INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Considerable progress has been made in the development of anti-human immunodeficiency virus type 1 (HIV-1) drugs and drug therapies. However, the emergence of drug-resistant viral strains presents a major problem; understanding the mechanisms that underlie drug resistance should be an important part of the effort to develop more effective drugs. Most of the available drugs target one of two viral enzymes, reverse transcriptase (RT) and protease (PR). There are two classes of RT inhibitors, nucleoside analogs and nonnucleosides. The nucleoside analogs used to treat HIV-1 infections lack the normal 3' OH of the ribose ring. The compounds are usually given to patients in an unphosphorylated state. The compounds are taken up by cells and converted to triphosphates by cellular enzymes. In this form, the analogs can be incorporated into viral DNA by HIV-1 RT; once incorporated, nucleoside analogs act as chain terminators, blocking viral DNA synthesis. One of the nucleoside analogs commonly used to treat HIV-1 infections is 3TC. 3TC treatment selects for drug-resistant viruses that have the methionine normally present at position 184 replaced either by isoleucine or valine. Viruses that have either the M184I or M184V mutation are quite resistant to 3TC; purified recombinant HIV-1 RTs that carry these mutations are resistant to 3TCTP in simple in vitro polymerization assays (9, 11, 18).

Nucleoside analogs inhibit reverse transcription (and viral replication) because they are incorporated into viral DNA by HIV-1 RT. Resistance to nucleoside analogs implies that the mutant RT has an increased discrimination between the analogs and their normal counterparts. In order for the virus to replicate, the mutant RT must be able to incorporate normal deoxynucleoside triphosphates (dNTPs) and synthesize viral DNA reasonably efficiently. Either the increased discrimination can occur at the time when the triphosphate form of the nucleoside analog is incorporated, or there can be enhanced excision of the analog after it has been incorporated. Resistance to AZT involves enhanced excision to AZTMP after it has been incorporated (2, 7, 21). In contrast, resistance to 3TC involves a block at the incorporation step (9, 11, 18).

Structural analysis (16, 24) led to the proposal that mechanism of resistance to 3TC of the M184V and M184I mutations is steric hindrance. In wild-type HIV-1 RT, there is a methionine at position 184, which is the X of the conserved YXDD motif; the two aspartates at positions 185 and 186 are part of the polymerase active site. 3TC has, instead of a ribose, an oxathiolane ring. The form of 3TC used as a drug is the opposite enantiomer relative to normal nucleosides. The introduction of the bulkier sulfur into the pseudoribose ring of 3TC and the use of the opposite enantiomer provide an opportunity for steric hindrance. The substitution of a -branched amino acid (either isoleucine or valine) for the methionine normally present at position 184 leads to steric hindrance; the  branch of the amino acid at position 184 clashes with the oxathiolane ring of 3TC. There is still enough room for normal dNTPs to be incorporated; the -branched amino acids create a steric gate that allows the incorporation of a normal dNTP but not 3TCTP (11, 16, 24). This exact mechanism also appears to account for the resistance of hepatitis B virus (HBV) RT and feline immunodeficiency virus (FIV) RT to 3TC (8, 24). In HBV RT, the -branched amino acid introduced into the YXDD motif is valine or isoleucine. In FIV RT, the amino acid selected at the X position is threonine (25). In contrast to HIV-1, in which the YXDD motif is YMDD, the RT of Moloney murine leukemia virus (MLV) has, in the wild-type enzyme, YVDD. In the case of MLV RT, the valine of the YVDD motif is at position 223. As might be expected from the results obtained with HIV-1, wild-type MLV is relatively resistant to 3TC in cell culture (14). However, MLV mutants with YMDD or YADD at the polymerase active site are also relatively resistant to 3TC. This prompted us to purify recombinant wild-type MLV RT and the YIDD, YMDD, and YADD mutants of MLV RT and to compare these MLV RTs to wild-type HIV-1 RT and the YADD mutant of HIV-1 RT. All of the MLV RTs were more resistant to 3TCTP than wild-type HIV-1 RT. The YMDD mutant of MLV RT had a moderate sensitivity to 3TCTP; the other MLV RTs were more resistant. Both the YADD enzymes (MLV and HIV-1) had obvious defects in simple polymerase assays. To explain these results, we have developed a model that incorporates the steric hindrance model for the resistance of the HIV-1 RT mutants M184V and M184I to 3TCTP (11, 16, 24).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

HIV-1 RT. The open reading frames encoding wild-type HIV-1 RT and each of the M184 mutants were cloned into a plasmid similar to p66HRT-PROT (3, 4, 19). The plasmid is based on the expression vector pT5m and was introduced into Escherichia coli strain BL21 (DE3)pLysE (4, 19, 23, 26). After induction with isopropyl--D-thiogalactopyranoside, the plasmid expresses both the p66 form of HIV-1 RT (either wild type or a mutant) and HIV-1 PR. Approximately 50% of the overexpressed p66 RT is converted to the p51 form by HIV-1 PR, and p66-p51 heterodimers accumulate in E. coli. The p66-p51 heterodimers were purified by metal chelate chromatography (4, 19, 20).

MLV RT. The expression clone for the MLV RT has been previously described (15). Briefly, codons for two extra amino acids (methionine and glycine) were added to the 5' end of the MLV RT coding region to generate an initiating ATG codon and an NcoI site. A termination codon and HindIII site were added to the 3' end of the coding region. The resulting construct was cloned into the expression vector pUC12N, which causes constitutive expression of the recombinant protein in E. coli. The protocol for purifying wild-type and mutant MLV RTs is given in reference 6.

MLV RT mutants. The mutations were generated using BspMI cassettes, similar to those used to generate mutations in HIV-1 RT (5). PCR amplification was used to generate two fragments from the MLV coding region. The first PCR amplification spans the 5' end of the MLV RT coding region. The 5' primer anneals to the pUC12N sequence 5' of the NcoI site and the initiation codon of MLV RT (5' GCGGGCAGTGAGCGCAACGC 3'). The 3' primer (5'GCGG CGGAATTCGCGACCTGCGGCCTGGGTGCTGGATCCGGAAGTCTGC 3') is complementary to the region of MLV RT near amino acid (aa) 213 and also includes BspMI and EcoRI recognition sequences (underlined). The PCR fragment was digested with NcoI and EcoRI and then purified. The second PCR amplification spans the 3' end of the MLV RT coding region. The 5' primer is complementary to the MLV RT coding region near aa 230 and also includes EcoRI and BspMI recognition sequences (5'GCGGCGGAATTCGCGACCTGCGGCCCCACTTCTGAGCTAGACTGCCAAC 3'). The 3' primer anneals to the pUC12N vector sequence 3' of the TAG termination codon and the HindIII recognition sequence (5' GTAAAACGACGGCCAGTGCCAAG 3'). The PCR fragment was digested with EcoRI and HindIII and purified. This fragment and the NcoI/EcoRI PCR fragment described above were coligated into NcoI/HindIII-digested pUC12N. The resulting clone, designated MLV RT 1-2, has a deletion in the coding region of MLV RT between aa 213 and 230 which is replaced by an EcoRI recognition sequence flanked by two BspMI recognition sequences oriented in opposite directions.

Construction of plasmid used to synthesize RNA in vitro. A 35-base fragment from the HIV-1 provirus clone pNL 4-3 (1), including the polypurine tract from the HIV-1 genome (positions 9049 to 9083), was linked to 30 adenines and inserted into the EcoRI and HindIII restriction sites of plasmid pGEM-3Zf (Promega, Madison, Wis.). The sequence and structure of the resulting plasmid, pGPA35, were verified both by restriction enzyme mapping and by double-stranded DNA sequencing (10).

RNA template preparation. The RNA template used in the RNase H cleavage assays was prepared from linearized plasmid DNA (pGPA35) by in vitro runoff transcription with T7 RNA polymerase, using a MEGAscripts RNA synthesis kit (Ambion, Inc., Austin, Tex.) in the presence of 120 µCi of [-³²P]UTP. The transcription reaction was heat inactivated at 70°C for 20 min. The resulting RNA was purified using the PolyA Tract mRNA isolation system, composed of biotinylated oligo(dT) and streptavidin-coated magnetic particles (Promega, Madison, Wis.). The amount of radioactive UTP incorporation into RNA was determined by scintillation counting (10).

DNA oligonucleotide preparation. DNA oligonucleotides were synthesized by BioServe Biotechnologies (Laurel, Md.). The lyophilized oligonucleotides were dissolved in diethyl pyrocarbonate-treated water and stored at 20°C.

RNase H cleavage assays. ³²P-labeled RNA template (50,000 cpm) that was synthesized from linearized pGPA35 (~100 ng) was hybridized to approximately 20 ng of the individual oligonucleotide as described above in the presence of 50 mM Tris-Cl (pH 8.0), 50 mM NaCl, 5 mM MgCl₂, 2.0 mM dithiothrectol (DTT), 100 µg of acetylated bovine serum albumin (BSA), and 10 mM CHAPS {3- [(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate}. The mixtures of RNA and oligonucleotides were heated to 70°C for 10 min and then slowly cooled to room temperature. The reactions were initiated by adding 45 ng of purified wild-type or mutant HIV-1 RT and MgCl₂ to a final concentration of 5 mM in a final volume of 12 µl and were then incubated at 37°C. Samples were removed at 0.25, 1, 4, and 16 min, and the reactions were terminated by adding 2× RNA loading buffer. The products were heat denatured and separated on a denaturing 15% polyacrylamide-7 M urea gel in Tris-borate-EDTA buffer at 1,600 V for approximately 90 min (10). The gel was dried and autoradiographed for several hours or overnight.

RNase H cleavage inhibition assay. The RNase H inhibition assay was based on a standard RNase H assay (10). Purified wild-type or mutant HIV-1 RT (120 ng) was first mixed with dCTP or 3TCTP at different concentrations, as indicated in the figure legends, and incubated at room temperature for 2 min. The low-salt reactions were initiated by adding annealed ³²P-labeled RNA template and DNA oligonucleotide in the presence of 50 mM Tris-Cl (pH 8.0), 50 mM NaCl, 2.0 mM DTT, 75 µg of acetylated BSA per ml, 10 mM CHAPS, and 5 mM MgCl₂. Aliquots of the reactions were removed at 0.25, 1, 4, and 16 min, and the reactions were terminated by adding these aliquots to equal amounts of 2× RNA loading buffer. The products were heat denatured and separated as described above (see also reference 10). The high-salt RNase H cleavage inhibition assay was done in 50 mM Tris-Cl-100 mM KCl-20 mM MgCl₂-2.5 mM DTT-75 µg of acetylated BSA per ml-10 mM CHAPS-2.5% glycerol (11).

Polymerization assays. (i) 3TCTP inhibition. For each sample, 0.25 µg of single-stranded M13mp18 DNA (New England Biolabs, Beverly, Mass.) was hybridized to 0.5 µl of the 47 sequencing primer (1.0 optical density unit [OD]/ml; New England Biolabs) by heating to 96°C and slow cooling to room temperature. The template-primer was extended by adding 1.0 µg of wild-type or mutant HIV-1 RT in a mixture containing 25 mM Tris (pH 8.0), 75 mM KCl, 8.0 mM MgCl₂, 100 µg of BSA per ml, 10 mM CHAPS, 2.0 mM DTT, 10 µM each dATP, dGTP, and dTTP, 5.0 µM [-³²P]dCTP, and the indicated concentrations of 3TCTP (Moravek Biochemicals, Brea, Calif.) in a 100-µl reaction volume. The mixture was incubated at 37°C for 30 min. The reaction was halted by the addition of 3 ml of ice-cold trichloroacetic acid (TCA), and the precipitated DNA was collected by suction filtration through Whatman GF/C glass filters. The amount of incorporated radioactivity was determined by liquid scintillation counting. Reactions involving MLV RT were done using similar conditions; however, for MLV RT, the concentration of KCl was 30 mM.

(ii) Low-dNTP extension assay. For each sample, 0.5 µl of the 47 primer (1.0 OD/ml; New England Biolabs) was 5' end labeled with [-³²P]ATP and T4 polynucleotide kinase. After incubation at 37°C for 30 min, the primer was separated from the unincorporated nucleotide by passage through Sephacryl S-200 HR (Sigma, St. Louis, Mo.); then 0.25 µg of single-stranded M13mp18 DNA (New England Biolabs) was annealed to the labeled primer as described above.

(iii) MLV RT polymerase incorporation assays. The MLV RT polymerase assays using either poly(rC) · oligo(dG) or M13mp18 and the 47 primer DNA as the template-primer are similar to the HIV-1 RT assays. For each sample to be assayed, either 0.25 µg of single-stranded M13mp18 DNA was hybridized to 0.5 µl of 47 sequencing primer (1.0 OD/ml) or 0.1 U of poly(rC) · oligo(dG) (Pharmacia) was used. For each sample, 1.0 µg of wild-type MLV RT or mutant MLV RT was added to the template-primer in 25 mM Tris-Cl (pH 8.0)-30 mM KCl-8.0 mM MgCl₂-2.0 mM DTT-100 µg of BSA per ml-and 10.0 mM CHAPS. For poly(rC) · oligo(dG), the reaction was supplemented with 10.0 µM dGTP and 0.025 µM [³²P]dGTP. For 47 plus M13mp18 DNA, the reaction was supplemented with 10.0 µM each dATP, dGTP, and dTTP, 5.0 µM dCTP, and 0.02 µM [³²P]dCTP. The final reaction volume was 100 µl. The reactions were allowed to proceed at 37°C for 30 min and then halted by the addition of 3 ml of ice-cold 10% TCA, and the precipitated DNA was collected by suction filtration through Whatman GF/C glass filters. The amount of incorporated DNA was determined by liquid scintillation counting.

(iv) Extension assay. The extension assay is similar to the low-dNTP assay described above. Briefly, 47 sequencing primer (New England Biolabs) was 5' end labeled with [-³²P]ATP and T4 polynucleotide kinase. After purification, the labeled primer was annealed to single-stranded M13mp18 DNA (New England Biolabs) by heating and slow cooling. For each sample, 1.0 µg of wild-type MLV RT or MLV RT variant was added to the labeled template-primer in 25 mM Tris-Cl (pH 8.0)-30 mM KCl-8.0 mM MgCl₂-2.0 mM DTT-100 µg of BSA per ml-10.0 mM CHAPS. The reaction mix was allowed to sit at room temperature for several minutes to allow the RT to bind to the template-primer. The reaction was initiated by the addition of dATP, dCTP, dGTP, and dTTP to a final concentration of 10.0 µM. The final reaction volume was 100 µl. The reactions were allowed to proceed at 37°C for 10 min and then halted by phenol-chloroform extraction. The samples were precipitated by the addition of 1 volume of isopropanol, fractionated by electrophoresis on a 6.0% polyacrylamide gel, and autoradiographed.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Halvas et al. (14) used an MLV-based vector to test the 3TC sensitivity or resistance of wild-type MLV and MLV mutants with alterations in the YXDD motif of RT. Not only was wild-type MLV resistant to 3TC, but so were the V223I, V223M, V223A, and V223S variants (position 223 of MLV RT is equivalent to position 184 of HIV-1 RT). This was a surprising finding; the simple form of the current model for the resistance of HIV-1 to 3TC predicts that wild-type MLV RT and the V223I mutant would be 3TC resistant but that the other MLV RTs would be sensitive. We prepared recombinant wild-type MLV RT and the V223M, V223I, and V223A mutants (Materials and Methods).

Judging from the relative titers of the MLV vectors carrying the various mutants (14), we expected the V223M and V223I mutants to have polymerase activities approximately equivalent to the wild-type level and the V223A to be slightly impaired. The various enzymes were compared for the ability to copy single-stranded DNA (M13) from the 47 primer and to use poly(rC) · oligo(dG) as a template-primer. All of the enzymes had similar activities with poly(rC) · oligo(dG); with the M13/47 template-primer, V223M was similar to wild-type MLV RT, the V223I enzyme was significantly more active than the wild-type MLV RT, and the V223A mutant was considerably less active than wild-type MLV RT (Table 1). We also did extension assays using M13 as a template; the results were similar to what was found with the M13 incorporation assay. V223M was similar to and V223I was better than wild-type MLV RT in the extension assay; V223A was less efficient (Fig. 1). The differences between wild-type MLV RT and the mutants are more apparent in the assay shown in Table 1 (reaction time, 30 min) than in the extension assays in Fig. 1 (reaction time, 10 min). We believe that the longer incubation time enhanced the differences. These recombinant enzymes were tested for their sensitivity or resistance to 3TCTP in an in vitro polymerization assay; wild-type HIV-1 RT was included as a control. As expected, wild-type HIV-1 RT was quite sensitive to 3TCTP; wild-type MLV RT was fully resistant (Fig. 2). The other MLV RTs showed intermediate levels of resistance. The V223A and V223I enzymes showed some sensitivity to 3TCTP but were clearly more resistant than the V223M variant.

View larger version (39K): [in this window] [in a new window]   FIG. 1.   Extension assay for HIV-1 RT, MLV RT, and the MLV RT mutants V223M, V223I, and V223A. The 47 sequencing primer was phosphorylated with [-32P]ATP, purified, and hybridized to M13mp18 DNA. Polymerization reactions were allowed to proceed for 10 min at 37°C and stopped by phenol-chloroform extraction (see Materials and Methods). The DNA was recovered by isopropanol precipitation and fractionated on a 6% polyacrylamide gel. Bands were visualized by autoradiography (see Materials and Methods). All reactions were done as duplicates (lanes 1 and 2), the nature of the RT used in the reactions is given above each lane. The sizes of DNA molecular weight markers are given on the left side. WT, wild type.

View larger version (14K): [in this window] [in a new window]   FIG. 2.   Effects of 3TCTP on polymerization of HIV-1 RT, MLV RT, and the MLV RT mutants V223M, V223I, and V223A. Polymerization assays were performed with M13mp18 DNA as a template in the presence of [-32P]dCTP (see Materials and Methods). dATP, dGTP, and dTTP were present at a concentration of 10 µM, and the dCTP concentration was 5 µM. Increasing amounts of 3TCTP were added to the reactions. The reactions were allowed to proceed for 30 min and were stopped by the addition of ice-cold TCA; the DNA was collected on GF/C glass fiber filters. Radioactivity was measured using a liquid scintillation counter. The enzymes synthesized different amounts of DNA; to simplify the comparisons, the data for each enzyme were normalized. WT, wild type.

The substantial resistance of the V223A mutant suggested a parallel experiment with the M184A mutant of HIV-1 RT. This enzyme was also quite resistant to 3TCTP in an in vitro polymerase assay; in an in vitro assay, it appeared to be as resistant to 3TCTP as the M184I and M184V mutants (Fig. 3). Although M184A was quite resistant to 3TCTP in a simple in vitro assay, it is not selected by 3TC treatment in patients or in cell culture. We tested the ability of the M184A mutant to extend a primer at low dNTP concentrations. The enzyme is moderately deficient compared to wild-type HIV-1 RT and to the M184I and M184V mutants that can be selected in vivo (Fig. 4). A simple interpretation of these data is that the enzyme cannot interact with an incoming dNTP to form the catalytically relevant closed complex as well as wild-type RT (see Discussion).

View larger version (14K): [in this window] [in a new window]   FIG. 3.   Effects of 3TCTP on the polymerization of the HIV-1 RT mutants M184V, M184I, and M184A. The assay is similar to the assay described in the legend to Fig. 2 (see also Materials and Methods). The data for each enzyme were normalized.

View larger version (69K): [in this window] [in a new window]   FIG. 4.   Extent of DNA synthesis at limiting dNTP concentration. The 47 primer was 5' end labeled with T4 polynucleotide kinase. The primer was annealed to M13mp 18 single-stranded DNA. Polymerase reactions were performed with wild-type HIV-1 RT (WT) and the M184V, M184I, and M184A mutants at various dNTP concentrations (0.1, 0.5, and 2.0 µM) at 37°C for 30 min (see Materials and Methods). The reactions were stopped by extraction with phenol-chloroform, and the DNA was recovered by precipitation with isopropanol. The samples were fractionated by electrophoresis on a 6% gel and autoradiographed.

If the nucleotide at the 3' end of the primer strand lacks a 3' OH (here a dideoxynucleotide), and the appropriate incoming dNTP is provided, HIV-1 RT will form a stable closed ternary complex. In the ternary complex, the fingers of the p66 subunit close down on the incoming dNTP, forming the top of the dNTP binding pocket (16). This complex can be detected in a gel shift assay (11, 27), and the pattern of RNase H cleavage is changed relative to a binary complex composed of HIV-1 RT and the nucleic acid substrate (11). The most obvious change in the RNase H cleavage pattern is that the ternary complex fails to make the secondary cleavages that center around a position approximately 8 bases from the 3' end of the primer strand. We used this assay to monitor the ability of dCTP and 3TCTP to induce the closed ternary complex with wild-type HIV-1 RT and the M184V and the M184I mutants. In low salt, all of the enzymes can form complexes with either 3TCTP or dCTP; however, the complexes formed with the 3TC-resistant enzymes and 3TCTP are salt sensitive, suggesting that these complexes are strained due to steric hindrance between the -branched amino acid (I or V) at position 184 and the oxathiolane ring of 3TCTP (11).

Similar experiments were done with the M184A mutant of HIV-1 RT. The pattern of RNase H cleavage by the binary complex is the same for wild-type HIV-1 RT and the M184A mutant (Fig. 5A). As expected, the M184A mutant can bind dCTP and form the ternary complex in low salt (Fig. 5B). However, in low salt, a significantly higher concentration of 3TCTP is required for the M184A mutant to form the closed complex than for wild-type HIV-1 RT (11). The M184A enzyme can also form a ternary complex with 3TCTP in low salt (Fig. 5C). However, in high salt the M184A mutant does not form a stable complex with either dCTP or 3TCTP (Fig. 6). This suggests that in the case of M184A, there is some sort of steric hindrance for both 3TCTP and dCTP that prevents the formation of a stable closed complex (see Discussion).

View larger version (40K): [in this window] [in a new window]   FIG. 5.   Effects of dCTP and 3TCTP on the specificity of RNase H cleavage by the M184A mutant of HIV-1 RT. With appropriate nucleic acid substrates, HIV-1 RT makes two sets of RNase H cleavages. The primary cleavages are centered on a region about 17 nucleotides from the polymerase active site. Secondary cleavages occur in a region centered about 8 nucleotides from the polymerase active site. The RNA used in these assays is 81 nucleotides in length; it was hybridized to a DNA oligonucleotide that is 20 bases long. The RNA was labeled during synthesis; the sequence of the RNA and the choice of the labeled nucleotide preferentially labels the 5' half of the RNA. In these assays, the 17 cleavages produce labeled RNAs that migrate to a position just larger than the DNA marker that is 50 bases in length; the 8 cleavages produce labeled RNAs that migrate between the DNA markers at 40 and 50 bases (10, 11). If the end of the primer strand is a dideoxynucleotide and the cognate dNTP is provided, HIV-1 RT can form a stable ternary complex (see text). The ternary complex makes the 17 but not the 8 cleavages (11). The specific lack of the 8 cleavages can be used to monitor the ability of HIV-1 RT to form the ternary complex. (A) Comparison of the cleavages made by wild-type HIV-1 RT (WT) and the M184A mutant in the absence of added dCTP or 3TCTP. (B) Effect of binding increasing amounts of dCTP on the ability of the M184A mutant to make the 8 cleavages. The intact RNA is shown on the left (no RT). Each set of four lanes is a time course (0.25, 1, 4, and 16 min). Assays were done in the absence of dCTP () and at 300, 30, 3, and 0.3 µM dCTP. These assays were performed under conditions that we have defined as low salt (see reference 11 and Materials and Methods). (C) As for panel B except that increasing concentrations of 3TCTP were used.

View larger version (57K): [in this window] [in a new window]   FIG. 6.   Effects of high salt, dCTP, and 3TCTP on the specificity of RNase H cleavage. The assays are similar to those shown in Fig. 5 except that the reactions were done in high salt (see reference 11 and Materials and Methods). (A) Effects of increasing amounts of dCTP; (B) effects of increasing amounts of 3TCTP.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

HIV-1, HBV, and FIV are all sensitive to 3TC. In each case, 3TC-resistant viruses can be selected. All of the 3TC-resistant viruses have mutations in the YXDD motif of RT; in all cases, the amino acid normally present at the second position of the motif is replaced by a -branched amino acid (8, 24). Models based on the structures of a ternary complex of wild-type HIV-1 RT, DNA, and a dNTP (16) or of a binary complex of M184I and DNA (24) suggested that resistance is caused by steric hindrance between the -branched amino acid and the oxathiolane ring of 3TCTP. We have obtained biochemical data to support the model (11). However, Halvas et al. (14) showed, for MLV, that a -branched amino acid in the second position of the YXDD motif of RT was not required for 3TC resistance. We tested several MLV RTs, including the wild-type enzyme and the V223I, V223M, and V223A mutants, for their resistance to 3TCTP. The wild-type enzyme is completely resistant; V223M is modestly resistant. V223I and V223A show intermediate resistance, greater than that of V223M and less than that of wild-type MLV RT. These results prompted an examination of the M184A mutant of HIV-1 RT; the M184A mutant of HIV-1 RT is quite resistant to 3TCTP in vitro.

This raises two questions. First, for HIV-1, why isn't the M184A mutant selected in response to 3TC in vivo? Second, since alanine is not a -branched amino acid, what is the mechanism of resistance? In terms of the possible selection of M184A by 3TC treatment, there are two potential problems. First, it would take two mutations to convert the methionine codon to an alanine codon; only one change is needed to obtain either a valine or isoleucine codon. Second, and probably more important, the M184A mutant appears to be less efficient than wild-type HIV-1 RT or either the M184I or M184V mutant. Both data in the literature (22, 29) and our data suggest that the M184A mutant is less able to form the closed complex and incorporate normal dNTPs than the wild-type enzyme. The magnitude of the problem appears to be a function of both the template-primer (22, 29) and the salt concentration (Fig. 6). However, the M184A enzyme is not only less able to form the closed complex and incorporate normal dNTPs less well than wild-type HIV-1 RT; it is also less able than either M184V or M184I. It seems reasonable to assume that this difference accounts for the selection of M184V (or M184I) instead of M184A when the virus is challenged with 3TC either in culture or in a patient. It is difficult to know to what degree low dNTP levels actually limit HIV-1 replication in vivo, in part because HIV-1 can replicate in a variety of cell types. In quiescent peripheral blood lymphocytes (PBLs), the level of dNTPs range from a low of 0.32 µM for dATP up to 5.6 µM for dTTP. Viral DNA synthesis is completed by the wild-type enzyme, but slowly and inefficiently (12). It has been suggested that HIV-1 RT may act in a partially distributive manner at these low levels of dNTPs (12). In activated PBLs, the level of dNTPs is higher, ranging from 3.24 µM for dATP up to 26.13 µM dTTP. Measured K_m values in this system were approximately 2.6 to 4.0 µM for the various dNTPs (12). Our data suggest that the M184A enzyme is less efficient than either the M184I or the M184V enzyme over a considerable range of dNTP concentrations (0.1 to 2 µM), which would put it at a disadvantage in quiescent PBLs. It may be very inefficient at viral DNA synthesis in quiescent PBLs. In support of this idea, HIV-1 carrying the M184A mutation replicates quite poorly in vitro (28). Relatively poor replication, in at least some cell types, also appears to limit the selection of another 3TC-resistant mutation, M184T (17).

What then is the mechanism of resistance of the M184A mutant? We believe that it is still steric hindrance, even though alanine is not a -branched amino acid. The M184I structure is a binary complex (24). In this complex, the position of the DNA (both primer and template) is clearly different from the position occupied by the template-primer in the structure of wild-type HIV-1 RT. The significance of this observation for the structure of the ternary complex is unclear; whether there is a repositioning of the nucleic acid in the ternary complex is not known.

If we assume that there is repositioning of the template-primer in the ternary complex, and that repositioning also occurs in the M184A mutant, then there is a simple explanation for the data. The repositioning of the template-primer creates an opportunity for steric hindrance between the alanine at position 184 (or some other nearby amino acid) and the oxathiolane ring of 3TCTP. Although the simplest model would have the oxathiolane ring of 3TCTP interacting directly with 184A, it is possible that changes in the positions of the amino acid side chains could alter the polymerase active site in a fashion such that some other amino acid in the active site of the enzyme could have a direct role in creating the steric clash. A better understanding of the precise mechanism will probably require an X-ray crystallography structure complex of the M184A mutant, nucleic acid, and bound dNTP.

There are data in the literature which show that, for the M184A mutant, the nature of the nucleic acid substrate can profoundly affect the behavior of the enzyme. Wakefield et al. (28) reported that the M184A mutant can copy a poly(rA) template from an oligo(dT) primer almost as well as wild-type HIV-1 RT; however, it is much less active than wild-type RT with a poly(rA) template and an oligo(U) primer. Wilson et al. (29) showed, with homopolymeric RNA templates, that the M184A mutant has a substantially higher K_m for dCTP and dTTP. This shows that wild-type HIV-1 RT and the M184A mutant interact differently with different nucleic acid substrates. These data, taken together with the low-dNTP extension data (Fig. 4) and the salt sensitivity of the ternary complex (Fig. 6), suggest that changes in the structure of the polymerase active site of the YADD mutant or changes in the positioning of the nucleic acid (or both) produce a more confined dNTP binding site that leads to steric hindrance even with normal dNTPs. The differences seen with the different template-primers also suggest that the structure and/or position of the nucleic acid can play an important role in defining the dNTP binding site.

The model can be extended to explain the MLV RT data. The polymerase active site of MLV RT is quite similar to the polymerase active site of HIV-1 RT (13). For MLV RT, like HIV-1 RT, the YMDD version of the enzyme is considerably more sensitive to 3TCTP than the YVDD version. However, the YMDD version or MLV RT is more resistant to 3TCTP than the YMDD version of HIV-1 RT. If the ribose (or oxathiolane) ring is a little closer to the active site in the vicinity of the YXDD motif in MLV RT than in HIV-1 RT, the data can be explained. The oxathiolane ring is larger than the ribose ring, and since it is the opposite enantiomer there will be a steric clash with 3TCTP for wild-type MLV RT and the YMDD mutant; as expected, the extent of the clash (and the level of 3TC resistance) is greater for the -branched amino acid valine than for methionine. Although the polymerase active site of MLV RT is similar to the polymerase active site of HIV-1 RT, it is not identical. Differences in the exact positioning of the template-primer or in the relative positions of the amino acid side chains (or both) could account for the subtle differences seen with the individual amino acid substitutions in the YXDD motif of HIV-1 RT and MLV RT. The phenotypes of the individual YXDD mutants of MLV RT, and their relative sensitivity to 3TCTP, can be accounted for in terms of the effects of the individual amino acid substitutions on the precise structure at the polymerase active site, the exact position of the template-primer, or both.