerization reactions. Although combination therapies that contain two or more RT inhibitors have profoundly reduced morbidity and mortality from HIV-1 infection, their long-term efficacy is limited by the selection of drug-resistant variants of HIV-1. Antiviral drug resistance is defined by the presence of viral mutations that reduce drug susceptibility compared with the drug susceptibilities of wild-type (WT) viruses. Whether or not a particular drug-resistant mutant develops depends on the extent to which virus replication continues during therapy, the ease of acquisition of the particular mutation, and the effect that the mutation has on drug susceptibility and viral fitness. In this regard, we recently detected a novel but rare NNRTI resistance mutation at codon 132 (I132M) in RTs of clinical * Corresponding author. Mailing address: Division of Infectious Diseases, University of Pittsburgh School of Medicine, 817B Scaife Hall, 3550 Terrace St., Pittsburgh, PA Phone: (412) Fax: (412) ambrosez@dom.pitt.edu. Published ahead of print on 4 February isolates from patients for whom NNRTI therapy was failing (6, 16). In vitro analyses showed that the I132M mutation in HIV-1 RT conferred high-level resistance to nevirapine and delavirdine ( 10-fold that of the WT) and low-level resistance ( 2- to 3-fold that of the WT) to efavirenz (18). In fact, the levels of resistance conferred by I132M in RT were essentially similar to those conferred by the Y181C mutation (J. Radzio, C. W. Sheen, and N. Sluis-Cremer, unpublished results). According to the Stanford University HIV Drug Resistance Database, combination therapies that contain nevirapine select for the Y181C mutation in approximately 35% of patients for whom the therapies are failing. However, these therapies select for the I132M mutation in less than 0.5% of these patients. Accordingly, the primary objective of the present study was to determine why the I132M mutation in HIV-1 RT is infrequently selected in patients for whom NNRTI-containing therapies are failing. MATERIALS AND METHODS Enzymes and viruses. The I132M, A62V, and L214I mutations were introduced into the WT HIV-1 LAI molecular clone or the p6hrt-prot prokaryotic expression vector (21) by site-directed mutagenesis using the QuikChange mutagenesis kit (Stratagene, La Jolla, CA). Full-length sequencing of mutant RTs was performed to confirm the presence of the desired mutations and to exclude adventitious mutations introduced during mutagenesis. WT and mutant recombinant HIV-1 RTs were expressed and purified to homogeneity as described previously (12, 13). For subunit-selective mutagenesis, the p66 and p51 RT genes were cloned into the pet-duet vector (Novagen-EMD Biosciences Inc., San Diego, CA) and enzymes were purified using a double-tag strategy as described on October 31, 2018 by guest 3826

2 VOL. 83, 2009 NOVEL HIV MUTANT HYPERSENSITIVE TO NUCLEOSIDE ANALOGS 3827 previously (14, 15). The RT concentration was determined spectrophotometrically at 280 nm by using an extinction coefficient (ε 280 ) of 260,450 M 1 cm 1. Virus stocks were made by the transfection of 293T cells with proviral plasmids by using Lipofectamine 2000 (Invitrogen, Carlsbad, CA). The titers of the viruses were determined using GHOST cells expressing the human CD4 and CXCR4 receptors (27) under single-cycle conditions, and the cells were analyzed for infection by flow cytometry with a FACSCaliber instrument (BD Biosciences, San Jose, CA). Antiviral assays. Antiviral assays were performed with TZM cells by using different concentrations of zidovudine (AZT; Sigma, St. Louis, MO), lamivudine (3TC; Moravek, Brea, CA), and tenofovir (NIH AIDS Research and Reference Reagent Program, Rockville, MD) as described previously (1). Briefly, TZM-bl cells (4) seeded into 24-well plates were infected in duplicate in the presence or absence of the drug. After 48 h, cells were lysed and analyzed for luciferase expression. Results were expressed as luciferase counts per second and are shown as the percentage of cells infected with each virus at each dilution of the drug relative to the proportion of cells infected without the drug. Inhibition of WT and mutant HIV-1 RTs by nucleoside analogs. Fixed-timepoint assays were used to evaluate HIV-1 RT-associated RNA-dependent DNA polymerase activity, as reported previously (18). Assays were carried out using both heteropolymeric and homopolymeric template/primer (T/P) substrates. For the heteropolymeric T/P, the sequences of the DNA primer and template were 5 -TCGGGCGCCACTGCTAGAGA-3 and 5 -CTCAGACCCTTTTAGTCAG AATGGAAAGTCTCTAGCAGTGGCGCCCGAACAGGGACA-3, respectively. Poly(rA)-oligo(dT) 18 was used as the homopolymeric T/P substrate. Both the heteropolymeric and oligo(dt) 18 primers were synthesized with a biotin label on their 5 terminus. DNA polymerase reactions using heteropolymeric T/P were carried out in mixtures of 50 mm Tris-HCl, ph 7.5 (37 C), 50 mm KCl, and 10 mm MgCl 2 containing 600 nm T/P, 10 M each 3 H-labeled deoxynucleoside triphosphates (dntp), and various concentrations (0 to 500 nm) of AZTtriphosphate (AZT-TP), 3TC-triphosphate (3TC-TP), or tenofovir-diphosphate (tenofovir-dp). Reactions were initiated by the addition of 25 nm RT, reaction mixtures were incubated for 20 min at 37 C, and then reactions were quenched with 0.5 M EDTA. Streptavidin scintillation proximity assay beads (GE Healthcare, Piscataway, NJ) were then added to each reaction mixture, and the extent of radionucleotide incorporation was determined by scintillation spectrometry using a 1450 Microbeta liquid scintillation counter (PerkinElmer, Waltham, MA). DNA polymerase assays using homopolymeric T/P were carried out using identical experimental conditions except that 600 nm poly(ra)-oligo(dt) 18 and 10 M [ 3 H]dTTP were used as substrates. Pre-steady-state kinetic experiments. A rapid-quench instrument (model RQF-3; KinTek Corporation, Clarence, PA) was used for pre-steady-state experiments with reaction times ranging from 5 ms to 30 min. The typical experiment was performed at 37 C with solutions of 50 mm Tris-HCl (ph 8.0) containing 50 mm KCl, 10 mm MgCl 2, and various concentrations of nucleotides. All concentrations reported refer to the final concentrations after mixing. HIV-1 RT at 300 nm was preincubated with 50 nm DNA T/P substrate, prior to rapid mixing with nucleotides and divalent metal ions to initiate the reaction, which was quenched with 0.5 M EDTA. The sequences of the primer and template were 5 -TCGGGCGCCACTGCTAGAGA-3 and 5 -CTCAGACCCT TTTAGTCAGAATGGAAAGTCTCTAGCAGTGGCGCCCGAACAGGGA CA-3, respectively. The quenched samples were then mixed with an equal volume of gel loading buffer (98% deionized formamide, 10 mm EDTA, and 1 mg/ml each of bromophenol blue and xylene cyanol) and denatured at 85 C for 5 min, and the products were separated from the substrates on a 7 M urea 16% polyacrylamide gel. The disappearance of the substrate (20-mer) and the formation of the product (21-mer) were quantified using a GS525 molecular imager (Bio-Rad Laboratories, Inc., Hercules, CA). The data were fitted by nonlinear regression using Sigma Plot software (Jandel Scientific) and the appropriate equations (15). The apparent burst rate constant (k obs ) for each particular concentration of dntp was determined by fitting the time courses for the formation of the product (21-mer) with the following equation: [21-mer] A[1 exp( k obs t)], where A represents the burst amplitude and t represents time. The turnover rate (k pol ) and apparent dissociation constant (K d ) for dntp were then obtained by plotting the apparent catalytic rates (k obs values) against dntp concentrations and fitting the data with the following hyperbolic equation: k obs (k pol [dntp])/([dntp] K d ). Yeast two-hybrid assays. Protein-protein interactions were quantified using the -galactosidase ( -Gal) liquid assay performed with permeabilized Saccharomyces cerevisiae cells grown from at least three independent transformants, as described previously (25). Briefly, individual transformants were grown in 1 ml of SC-His-Leu (synthetic complete medium lacking histidine and leucine and containing 2% [wt/vol] glucose) at 30 C with aeration for 16 h before being diluted to an absorbance (at 600 nm) of 0.2 in 2.5 ml of SC-His-Leu. Cells were then allowed to grow with aeration at 30 C until the absorbance (at 600 nm) reached 0.5 to 0.8 before they were pelleted and stored at 20 C. Thereafter, they were permeabilized in 50 l of Y-PER yeast protein extraction reagent (Pierce) and assayed for -Gal activity by adding a 1-ml mixture of 60 mm Na 2 HPO 4,40mM NaH 2 PO 4, 10 mm KCl, and 1 mm MgSO 4 containing 40 mm 2-mercaptoethanol and then adding 200 l of a 4-mg/ml stock of orthonitrophenyl- -D-galactopyranoside. Reaction mixtures were incubated at 30 C, reactions were quenched by the addition of 0.5 ml of 1 M Na 2 CO 3, and the absorbance at 420 nm was read. -Gal activity (in Miller units) was calculated by using the following equation: Miller units A 420 1,000/incubation time (in minutes). Protein expression in yeast. To prepare yeast protein extracts for Western blot analysis, yeast transformants were grown and pelleted as described above. To extract protein, the cells were lysed at room temperature in 160 l of Yeast- Buster protein extraction reagent (Novagen) containing 1- g/ml concentrations of pepstatin, leupeptin, and aprotinin. The cells were subjected to a vortex for 10 s and allowed to mix on a rotating wheel for 20 min at room temperature. The supernatant was clarified after centrifugation at 13,000 g for 15 min. The total protein concentration for each protein extract was determined using Bradford reagent (Bio-Rad), and 5 to 15 g of total protein per well was loaded for Western blot analysis. Fusion protein expression in yeast was evaluated by Western blot analyses of lysates with Gal4 activation domain (Gal4AD) polyclonal antibodies (Upstate Biotechnology) and anti-lexa polyclonal antibodies (Invitrogen). Immunodetection was accomplished using an ECL-Plus kit (Amersham). HIV-1 replication capacity (RC). HUT-CCR5 cells were infected with each virus at a multiplicity of infection of Cell cultures were split every 2 to 3 days, and replication was monitored by the quantitation of HIV-1 p24 in the cell supernatants with an enzyme-linked immunosorbent assay system (Beckman Coulter, Miami, FL, or PerkinElmer, Waltham, MA). Supernatants from cultures in which the replication of WT or I132M virus had peaked were used to infect fresh HUT-CCR5 cells. After 3 to 4 days, genomic DNA was extracted from the cells using the QiaAmp DNA blood kit (Qiagen, Valencia, CA). The RT coding region was amplified by PCR from proviral DNA and TA cloned into the pcr2.1-topo vector (Invitrogen). From each culture, 16 clones containing an insert were sequenced. RESULTS Susceptibilities of WT and I132M HIV-1 and RTs to NRTIs. The I132M mutation in HIV-1 RT was identified in a phenotypic screen of clinical HIV-1 isolates from patients for whom NNRTI-containing therapies were failing (6, 16). While previous studies established the role of this mutation in NNRTI resistance (18), they did not evaluate cross-resistance to other RT inhibitors. Accordingly, we determined the susceptibilities of WT and I132M HIV-1 to the NRTIs AZT, 3TC, and tenofovir (Fig. 1). In comparison with the WT virus, I132M HIV-1 exhibited modest hypersusceptibility (a 0.5-fold increase in susceptibility) to AZT: the 50% effective concentrations (EC 50 s) of AZT, the concentrations required to inhibit HIV-1 replication by 50%, were 0.04 and 0.02 M for the WT and I132M viruses, respectively (Fig. 1A). Surprisingly, however, I132M HIV-1 showed significant hypersensitivity (an approximately 0.15-fold increase in sensitivity compared to that of the WT virus) to both 3TC and tenofovir. The EC 50 s of both of these analogs decreased from 2 M for WT HIV-1 to 0.3 M for I132M HIV-1 (Fig. 1B and C). To ascertain whether the observed 3TC and tenofovir hypersusceptibilities occurred at the enzyme level, we next expressed and purified recombinant WT and I132M HIV-1 RTs and determined the 50% inhibitory concentrations (IC 50 s) of 3TC-TP, tenofovir-dp, and AZT-TP for the RTs, the concentrations required to inhibit the RNA-dependent DNA polymerase activities of the enzymes by 50% (Table 1). The data derived from these in vitro experiments are consistent with the

3 3828 AMBROSE ET AL. J. VIROL. FIG. 1. Inhibition of WT and I132M HIV-1 by AZT, 3TC, and tenofovir. Assays were carried out as described in Materials and Methods, and each curve represents the average of results from three to four independent experiments. Error bars correspond to standard deviations. Virus infectivity is expressed as the percentage of cells infected with each virus at each concentration of the drug relative to the proportion of cells infected without the drug. The EC 50 s of AZT calculated for the WT and I132M viruses were 0.04 and 0.02 M, respectively, those of 3TC were 2 and 0.3 M, respectively, and those of tenofovir were 2 and 0.3 M, respectively. antiviral data presented in Fig. 1. In comparison with the WT enzyme, I132M RT showed 0.3-, 0.08-, and 0.2-fold increases in susceptibilities to inhibition by AZT-TP, 3TC-TP, and tenofovir-dp, respectively. Residue 132 forms part of the 7-8 loop (residues 132 to 140) in HIV-1 RT. In the 66-kDa (p66) subunit of RT, this loop is situated in the fingers domain and resides a substantial distance from both the DNA polymerase active site and the NNRTI-binding pocket. In the 51-kDa (p51) subunit of RT, the 7-8 loop is located at the dimer interface and contributes to the formation of the base of the NNRTI-binding pocket. We previously demonstrated that I132M in the p51 subunit of HIV-1 RT was responsible for the observed NNRTI resistance (18). To delineate the effects of I132M in each of the RT subunits on 3TC-TP susceptibility, we performed a subunitselective analysis using purified enzyme that harbored I132M in either the p66 or p51 subunit. The data (Table 1) clearly demonstrate that 3TC-TP resistance is conferred by the mutation present in the p51, and not the p66, subunit. I132M impairs HIV-1 heterodimer formation, but this phenotype does not account for the observed NRTI hypersensitivity. Mutations of residues I135, N136, N137, and E138 in the 7-8 loop of the p51 subunit of RT significantly decrease the dimeric stability of the enzyme (18). To determine whether the I132M mutation also had an impact on RT dimerization, TABLE 1. NRTI susceptibilities of WT and I132 M HIV-1 RTs containing subunit-specific mutations RT composition IC 50 a ( M) of: AZT-TP 3TC-TP Tenofovir-DP WT p66/wt p (1.0) (1.0) (1.0) I132M p66/i132m p (0.3) (0.08) (0.2) I132M p66/wt p (1.5) WT p66/i132m p (0.2) a The IC 50 s are reported as averages standard deviations of results from three independent experiments. Values in parentheses indicate the change in resistance (n-fold) compared to that of the WT. we analyzed the ability of the I132M RT to form functional heterodimers by using the yeast two-hybrid assay (25). In this assay system, p66 is fused to the LexA DNA-binding domain (the bait) and p51 is fused to Gal4AD (the prey). Specific interaction between the two subunits results in the transactivation of the lacz reporter gene in the yeast strain CTY10-5d, which permits the interaction of the RT subunits to be quantified by assaying for -Gal activity. Initially, -Gal activity in yeast coexpressing mutant p66 bait and p51 prey fusions engineered with the same mutations in each subunit was measured. This analysis revealed that I132M significantly decreased -Gal activity compared with that in yeast expressing the WT p66 bait and p51 fusion proteins (Fig. 2A). The observed defect in I132M HIV-1 RT dimerization was similar to that in I135A HIV-1 RT dimerization. Consistent with previously published data, the E138K mutation did not have an impact on RT dimerization (18). To delineate the effects of I132M in each of the subunits on RT dimerization, we performed a subunit-selective analysis. To analyze the effects of mutations in the p66 subunit, we cotransformed yeast cells with mutant p66 bait and WT p51 prey, and to analyze the effects of mutations in the p51 subunit, we cotransformed yeast cells with WT p66 bait and mutant p51 prey. Our data show that I132M in both subunits contributes to the observed decrease in -Gal activity (Fig. 2A). Decreases in -Gal activity, as measured in a yeast twohybrid assay, may represent a true diminution in RT subunit interaction or may be due to decreased expression levels of either the bait or prey fusions compared with that of the WT protein. To distinguish between these two possibilities, we compared the steady-state protein levels in yeast expressing I132M p66 bait and p51 prey fusion proteins with those in yeast expressing the WT protein. Yeast protein extracts from exponentially growing cells were prepared and subjected to Western blot analysis. The p66 bait was detected using LexA antibodies, and p51 prey was detected using Gal4AD antibodies. No differences in the steady-state levels of the p66 bait and p51

4 VOL. 83, 2009 NOVEL HIV MUTANT HYPERSENSITIVE TO NUCLEOSIDE ANALOGS 3829 FIG. 2. Effects of the I132M, I135A, and E138K mutations on RT dimerization and 3TC-TP susceptibility. (A) Effects of I132M, I135A, and E138K on the -Gal readout in the yeast two-hybrid RT dimerization assay. The yeast reporter strain CTY10-5d was cotransformed with p66 bait and p51 prey constructs containing the I132M, I135A, and E138K mutations and assayed for -Gal activity. The reported -Gal activity (abscissa) represents the average for three transformants from three independent experiments. (B) Autoradiogram showing the incorporation of 3TC-TP by I132M, I135A, and E138K HIV-1 RTs. A 200 nm concentration of WT or mutant RT was incubated with 20 nm heteropolymeric T/P (sequence provided in the figure) in mixtures of 50 mm Tris, ph 7.5, 50 mm KCl, and 10 mm MgCl 2 containing 0.5 M dctp, various concentrations (0 to 50 M) of 3TC-TP, and 20 M ddttp. Following incubation at 37 C for 15 min, reactions were quenched and samples were analyzed by denaturing gel electrophoresis as described in Materials and Methods. prey mutants and the WT subunits were observed (data not shown). This result indicates that the observed decreases in -Gal activity were entirely due to a diminution in RT subunit interactions. To determine whether this decrease in dimerization for I132M RT accounted for the observed NRTI hypersusceptibility, we next evaluated the susceptibilities of recombinant I132M, I135A, and E138K HIV-1 mutants to 3TC-TP. As described above, I135A decreases RT dimer stability whereas E138K in RT has no effect. Interestingly, all three mutations (i.e., I132M, I135A, and E138K) confer NNRTI resistance (17). The data from this study demonstrated that neither the I135A nor the E138K mutation in RT had an impact on 3TC-TP susceptibility (Fig. 2B). Taken together, these results suggest that decreases in RT dimerization do not cause NRTI hypersusceptibility. Therefore, the hypersusceptibility of I132M HIV-1 RT to NRTIs must be due to another mechanism. Pre-steady-state kinetic analyses of dctp and 3TC-TP incorporation by WT, I132M, and M184V HIV-1 RTs. To determine the mechanisms responsible for I132M HIV-1 RT hypersusceptibility to 3TC-TP, pre-steady-state analyses were carried out to elucidate the interactions of dctp and 3TC-TP with the polymerase active sites of WT, I132M, and M184V HIV-1 RTs (Fig. 3; Table 2). M184V RT was included as a control in this study because the M184V mutation confers significant resistance to 3TC (3, 26). The pre-steady-state kinetic experiments, the results of which are presented in Table 2, defined the maximum rates of nucleotide incorporation (k pol values), the nucleotide K d values, and the catalytic efficiencies of incorporation (k pol /K d values). The k pol /K d value for the incorporation of dctp by I132M RT was 5-fold less than the k pol /K d value calculated for the WT enzyme. This change was driven primarily by a decrease in affinity (K d ) for dctp at the polymerase active site of I132M HIV-1 RT. In contrast, the k pol /K d value for the incorporation of 3TC-TP by I132M RT was 3-fold higher than the k pol /K d value calculated for the WT enzyme, and this effect was driven by an increase in K d for 3TC-TP at the polymerase active site of I132M HIV-1 RT. As reported previously, 3TC-TP was not efficiently incorporated by M184V HIV-1 RT due to weak binding of the nucleoside analog at the mutant enzyme s active site (5). The selectivity of

5 3830 AMBROSE ET AL. J. VIROL. Downloaded from FIG. 3. Time and concentration dependence for the incorporation of 3TC-TP by WT and I132M HIV-1 RTs. (A and B) The concentration of 3TC-monophosphate-terminated primer product formed as a function of time is shown for WT (A) and I132M (B) HIV-1. Reactions used a 6:1 ratio of RT to T/P, as described in Materials and Methods. The experimental data were fit to a single exponential, as described in Materials and Methods. (C and D) The hyperbolic variation of k pol with the concentration of 3TC-TP in these reactions is plotted. Panel D highlights the differences in apparent rate constants (k app ) between the WT and I132M RTs at low 3TC-TP concentrations. The calculated K d and k pol values are reported in Table 2. RT, which is defined by the ratio of the k pol /K d value for the incorporation of dctp to the k pol /K d value for the incorporation of 3TC-TP, is an indication of the ability of the WT or mutant RT to discriminate between dctp and 3TC-TP. The WT enzyme favors the incorporation of the natural dntp over the nucleoside analog (selectivity 11.9). Whereas the M184V mutation increased this selectivity 170-fold, the I132M mutation decreased selectivity 0.06-fold. This value is consistent with the changes in EC 50 s and IC 50 s for the WT and I132M viruses or RTs reported in Fig. 1 and Table 1. on October 31, 2018 by guest TABLE 2. Pre-steady-state kinetic constants for binding and incorporation of dctp and 3TC-TP by WT, I132M, and M184V HIV-1 RTs a RT k pol (s 1 ) dctp K d ( M) k pol /K d value ( M 1.s 1 ) k pol (s 1 ) 3TC-TP K d ( M) k pol /K d value ( M 1.s 1 ) Selectivity b Change in resistance (n-fold) c WT I132M mutant M184V mutant a Data are the means standard deviations of results from at least three independent experiments. b Selectivity is defined as follows: (k pol /K d value for incorporation of dntp)/(k pol /K d value for incorporation of NRTI-TP). c The change in resistance was calculated as follows: selectivity of the mutant/selectivity of the WT.

6 VOL. 83, 2009 NOVEL HIV MUTANT HYPERSENSITIVE TO NUCLEOSIDE ANALOGS 3831 FIG. 4. RCs of WT and mutant HIV-1. (A) The replication kinetics of WT, I132M, A62V/I132M, and A62V HIV-1 in HUT-CCR5 cells are shown. (B) The replication kinetics of WT, I132M, I132M/L214I, and L214I HIV-1 in HUT-CCR5 cells are shown. RCs of WT and I132M HIV-1. Because viral replication fitness may substantially contribute to the pattern of resistance mutations selected during NNRTI therapy, we also measured the impact that the I132M mutation has on the HIV-1 RC. The RC was investigated by infecting the T-cell line HUT-CCR5 with WT HIV-1 or I132M HIV-1 at equal multiplicities of infection (Fig. 4A). Although approximately 100-fold more I132M HIV-1 particles than WT HIV-1 particles were used in the infection, the mutant virus exhibited a significantly delayed growth curve, with detectable replication occurring after 16 days postinfection (Fig. 4A). Therefore, these data suggest that the I132M mutation significantly diminishes viral RC. Previously, a variant containing A62V and I132M was isolated from samples generated for the detection of minority drug-resistant variants (8, 17). Because studies have shown that A62V can partially eliminate the replication defect associated with the K65R or Q151M mutation (10, 24), we also introduced this mutation into our WT and I132M HIV-1 molecular clones. While A62V alone had no effect on RC, it significantly improved the replication of I132M HIV-1 (Fig. 4A). However, HIV-1 carrying mutations A62V and I132M (A62V/I132M HIV-1) still exhibited a replication lag of approximately 7 days compared to the WT virus. To determine whether there was reversion at codon 132 or whether a compensatory mutation in I132M HIV-1 was selected, virus that eventually grew out from two independent replication assays was isolated and sequenced. In one experiment, M132 had reverted to the WT residue I132; in the second experiment, the L214I mutation was found to coexist with the I132M mutation. We introduced L214I into the WT and I132M molecular clones to further investigate the role of this mutation in viral RC. Both L214I HIV-1 and I132M/L214I HIV-1 replicated similarly to WT virus, suggesting that L214I is a compensatory mutation that improved the RC of virus containing the I132M mutation in RT (Fig. 4B). DNA polymerase activities of recombinant purified WT and mutant HIV-1 RTs. To determine the mechanisms responsible for the decreased RCs of viruses containing I132M, we expressed and purified WT, I132M, A62V/I132M, and I132M/ L214I HIV-1 RTs. The RNA-dependent DNA polymerase activities of these enzymes were evaluated. As reported previously, the I132M recombinant HIV-1 RT is 40 to 50% less active than the WT enzyme. However, both the A62V and L214I mutations restored the activities of the enzymes to near-wt levels (Fig. 5). We also introduced the A62V/I132M RT into yeast to perform the yeast two-hybrid assay for RT dimerization. However, the steady-state protein expression levels in yeast were significantly decreased compared to those of WT RT, thus preventing the quantification of RT dimerization. DISCUSSION The I132M mutation in HIV-1 RT was identified in a phenotypic screen of clinical HIV-1 isolates from patients for whom therapy containing an NNRTI was failing (6, 16), and we previously demonstrated that this mutation confers high-level resistance to nevirapine and delavirdine (18). Remarkably, in this study we have shown that I132M also leads to marked hypersusceptibility (0.06- to 0.1-fold-increased susceptibility compared to that of the WT) to the nucleoside analogs 3TC and tenofovir. While previous reports have demonstrated that NRTI resistance mutations elicit NNRTI hypersusceptibility (9, 22), the reverse situation (i.e., NNRTI resistance mutations conferring NRTI hypersusceptibility) is rare. The Y181C mutation has been shown previously to confer modest hypersus-

7 3832 AMBROSE ET AL. J. VIROL. FIG. 5. Activities of WT and mutant HIV-1 RTs determined using recombinant purified enzymes. Experiments were carried out as described in Materials and Methods using 600 nm poly(ra)-oligo(dt) 18 and 10 M [ 3 H]dTTP as substrates. Data are reported as the averages standard deviations of results from at least three separate experiments. The asterisk indicates a P value of 0.01 for comparison with the WT (Student s t test). ceptibility ( 0.5- to 0.8-fold-greater susceptibility) to AZT when present with AZT resistance-conferring mutations (11), but it may decrease susceptibility to other NRTIs, such as stavudine (2, 7). Furthermore, the mechanisms by which I132M and Y181C confer NRTI hypersusceptibility are different. Whereas Y181C diminishes the capacity of RT to excise AZT-monophosphate from chain-terminated primers (20), our study shows that I132M directly affects the ability of the enzyme to discriminate between the natural nucleotide and the nucleoside analog. To our knowledge, this is the first study that has identified a mutation outside of the polymerase active site and within the p51 subunit of HIV-1 RT that can significantly influence nucleotide selectivity. Pre-steady-state kinetic experiments demonstrated that, in comparison with the WT enzyme, I132M HIV-1 RT bound the natural dctp substrate with decreased affinity but 3TC-TP with increased affinity. While this finding provides a kinetic explanation for the observed NRTI hypersusceptibility, it does not address a structural mechanism. In the p51 subunit of RT, I132M is situated at the base of the 7-8 loop, which contributes to the dimer interface and also the formation of the base of the NNRTI-binding pocket. In this regard, the results of our studies do not suggest a link between RT dimerization and nucleotide selectivity at the DNA polymerase active site. However, previous studies have demonstrated communication between the NNRTI-binding pocket and the DNA polymerase active sites of RT that has an impact on nucleotide binding (23, 28), although the precise nature of the conformational changes responsible for the observed communication between the NNRTI-binding pocket and the polymerase active site have not been identified (28). In this regard, preliminary modeling experiments have also failed to provide a plausible structural explanation for the I132M-induced NRTI hypersensitivity (data not shown). HIV-1 containing the I132M mutation in RT was also found to replicate significantly less efficiently than the WT virus. This decrease in replication efficiency can be explained, in part, by the decreased DNA polymerase activity of the mutant RT observed in both the pre-steady-state and steady-state kinetic analyses. Characterization of A62V/I132M HIV-1 indicates that A62V partially eliminates the growth defect conferred by the I132M mutation. The A62V mutation has also been reported to eliminate the replication defect associated with the K65R and Q151M mutations in RT (10, 24). Of interest, Olivares et al. reported that the F130W substitution significantly diminishes viral replication efficiency but that a compensatory change at codon 58 (T58S) can mitigate this effect (19). Taken together, the results of that study and ours may suggest a functional link between the 7-8 loop (which contains residues 130 and 132) and the 3-4 loop (which contains residues 58 and 62). In addition, we selected a second-site compensatory mutation, L214I, in RT. According to the Stanford HIV Drug Resistance Database, L214I is a polymorphism (i.e., it exists in both treatment-naïve and -experienced individuals) and is not associated with drug resistance. Nevertheless, the results of our study show that this substitution almost completely compensated for the replication defect of I132M HIV-1, as well as the decrease in RT activity. This finding highlights the notion that the genetic backbone of a given virus may significantly affect the selection of drug resistance mutations or the interpretation of the effects of a given drug resistance mutation on viral fitness. In conclusion, our data show that the I132M mutation in the p51 subunit of HIV-1 RT confers NRTI hypersusceptibility and also decreases viral RC. Taken together, these findings help to explain why I132M is infrequently selected by treatment regimens containing either nevirapine or delavirdine, which most likely also include NRTIs, and further demonstrate that a single mutation outside of the polymerase active site and within the p51 subunit of HIV-1 RT can significantly influence nucleotide selectivity. ACKNOWLEDGMENTS We thank Steve Hughes and Vineet KewalRamani for helpful discussions and Matthew Bess and Tamera Kirkland for technical assistance. The research was supported by grant R01 GM from the National Institutes of Health to N.S.-C. and by federal funds from the National Cancer Institute, National Institutes of Health, under contract N01-CO (D.V.N.). G.T. was supported by NHMRC senior research fellowship , and the research was supported by NHMRC project grants and The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. REFERENCES 1. Ambrose, Z., V. Boltz, S. Palmer, J. M. Coffin, S. H. Hughes, and V. N. Kewalramani In vitro characterization of a simian immunodeficiency virus-human immunodeficiency virus (HIV) chimera expressing HIV type 1 reverse transcriptase to study antiviral resistance in pigtail macaques. J. Virol. 78: Baldanti, F., S. Paolucci, G. Maga, N. Labo, U. Hubscher, A. Y. Skoblov, L. Victorova, S. Spadari, L. Minoli, and G. Gerna Nevirapine-selected mutations Y181I/C of HIV-1 reverse transcriptase confer cross-resistance to stavudine. AIDS 17:

8 VOL. 83, 2009 NOVEL HIV MUTANT HYPERSENSITIVE TO NUCLEOSIDE ANALOGS Boucher, C. A., N. Cammack, P. Schipper, R. Schuurman, P. Rouse, M. A. Wainberg, and J. M. Cameron High-level resistance to ( ) enantiomeric 2 -deoxy-3 -thiacytidine in vitro is due to one amino acid substitution in the catalytic site of human immunodeficiency virus type 1 reverse transcriptase. Antimicrob. Agents Chemother. 37: Derdeyn, C. A., J. M. Decker, J. N. Sfakianos, X. Wu, W. A. O Brien, L. Ratner, J. C. Kappes, G. M. Shaw, and E. Hunter Sensitivity of human immunodeficiency virus type 1 to the fusion inhibitor T-20 is modulated by coreceptor specificity defined by the V3 loop of gp120. J. Virol. 74: Feng, J. Y., and K. S. Anderson Mechanistic studies examining the efficiency and fidelity of DNA synthesis by the 3TC-resistant mutant (184V) of HIV-1 reverse transcriptase. Biochemistry 38: Flys, T., D. V. Nissley, C. W. Claasen, D. Jones, C. Shi, L. A. Guay, P. Musoke, F. Mmiro, J. N. Strathern, J. B. Jackson, J. R. Eshleman, and S. H. Eshleman Sensitive drug-resistance assays reveal long-term persistence of HIV-1 variants with the K103N nevirapine (NVP) resistance mutation in some women and infants after the administration of single-dose NVP: HIVNET 012. J. Infect. Dis. 192: Gilleece, Y., C. Torti, S. Mandalia, B. G. Gazzard, D. Pillay, and A. L. Pozniak The prevalence of reduced zidovudine susceptibility in zidovudine-naive, antiretroviral-experienced HIV-1-infected patients. HIV Med. 4: Halvas, E. K., G. M. Aldrovandi, P. Balfe, I. A. Beck, V. F. Boltz, J. M. Coffin, L. M. Frenkel, J. D. Hazelwood, V. A. Johnson, M. Kearney, A. Kovacs, D. R. Kuritzkes, K. J. Metzner, D. V. Nissley, M. Nowicki, S. Palmer, R. Ziermann, R. Y. Zhao, C. L. Jennings, J. Bremer, D. Brambilla, and J. W. Mellors Blinded, multicenter comparison of methods to detect a drug-resistant mutant of human immunodeficiency virus type 1 at low frequency. J. Clin. Microbiol. 44: Haubrich, R. H., C. A. Kemper, N. S. Hellmann, P. H. Keiser, M. D. Witt, D. N. Forthal, J. Leedom, M. Leibowitz, J. M. Whitcomb, D. Richman, and J. A. McCutchan The clinical relevance of non-nucleoside reverse transcriptase inhibitor hypersusceptibility: a prospective cohort analysis. AIDS 16:F33 F Kosalaraksa, P., M. F. Kavlick, V. Maroun, R. Le, and H. Mitsuya Comparative fitness of multi-dideoxynucleoside-resistant human immunodeficiency virus type 1 (HIV-1) in an in vitro competitive HIV-1 replication assay. J. Virol. 73: Larder, B. A Azido-3 -deoxythymidine resistance suppressed by a mutation conferring human immunodeficiency virus type 1 resistance to nonnucleoside reverse transcriptase inhibitors. Antimicrob. Agents Chemother. 36: Le Grice, S. F., C. E. Cameron, and S. J. Benkovic Purification and characterization of human immunodeficiency virus type 1 reverse transcriptase. Methods Enzymol. 262: Le Grice, S. F., and F. Gruninger-Leitch Rapid purification of homodimer and heterodimer HIV-1 reverse transcriptase by metal chelate affinity chromatography. Eur. J. Biochem. 187: Maier, G., U. Dietrich, B. Panhans, B. Schroder, H. Rubsamen-Waigmann, L. Cellai, T. Hermann, and H. Heumann Mixed reconstitution of mutated subunits of HIV-1 reverse transcriptase coexpressed in Escherichia coli: two tags tie it up. Eur. J. Biochem. 261: Miranda, L. R., M. Gotte, F. Liang, and D. R. Kuritzkes The L74V mutation in human immunodeficiency virus type 1 reverse transcriptase counteracts enhanced excision of zidovudine monophosphate associated with thymidine analog resistance mutations. Antimicrob. Agents Chemother. 49: Nissley, D. V., E. Halvas, D. Garfinkel, J. Mellors, and J. Strathern Detection of rare drug-resistant HIV-1 reverse transcriptase variants using a sensitive phenotypic assay. Antivir. Ther. 8:S Nissley, D. V., J. Julias, J. Mellors, S. Hughes, and J. Strathern Detection and characterization of rare drug resistant HIV-1 RT variants using a sensitive phenotypic assay. Antivir. Ther. 9:S Nissley, D. V., J. Radzio, Z. Ambrose, C. W. Sheen, N. Hamamouch, K. L. Moore, G. Tachedjian, and N. Sluis-Cremer Characterization of novel nonnucleoside reverse transcriptase (RT) inhibitor resistance mutations at residues 132 and 135 in the 51 kda subunit of HIV-1 RT. Biochem. J. 404: Olivares, I., A. Mulky, P. I. Boross, J. Tozser, J. C. Kappes, C. Lopez- Galindez, and L. Menendez-Arias HIV-1 protease dimer interface mutations that compensate for viral reverse transcriptase instability in infectious virions. J. Mol. Biol. 372: Selmi, B., J. Deval, K. Alvarez, J. Boretto, S. Sarfati, C. Guerreiro, and B. Canard The Y181C substitution in 3 -azido-3 -deoxythymidine-resistant human immunodeficiency virus, type 1, reverse transcriptase suppresses the ATP-mediated repair of the 3 -azido-3 -deoxythymidine 5 -monophosphate-terminated primer. J. Biol. Chem. 278: Shi, C., and J. W. Mellors A recombinant retroviral system for rapid in vivo analysis of human immunodeficiency virus type 1 susceptibility to reverse transcriptase inhibitors. Antimicrob. Agents Chemother. 41: Shulman, N., A. R. Zolopa, D. Passaro, R. W. Shafer, W. Huang, D. Katzenstein, D. M. Israelski, N. Hellmann, C. Petropoulos, and J. Whitcomb Phenotypic hypersusceptibility to non-nucleoside reverse transcriptase inhibitors in treatment-experienced HIV-infected patients: impact on virological response to efavirenz-based therapy. AIDS 15: Spence, R. A., W. M. Kati, K. S. Anderson, and K. A. Johnson Mechanism of inhibition of HIV-1 reverse transcriptase by nonnucleoside inhibitors. Science 267: Svarovskaia, E. S., J. Y. Feng, N. A. Margot, F. Myrick, D. Goodman, J. K. Ly, K. L. White, N. Kutty, R. Wang, K. Borroto-Esoda, and M. D. Miller The A62V and S68G mutations in HIV-1 reverse transcriptase partially restore the replication defect associated with the K65R mutation. J. Acquir. Immune Defic. Syndr. 48: Tachedjian, G., H. E. Aronson, M. de los Santos, J. Seehra, J. M. McCoy, and S. P. Goff Role of residues in the tryptophan repeat motif for HIV-1 reverse transcriptase dimerization. J. Mol. Biol. 326: Tisdale, M., S. D. Kemp, N. R. Parry, and B. A. Larder Rapid in vitro selection of human immunodeficiency virus type 1 resistant to 3 -thiacytidine inhibitors due to a mutation in the YMDD region of reverse transcriptase. Proc. Natl. Acad. Sci. USA 90: Trkola, A., T. Ketas, V. N. Kewalramani, F. Endorf, J. M. Binley, H. Katinger, J. Robinson, D. R. Littman, and J. P. Moore Neutralization sensitivity of human immunodeficiency virus type 1 primary isolates to antibodies and CD4-based reagents is independent of coreceptor usage. J. Virol. 72: Xia, Q., J. Radzio, K. S. Anderson, and N. Sluis-Cremer Probing nonnucleoside inhibitor-induced active-site distortion in HIV-1 reverse transcriptase by transient kinetic analyses. Protein Sci. 16: