HIV-1 Reverse Transcriptase Discriminates against Non-self tRNA Primers
Descrição do Produto
J. Mol. Biol. (1996) 264, 243–254
HIV-1 Reverse Transcriptase Discriminates against Non-self tRNA Primers Belinda B. Oude Essink, Atze T. Das and Ben Berkhout* Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, Meibergdreef 15 1105 AZ Amsterdam The Netherlands
The interactions between the Reverse Transcriptase (RT) of human immunodeficiency virus type 1 (HIV-1) and the natural tRNALys3 primer for initiation of viral DNA synthesis were examined. We constructed a set of HIV-1 RNA templates in which the wild-type primer binding site (PBSLys3) is replaced by sequences complementary to tRNAIle, tRNALys1,2, tRNAPhe, tRNAPro or tRNATrp and tested the ability of RT enzymes of different retroviral species to initiate cDNA synthesis from self versus non-self tRNA primers. We demonstrate that initiation of HIV-1 reverse transcription is a specific process that is most efficient with the self tRNALys3 primer. Interestingly, the property of HIV-1 RT to discriminate against non-self tRNA primers is lost upon extension of the tRNA by only two deoxyribonucleotides. Furthermore, selective tRNA priming by HIV-1 RT was not observed with viral RNA-tRNALys3 duplexes isolated from HIV-1 virion particles, suggesting that the majority of tRNALys3 primers annealed to viral RNA in particles is extended by a variable number of deoxyribonucleotides. This result indicates that reverse transcription is initiated relatively early in nascently assembled virions. 7 1996 Academic Press Limited
*Corresponding author
Keywords: reverse transcription; HIV-1 retrovirus; tRNA primer; primer-binding site
Introduction Retroviruses encode an RNA-dependent DNA polymerase (reverse transcriptase or RT) that generates a double-stranded DNA genome from a genomic RNA template (Baltimore, 1970; Temin & Mizutani, 1970; Coffin, 1984). Retroviral reverse transcription is initiated from a cellular tRNA molecule annealed to the viral genome at the primer-binding site (PBS). Different retroviruses utilize different tRNAs as primer for reverse transcription (Leis et al., 1993). For instance, Moloney murine leukemia virus and the human T-cell leukemia virus use the tRNAPro primer, the avian myeloblastosis virus uses tRNATrp and the human immunodeficiency viruses type 1 and 2 (HIV-1 and HIV-2) utilize tRNALys3. The tRNA primer molecule is likely to be involved in specific molecular interactions with viral proteins and the viral genomic RNA at several Abbreviations used: RT, reverse transcriptase; HIV-1 human immunodeficiency type; PBS, primer binding site; vRNA, viral RNA; wt, wild-type; AZT, 3'-azido-3'-deoxythymidine; LTR, long terminal repeat; PCR, polymerase chain reaction. 0022–2836/96/470243-12 $25.00/0
stages of the replication cycle (reviewed by Berkhout, 1996). Biochemical and virological evidence exists for specific packaging of the HIV-1 tRNALys3 primer species into virions by the viral RT protein (Barat et al., 1989, 1991, 1993; Richter-Cook et al., 1992; Mak et al., 1994; Oude Essink et al., 1995). In the virus particle the tRNA primer is attached to the viral RNA (vRNA) genome through base-pairing of the 3'-terminal 18 nucleotides with the complementary PBS sequence situated near the 5'end of the viral genome. Both the RT protein (Barat et al., 1993; Sarih-Cottin et al., 1992; Oude Essink et al., 1995; Wohrl et al., 1993) and the nucleocapsid protein NCp7 (Mely et al., 1995; Khan & Giedroc, 1992; Lapadat-Tapolsky et al., 1995; Barat et al., 1993; De Rocquigny et al., 1992) have been implicated in unwinding of the tRNA cloverleaf and subsequent PBS annealing. Little is known about the molecular interactions regulating the process of initiation of reverse transcription, but recent studies suggest that efficient initiation requires the combination of HIV-1 RNA, HIV-1 RT enzyme and fully modified tRNALys3 molecule (Arts et al., 1994a; Isel et al., 1996). Additional base-pairing contacts between the HIV RNA template and tRNALys3 have been suggested to play a role in 7 1996 Academic Press Limited
244 initiation of reverse transcription (Berkhout & Schoneveld, 1993; Isel et al., 1993, 1995), similar to models originally described for the avian retroviruses (Cobrinik et al., 1991; Aiyar et al., 1992, 1994). However, recent work suggests that the additional template–primer interactions do not affect the initiation, but rather the elongation process (Isel et al., 1996). In this study we compared the ability of different retroviral RT enzymes to initiate reverse transcription with various tRNA primers, including the natural or self tRNA and several heterologous or non-self tRNAs. With template–primer complexes assembled in vitro we demonstrate that HIV-1 reverse transcription is a specific process that is most efficiently initiated from the self tRNALys3 primer. Interestingly, such a specificity of reverse transcription is not observed with template– primers purified from HIV-1 virion particles. Subsequent analyses suggest that the genome-associated tRNALys3 primer isolated from virions is already extended by a variable number of deoxyribonucleotides. These results provide new insights into HIV-1 reverse transcription by demonstrating that initiation is a specific event that occurs relatively early in newly assembled virions.
Results Reverse transcription with vRNA–tRNA complexes isolated from PBS-mutated HIV-1 virions For the production of HIV-1 virions with RNA genomes that are in association with non-self tRNA primers, we replaced the wild-type (wt) PBS site of the infectious plasmid pLAI by sequences complementary to the 3' end of tRNAIle, tRNALys1,2, tRNAPhe, tRNAPro or tRNATrp. We found that these HIV-1 mutants are severely replication defective, and all mutants reverted upon prolonged culture to the wild-type PBSLys3 sequence through infrequent annealing of the wt tRNALys3 to the mutant PBS sites (Das et al., 1995). However, the transiently produced viruses contain the tRNA primer complementary to the new PBS site, albeit at reduced levels compared with wt HIV-1. Compared with the wt PBS, we measured tRNA-occupancy levels ranging from 3% for PBSTrp to 20% for PBSLys1,2 (ranking order: wtLys1,2 > Pro > Ile > Phe > Trp) (Das et al., 1995). We used this set of template–primer combinations for a tRNA-extension reaction in vitro. Similar amounts of vRNA template were used in a reverse transcription reaction that extends the associated tRNA primer into a radiolabeled tRNA–cDNA product (Figure 1B). The migration properties of these products on the gel vary according to the different lengths (Figure 1A) and base modifications of the tRNA primers (Das et al., 1995). A PBS-deletion mutant (PBS− ) was included as a negative control. To compare the priming
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
Figure 1. Reverse transcription of different tRNA– vRNA hybrids isolated from wild-type and mutant HIV-1 virions. A, Length in nucleotides of the individual tRNA primers and the tRNA–cDNA reverse transcription products. B, Reverse transcription assay of the set of virion-extracted tRNA–vRNA hybrids (indicated on top of the panel) by addition of all dNTPs and RT enzyme from HIV-1, AMV or MoMLV. The cDNA products were analyzed on a 6% polyacrylamide sequencing gel, of which a small segment is shown. (Please note that the amount of non-self tRNA primer annealed to the mutant PBS site is considerably lower than the occupancy of the wt PBSLys3 site (Das et al., 1995).) Thus, the intensities of the tRNA–cDNA signals reflect not only the actual priming efficiencies, but also the tRNA occupancies of the wt and mutated PBS sites. Therefore, one has to compare the activity spectra obtained with the three RT enzymes.
efficiencies of self versus non-self combinations of RT enzyme and tRNA primer, we performed this assay with the RT enzymes of HIV-1 (natural primer is tRNALys3 ), avian myeloblastosis virus (AMV, tRNATrp ) and Moloney murine leukemia virus (MoMLV, tRNAPro ). The three enzymes demonstrated a similar activity spectrum with this set of vRNA–tRNA complexes (Figure 1B). (Please note that the great difference in cDNA synthesis with the different primers does reflect differences in tRNA occupancy (wt Lys1,2 > Pro > Ile > Phe > Trp).) This typical activity spectrum was observed with all three retroviral RT proteins, suggesting that these enzymes do not preferentially extend their own tRNA primer species. These initial assays were performed with an excess of RT enzyme, a condition that may conceal subtle differences in priming capacity of the different tRNA–RT combinations. We therefore repeated a subset of the tRNA-extension reactions with progressively reduced amounts of RT enzyme (Figure 2). Extension of tRNALys3 (top panel) was assayed with three different amounts of HIV-1 RT (lanes 1 to 3), AMV RT (lanes 4 to 6) and MoMLV RT (lanes 7 to 9). In addition, these HIV-1 RT enzyme dilutions were tested in combination with
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
245
the non-self tRNAIle primer (second panel), and the AMV and MoMLV RT enzymes were also tested in combination with the self primers tRNATrp and tRNAPro (third and fourth panel, respectively). Furthermore, an antisense oligonucleotide DNA primer was annealed to the vRNA and extended by the different RT samples (bottom panel). All primers, the self and non-self tRNAs and the DNA oligonucleotide, were equally sensitive to reduction of the amount RT enzyme available for primer extension. Thus, no major differences in tRNApriming efficiency were measured with the retroviral RT enzymes in assays with virus-extracted template–primers. HIV-1 RT discriminates against non-self tRNA primers on in vitro assembled vRNA–tRNA templates The obvious disadvantage of the vRNA–tRNA combinations extracted from virion particles is the variable amount of tRNA annealed to the different templates. We therefore decided to repeat these experiments with in vitro synthesized HIV-1 templates containing the PBS mutations and calf liver tRNA as a source of all the tRNA primer species. Another reason to investigate tRNA priming efficiencies in this simplified system is that, in contrast to our results with virion-derived material, several studies reported specific priming with the HIV-1 RT–tRNALys3 combination (Arts et al., 1994a; Li et al., 1994; Isel et al., 1996). Initial reverse transcription assays were performed in the presence of all dNTPs, generating a strong-stop minus-strand cDNA product (see e.g. Figure 7). However, with these in vitro assembled vRNA–tRNA duplexes we consistently obtained better results in reactions that restrict the extent of cDNA synthesis by withholding a single dNTP nucleotide. In this protocol, RT elongates up to a position corresponding with the missing nucleotide. This occurs in a ‘‘ − dGTP reaction’’ on the authentic HIV-1 template RNA after incorporation of two nucleotides (including a radiolabeled dCTP). An additional advantage of this assay is that a distinct cDNA band is produced. In contrast, the ‘‘all dNTP assay’’ generates multiple cDNAs that are shorter than strong-stop cDNA because of pausing and/or premature termination of the low-processive RT enzyme. Because the extent of pausing varies strongly among the different RT enzymes (data not shown), it seems inappropriate to analyze tRNA-priming efficiencies by quantitating strong-stop cDNA production. We therefore used the ‘‘ − dGTP assay’’ with the in vitro assembled vRNA–tRNA duplexes. Interestingly, this assay produced inferior results when used in combination with the virus-extracted template– primers (see Figure 7 and Discussion). The PBS-mutated templates were synthesized by bacteriophage T7 RNA polymerase, and heatannealed to the appropriate primer present in the
Figure 2. Extension of self and non-self tRNAs with virion-derived template primers and limiting RT levels. Reverse transcription assays were performed as for Figure 1 with a tRNA primer (upper four panels, specific tRNA is indicated on the left of the panels) and the DNA primer (bottom panel). The DNA primer extension was performed on the PBSLys3 template. The DNA primer (CN1, bottom panel) is complementary to nucleotides 123 to 151 of the HIV-1 genome and produces an 151-nucleotide cDNA. Varying amounts of the three RT enzymes were used (schematically depicted on top of the panel): the standard amount and threefold and ninefold dilutions of HIV-1 RT enzyme (lanes 1 to 3), AMV RT enzyme (lanes 4 to 6), and MoMLV RT enzyme (lanes 7 to 9). All products were analyzed on a 6% sequencing gel, of which small segments are shown.
calf liver tRNA pool. This template–primer sample was split in three and reverse transcription was initiated by addition of dNTPs (−dGTP) and one of the three RT enzymes (Figure 3, left). The different migration of the tRNA–cDNA products corresponds to the unique length and base modifications of the tRNA primers (Das et al., 1995), confirming that selective annealing of the tRNA complementary to the PBS site did occur in this system. Product bands were quantitated with a Phosphor Imager (Figure 3 right, with the self primers marked by a filled-in bar for each individual RT enzyme). In this system, the tRNALys3 molecule is an efficient primer for the HIV-1 RT polymerase (Figure 3, upper panel). This experiment was performed with the p66/p66 homodimeric form of the HIV-1 RT enzyme, but similar results were
246 obtained with other RT enzyme preparations, including p66/p51 heterodimer samples. All nonself primers were extended with much less efficiency by HIV-1 RT. Consistent with a previous report (Li et al., 1994), we measured inefficient priming with tRNALys1,2, which differs from tRNALys3 at only 13 nucleotide positions. Several control reactions were performed. First, a DNAprimer extension assay (Figure 3, bottom panel) was included to check that equal levels of the PBS-mutated vRNA templates were used. Second, we know that an equivalent amount of each of the different tRNA primers was annealed to the wt and mutant PBS sites because all tRNAs efficiently primed cDNA synthesis with the MoMLV RT enzyme and, although with reduced efficiency, by
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
the AMV RT enzyme. Some minor differences in tRNA-priming efficiency were observed for the latter two enzymes. For instance, MoMLV seems to favor the non-self tRNAIle primer and AMV RT apparently initiates very inefficiently from the tRNALys1,2 primer, but these enzymes clearly do not prefer the self tRNA species in this experimental setting with the HIV-1 RNA template. Therefore, the reverse transcription reactions with MoMLV and AMV RT are suitable control experiments for the marked difference in tRNA primer usage observed for the HIV-1 RT enzyme. The specific priming observed with in vitro assembled complexes is different from the results obtained with virion-extracted vRNA–tRNA complexes. It is possible that HIV-1 RT enzyme triggers
Figure 3. Reverse transcription of in vitro assembled vRNA–tRNA template–primer complexes. The panels on the left represent gel analyses of the cDNA products obtained with different tRNA–template duplexes (indicated on top of the panel) in combination with either HIV-1, AMV or MoMLV-RH− RT. The same amounts of RT enzyme were used as in the experiments with in vitro assembled duplexes in Figure 1. All tRNA extension reactions on these HIV-1 templates were stopped after incorporation of two nucleotides, by withholding the dGTP nucleotide (see Materials and Methods for details). Quantification of the tRNA priming efficiencies is shown in the right-hand panels for all three RT enzymes. The self tRNA is marked for each RT enzyme by a filled bar (HIV-1, tRNALys3; AMV, tRNATrp; MoMLV, tRNAPro ). As a control for template RNA levels, we performed a DNA primer extension reaction with the CN1 oligonucleotide in combination with the AMV RT (bottom left panel).
247
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
Figure 4. Pre-incubation of a non-self tRNA–vRNA duplex with HIV-1 RT does not enhance the priming efficiency. The tRNALys3 and tRNATrp primers were annealed to the corresponding template and then either incubated (+) with HIV-1 RT enzyme for 30 minutes or mock-incubated (−, indicated on top of the panel). The HIV-1 RT protein was removed by proteinase K incubation and phenol extraction. Reverse transcription was initiated by addition of dNTPs (−dGTP) and either MoMLV-RH− RT (lanes 1 to 4), HIV-1 RT (lanes 5 to 8), AMV RT (lanes 9 to 12) or no RT enzyme (lanes 13 to 16). The position and length (in nucleotides) of the tRNA– cDNA products are marked.
a conformational change in the tRNA–vRNA complex during assembly and maturation of the virion particle, such that an efficient initiation complex is formed. We therefore analyzed whether HIV-1 RT can trigger such a conformational switch of tRNA primers in a pre-incubation in vitro. Primer–template complexes with the non-self primer tRNATrp and self primer tRNALys3 were pre-incubated with HIV-1 RT, and then treated with proteinase K, phenol-extracted and reverse transcribed upon addition of dNTPs (−dGTP) and one of the three RT enzymes (Figure 4). The ability of MoMLV RT to extend both tRNALys3 and tRNATrp efficiently was not influenced by HIV-1 RT pre-incubation (lanes 1 to 4). Most importantly, pre-incubation did not stimulate the tRNATrp priming activity in combination with HIV-1 RT (lanes 5 to 8). Pre-incubation of the tRNATrp and tRNALys3 samples with HIV-1 RT did not change the priming pattern with AMV RT either (lanes 9 to 12). Thus, the efficiency of priming with tRNALys3 by the heterologous AMV RT enzyme was not stimulated by pre-incubation with the HIV-1 enzyme. To check that the cDNAs were not synthesized by HIV-1 RT remaining from the pre-incubation, we also performed the second reaction without adding RT enzyme. No cDNA products were synthesized in this control (lanes 15 and 16), demonstrating that all enzyme was effectively removed after the preincubation step. The standard RT reaction of Figure 3 was performed for 25 minutes. In order to exclude the possibility that the priming efficiencies measured for the different tRNA–RT combinations reflect differences in kinetics of the initiation reaction, we quantitated cDNA products at different time intervals (Figure 5, 0 to 150 minutes). Similar kinetics of cDNA priming were measured for all RT-tRNA primer combinations. Interestingly, all tRNA-primed reactions were relatively slow when compared with a DNA oligonucleotide-primed reaction.
Specific tRNA priming is lost upon extension of the tRNA molecule by two deoxyribonucleotides The observation that HIV-1 RT efficiently primes reverse transcription from its own tRNALys3 primer annealed to in vitro synthesized transcripts (Figures 3 to 5) is inconsistent with the results obtained with virion-extracted template–primers (Figures 1 and 2). A hypothetical difference between the two systems is the form of the tRNA primers. It is possible that the genome-associated tRNA in virions is extended by deoxyribonucleotides if reverse transcription is initiated early in newly assembled virions. Because the 3'end of such an extended tRNA–cDNA molecule may resemble a DNA primer, efficient priming with all tRNAs is anticipated. This can explain the apparent discrepancy between the two experimental systems. In order to study the priming properties of such an extended tRNA–cDNA molecule, we first generated such hybrids by partial extension of tRNALys3 and tRNATrp by means of reverse transcription on the HIV-1 template in the absence of dGTP (Figure 6A). This first −dGTP step was performed with HIV-1 RT or MoMLV RT enzyme. If priming occurs, two deoxyribonucleotides are incorporated to yield extended primers of 78 and 77 nucleotides for tRNALys3 and tRNATrp, respectively. Consistent with the previous assays, MoMLV was able to extend efficiently both tRNALys3 and tRNATrp primers (Figure 6B, lanes 1 and 2), whereas HIV-1 RT did preferentially extend the self tRNALys3 molecule (compare lanes 3 and 4). These extended tRNA samples were used as primer in a second reverse transcription reaction. The length of the pre-extended, radiolabeled tRNAs was verified by gel analysis (Figure 6B, lanes 1 to 4). Another tRNA sample was deproteinized by proteinase K and phenol extraction, and then further extended in a second reverse transcription step (−dATP, lanes 5 to 12). This
248
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
obtained with virion-derived vRNA–tRNA molecules (Figure 1), suggesting that tRNA primers within the virion are in the extended form. Obviously, if tRNA priming and 32P labeling were inefficient in the first reaction, as with tRNATrp in combination with HIV-1 RT (lane 4), a second incubation in the absence of radioactive nucleotides with either HIV-1 RT or MoMLV RT (lanes 8 and 12, respectively) will not yield a labeled cDNA product. tRNA primer extension is already initiated within extracellular virions
Figure 5. Kinetics of self and non-self tRNA priming with HIV-1, AMV and MoMLV-RH− RT enzymes. The reaction conditions are as described in the legend to Figure 3. cDNA production was measured at different time intervals (0, 4, 15, 40 and 150 minutes) and quantitated with a Phosphor Imager. A DNA primer extension was performed with the CN1 oligonucleotide (broken line). The level of reverse transcription is presented in counts ( × 104 ) for two of the DNA-primed reactions (with the HIV-1 and MoMLV-RH− RT enzymes, both produce significantly higher cDNA levels than the corresponding tRNA-primed assays).
second reaction was performed with either HIV-1 RT (Figure 6B, lanes 5 to 8) or MoMLV RT (lanes 9 to 12). A mock-incubation was included to demonstrate the complete removal of the RT enzyme that was present in the pre-extension reaction (lanes 13 to 16). Extension in the second step will produce cDNA lengths of 81 and 80 nucleotides, respectively (Figure 6A). The tRNALys3 and tRNATrp primers that had been extended first by MoMLV RT were efficiently extended by HIV-RT in the second reaction (lanes 5 and 6). The ability of HIV-1 RT to prime reverse transcription with extended forms of both the self and non-self tRNA primers resembles the activity of the MoMLV enzyme with either the extended or native tRNAs (lanes 9 and 10 and 1 and 2, respectively). The apparent loss of priming specificity of the HIV-1 RT enzyme upon pre-extension of the tRNA by two nucleotides is fully consistent with the results
The loss of priming specificity of HIV-1 RT upon extension of the tRNA primer by two deoxyribonucleotides can potentially explain the lack of primer specificity that was observed with virionderived template–primers if this tRNA is extended in the virion. Obviously, virus-derived tRNAprimers with a variable 3' extension will produce cDNA products of different lengths in a −dNTP reverse transcription assay. However, these differentially extended tRNALys3 primers will produce a single strong-stop cDNA product of unit-length in the all dNTP reaction. Indeed, we did consistently obtain superior results with the all dNTP assay compared with the −dNTP assay in experiments with virion-extracted material. This effect is not simply caused by higher sensitivity of the former assay because the two assays performed satisfactory with templates assembled in vitro. To corroborate this effect further, both assays were compared with the virus-derived template– primer and in vitro assembled duplex. In addition, this comparison was performed with HIV-1 and HIV-2 templates (Figure 7). We constructed a T7 plasmid that synthesizes HIV-2 transcripts with an authentic 5' end. The HIV-2 virion-extracted and in vitro-synthesized templates were tested in reverse transcription with HIV-1 RT in the presence of all dNTPs (Figure 7A, lanes 4 and 8) versus a series of three reactions lacking an individual dNTP (lanes 1 to 3 and 5 to 7). Whereas reverse transcription from the calf liver tRNALys3 primer did produce both early-stop products (lanes 6 and 7) and strong-stop cDNA (lane 8), a noticeable decrease in the ratio of early-stop to strong-stop products was measured for the viral primer-template (lanes 2 to 4). Quantitation of the different cDNA products suggests that the majority (>95%) of the tRNALys3 primer in HIV-2 particles is extended, thus confirming the hypothesis. A similar difference was observed with virion-extracted and in vitro assembled HIV-1 templates (Figure 7B). There is some additional evidence for the hypothesis that the majority of tRNALys3 in virions is in the extended form. The all dNTP reaction demonstrated a particular difference between the virion-derived template–primer complexes and the in vitro assembled complexes. In the latter case, dramatic stop products were observed upon incorporation of approximately +5/ + 7 nucleotides
249
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
A Figure 6. HIV-1 RT efficiently primes non-self tRNAs that are extended by two deoxyribonucleotides. A, Diagram of the two-step reverse transcription reaction initiated from tRNALys3 or tRNATrp annealed to the corresponding PBS sites of HIV-1 transcripts. The first step, in the absence of dGTP, leads to incorporation of two nucleotides (indicated by a black box). The resulting tRNA–cDNA length is summarized on the right-hand side. Three additional nucleotides can be incorporated in the second step in the absence of dATP. B, Extended forms of tRNALys3 and tRNATrp were generated in the first −dGTP reaction with HIV-1 or MoMLV-RH− RT (indicated at the top of the panel). A sample of the reaction products was analyzed directly on the gel (lanes 1 to 4), the remainder was deproteinized and reverse transcribed in a second −dATP reaction with HIV-1 RT (lanes 5 to 8), MoMLVRH− RT (lanes 9 to 12) or no RT enzyme (lanes 13 to 16). Radiolabeled dCTP was present only during the first reaction. The positions of the first and second step reaction products are indicated (77, 78, 80 and 81 nucleotides, see the diagram in A).
B
on the HIV-2 template and approximately +1/ + 3 nucleotides on the HIV-1 template (lanes 8 in Figure 7A and B, respectively). Recent studies suggest that these signals represent pause sites induced by a rate-limiting transition from initiation to elongation of reverse transcription (Isel et al., 1996). Other studies reported a similar switch to a processive mode of synthesis after a few distributive nucleotide incorporations have occurred (Huber et al., 1989; Reardon & Miller, 1990). The absence of such stop signals with the virusextracted template–primer complex confirms the idea that the tRNALys3 primer in association with HIV-2 RNA was already extended by at least seven deoxyribonucleotides. To date only indirect evidence for early tRNA extension in virions has been provided. We therefore considered alternative approaches to verify directly the length of the virus-extracted tRNA species. Total RNA isolated from HIV particles can be labeled with [g-32P]ATP and bacteriophage T4 DNA kinase before gel analysis. However, the genome-associated primer is a minority of the total number of tRNALys3 molecules present in HIV-1 particles (Jiang et al., 1993). Another complicating factor is that no discrete tRNA signal will be produced on a denaturing gel because of the variable length of the cDNA
extension. As an alternative approach, we tried to remove the putative deoxyribonucleotide extension from the tRNALys3 primer by DNase treatment of the virion-derived tRNA–cDNA hybrid. Subsequent extension of this unit-length tRNA should increase the cDNA signal in the −dNTP assay, but we did not observe such an effect (data not shown). We believe that this approach failed because the RNA template is degraded by RNase H during the first round of cDNA synthesis.
Discussion Our results show that HIV-1 RT forms a specific complex with the tRNALys3 primer and HIV-1 vRNA that allows efficient initiation of reverse transcription when compared with non-self primer–template combinations. This study further extends the elaborate relation between a retroviral RT enzyme and the tRNA primer. The combined results of multiple biochemical and virological studies indicate that specific RT–tRNA recognition is important at several steps in the viral life cycle: (1) selective binding/packaging of the primer tRNA (Barat et al., 1991; Oude Essink et al., 1995; Das et al., 1995; Mak et al., 1994; Mishima & Steitz, 1995); (2) melting of the tRNA structure and annealing to the vRNA (Sarih-Cottin et al., 1992;
250
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
Wohrl et al., 1993; Oude Essink et al., 1995); (3) efficient initiation of reverse transcription (Li et al., 1994; Arts et al., 1994a, 1996a; Isel et al., 1996; and this paper). What is the biological significance of having multiple levels of specificity? First, it is important to realize that packaging of tRNAs in virion particles is not fully selective because a number of non-self tRNAs are present in virions (Jiang et al., 1993). Thus, priming specificity may
restrict the chances for aberrant reverse transcription from these non-self tRNAs, which may anneal to partially complementary vRNA sequences. Also, selective priming may form an additional safeguard against reverse transcription of fortuitously packaged cellular transcripts (Hu & Temin, 1990). Although no preferential reverse transcription was observed for the AMV and MoMLV RT enzymes in combination with their self tRNA primers in our
B
A
Figure 7. Early-stop versus strong-stop cDNA synthesis on virion-extracted and in vitro assembled tRNA–vRNA duplexes. A, The tRNALys3–vRNA template was purified from HIV-2 particles (left) or assembled from an in vitro synthesized HIV-2 RNA and calf liver tRNA (right). Reverse transcription was performed with the HIV-1 RT enzyme in −dNTP assays (lanes 1 to 3 and 5 to 7) or the all dNTP assay (lanes 4 and 8). The HIV-2 template sequence upstream of the PBS reads 5'-GCAGGU-3'. As a consequence, the −dATP reaction will not produce any cDNA because the first nucleotide is missing. The −dTTP and −dGTP reactions produce extended tRNALys3 molecules of 79 and 80 nucleotides, and the HIV-2 strong-stop product is 379 nucleotides in length (positions indicated between the panels). B, The tRNALys3–vRNA template was purified from HIV-1 particles (left) or assembled from an in vitro synthesized HIV-1 RNA and calf liver tRNA (right). Reverse transcription was performed with the HIV-1 RT enzyme in −dNTP assays (lanes 1 to 3 and 5 to 7) or the all dNTP assay (lanes 4 and 8). The HIV-1 template sequence upstream of the PBS reads 5'-UAGCAG-3'. Thus, a radiolabeled dCTP will be incorporated first, and the −dATP, −dTTP and −dGTP reactions produce extended tRNALys3 molecules of 81, 77 and 78 nucleotides, respectively. The HIV-1 strong-stop cDNA is 257 nucleotides in length (lane 4). A similar discrete cDNA product is not observed with the in vitro synthesized HIV-1 template (lane 8) because this T7 transcript contains a 5' extension (the complete U3 region, see Materials and Methods) with an expected strong-stop cDNA of 886 nucleotides.
251
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
assays with HIV-1 templates, it is important to realize that the putative additional tRNA–vRNA interactions cannot take place. It would be interesting to test the specificity of these RT enzymes in combination with their cognate RNA templates. Because the HIV-1 RT enzyme is strongly committed to the tRNALys3 molecule, it seems rather difficult to change the identity of the tRNA primer for HIV-1 reverse transcription. Consistent with this idea, HIV-1 constructs with a different PBS specificity are unstable and revert to the wt PBSLys3 sequence (Li et al., 1994; Wakefield et al., 1994, 1995; Das et al., 1995). Interestingly, a recent study (Wakefield et al., 1996) suggests that a PBSHis mutant of HIV-1 can be stabilized by a second mutation in the vRNA template in the region implicated in the additional interaction with the tRNA anticodon loop (Isel et al., 1995). Delayed replication was initially measured for this variant compared with wild-type HIV-1. However, by several passages the mutant viruses demonstrated replication kinetics similar to those of wt virus (Wakefield et al., 1996). A DNA sequence analysis of the PBS region did not reveal mutations that explain the improved viral replication. It would be interesting to analyze the RT gene of this improved PBSHis variant to check for potential second-site reversions within the tRNA binding domain of RT. This genetic approach can provide independent evidence for the specificity of RT–tRNA interactions and can be helpful in dissecting the RT domains that are involved in primer binding. In this study, we were unable to detect tRNALys3 extension products in the −dNTP assay with virion-extracted template–primer (Figure 7), but strong-stop cDNA was efficiently synthesized. These results suggest that the majority of genomeassociated tRNALys3 in virions is in the extended conformation. Consistent with this finding, sensitive PCR techniques revealed that short minusstrand cDNAs are present at a low frequency within cell-free virus particles (Trono, 1992; Yu et al., 1996; Lori et al., 1992; Zhang et al., 1993). HIV-1 particles were demonstrated to contain about one copy of strong-stop minus strand DNA per 100 to 1000 viral particles and only one full-length DNA was detected per 105 virions (Trono, 1992; Lori et al., 1992; Arts et al., 1994b). If we combine these results, the amount of cDNA seems to decrease progressively as a function of the distance from the priming position. Because the majority of HIV-1 and HIV-2 virions contain partially extended tRNAs, any arguments about the specificity of RT-tRNA interactions with virion-derived material should be considered with caution. Partial tRNA extension within virion particles may also explain a remarkable observation on reverse transcription in AZT-treated cells. It was demonstrated that this chain terminator strongly inhibits reverse transcription, but the level of strong-stop cDNA synthesis was found to be relatively unaffected by AZT (Zack et al., 1990; Arts
& Wainberg, 1994; Arts et al., 1996b). Similar results were recently described for the non-nucleoside RT inhibitor nevirapine (Zhang et al., 1996). It was suggested that the conformation of the RT enzyme may be slightly different during strong-stop cDNA synthesis than during the rest of viral DNA synthesis, thus explaining a differential effect of RT drugs. Here, we propose an alternative and relatively simple scenario to explain these results. If strong-stop cDNA synthesis was largely completed before infection of the AZT-treated cells, this initial reverse transcription step will be resistant to all RT inhibitors. We should also emphasize that differential drug sensitivity of the various reverse transcription steps will not occur in continuously treated cells. Here, both HIV-infected and HIV-producing cells will be affected equally by the drug, such that all reverse transcription steps will be inhibited in a similar manner. Extension of the tRNA primer in extracellular virions may be restricted by the concentration of dNTP molecules present in the virus particles. We obtained evidence for tRNA extension in virions of HIV-1 and HIV-2 produced either by transiently transfected HeLa cells or by infected SupT1 T cell cultures, both being transformed cell types. However, primary cells contain much lower dNTP levels (Meyerhans et al., 1994; O’Brien et al., 1994; Back et al., 1996) and unextended tRNA may be the major form of the primer in virions produced by these cells. In contrast, the replication strategy of the human foamy virus (HFV) allows this retrovirus to produce full-length DNA genomes in about every six to nine virus particles (Yu et al., 1996), which suggests that reverse transcription is largely completed within the producer cells before HFV virion release. Although there is convincing evidence that the protein involved in selective tRNA incorporation is the Gag-Pol precursor protein (Mak et al., 1994), it is not known whether this precursor is proteolytically processed at the moment of tRNA-primed initiation of reverse transcription. Reverse transcription is traditionally considered to occur in the cytoplasm of infected cells (Varmus & Swanstrom, 1984). At this relatively ‘‘late’’ point in time, the RT domain will be fully processed into the mature heterodimer. If the initial tRNA-priming reaction occurs relatively ‘‘early’’ in nascent virions, it remains possible that the Gag-Pol polyprotein is the actual priming enzyme. Consistent with this idea, the purified Gag-Pol precursor can efficiently reverse-transcribe the viral genome in vitro (Kaplan et al., 1994).
Materials and Methods Construction of DNA plasmids and synthesis of RNA templates The infectious molecular clones HIV-2 pROD10 (Peden & Martin, 1996) and HIV-1 pLAI (Peden et al., 1991) were kindly provided by Dr K. Peden. Construction of the
252 PBS-mutated HIV-1 plasmids has been described (Das et al., 1995). In order to produce PBS-mutated HIV-1 transcripts in vitro, we used a plasmid containing the complete 5'LTR-gag region cloned into Bluescript KS + . This plasmid Blue-5'LTR has been described (Berkhout & Klaver, 1993). This plasmid was linearized with XhoI (in the polylinker downstream of gag) and transcribed by T7 RNA polymerase according to a standard method (Berkhout et al., 1993). This transcript contains the complete HIV-1 LTR sequences (including U3) and terminates within the gag gene (position +830 of the LAI sequence, relative to the transcriptional start site at +1). An HIV-2 plasmid that initiates T7 transcription precisely at the +1 position (the U3-R border) of wild-type HIV-2 RNA was constructed as follows. A PCR-amplified HIV-2 DNA fragment containing the complete leader region and part of the gag gene (coordinates +1/ + 892 of the ROD sequence) was cloned as BamHI-EcoRI fragment in plasmid pUC8. The PCR primers used did encode the upstream BamHI site and T7 RNA polymerase promoter sequence and the downstream EcoRI restriction site to allow run-off transcription. Oligodeoxyribonucleotides 5'-CAT GGA TCC TAA TAC GAC TCA CTA TAG GGT CGC TCT GCG GAG AGG-3' (BamHI site underlined, T7 promoter in bold) and 5'-CAT GAA TTC CTG CAG TTC CTG TTT CTG-3' (EcoRI site underlined) were used in a standard PCR reaction (five minutes, 95°C; 35 cycles of one minute, 95°C; one minute, 55°C; and two minutes, 72°C; and then ten minutes, 72°C; and ten minutes, 4°C) with 100 ng of both oligonucleotides and 1 ng of the HIV-2 pBluescript GAG2+ plasmid (Berkhout et al., 1993). Isolation of virion RNA templates The wild-type HIV-1 and HIV-2 virion particles, as well as the PBS-mutated HIV-1 variants, were produced by transiently transfected HeLa cells (Figures 1 and 2). The protocols for transfection, virion purification and vRNA isolation by proteinase K treatment and phenol extraction have been described in detail (Das et al., 1995). The wild-type HIV-1 and HIV-2 virions were also produced by infection of the SupT1 T cell line (Figure 7). Reverse transcription reactions with the virion-extracted vRNA–tRNA complexes was performed according to the all dNTP assay or the −dNTP assay as described below. Reverse transcription assays vRNA–tRNA duplexes were either assembled in vitro or purified from virus particles. In vitro assembly was performed with approximately 10 ng in vitro synthesized vRNA template (0.003 pmol) and 1.5 mg calf liver tRNA (6 pmol total tRNA, Boehringer Mannheim). These two RNA samples were annealed in 12 ml annealing buffer (83 mM Tris-HCl (pH 7.5), 125 mM KCl) at 85°C for two minutes, at 65°C for ten minutes, and then cooled to room temperature in approximately 30 minutes. Control DNA oligonucleotide primers (20 ng) were annealed in an identical manner to both the extracted and in vitro synthesized vRNA templates. The tRNA/DNA primer was extended by addition of either all dNTPs or a mixture lacking one dNTP. In the −dNTP assay (e.g. −dGTP) we added 6 ml 3 × RT(−dGTP) buffer (9 mM MgCl2 , 30 mM dithiothreitol, 150 mg actinomycin D per ml , 30 mM dATP, 30 mM dTTP, 1.5 mM dCTP), 0.3 ml of [a-32P]dCTP (800 Ci/mmol, 10 mCi/ml) and RT enzyme (approximately 0.5 pmol).
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
The standard incubation was for 25 minutes at 42°C with 2.5 units of HIV-1 p66/p66 RT (final concentration 22 nM), 100 units of MoMLV RT (95 nM, Life Technologies), 100 units of MoMLV RNaseH− RT (137 nM, Promega), 12.5 units of AMV RT (40 nM, Boehringer). The units were as defined by the suppliers, based on different assay systems. In the all dNTP assay, the annealed primer (tRNA or DNA primer) was extended for three minute at 42°C by addition of 3 × RT buffer (see above, with 30 mM dATP/dTTP/dGTP and 1.5 mM dCTP), radiolabeled dCTP and RT enzyme. After addition of 1 ml dNTP mix (10 mM each dNTP), incubation was continued at 42°C for 25 minutes. The cDNA products of the −dNTP assay and the all dNTP assay were precipitated in 25 mM EDTA, 0.3 M sodium acetate (pH 5.2), 70% ethanol at −20°C. The pellets were resuspended in formamide loading buffer, heated at 85°C for three minutes, and analyzed on a denaturing 6% polyacrylamide/urea sequencing gel. Several variations on the basic reverse transcription assay were used. In order to measure the kinetics of reverse transcription, we used increased reaction volumes and took samples over a period of time. In other experiments, the vRNA–tRNA complex was pre-incubated with HIV-1 RT (30 minutes at 42°C with 2.5 units of HIV-1 RT in annealing buffer). HIV-1 RT was removed by addition of proteinase K (final concentration 100 mg/ml) and EDTA (final concentration 5 mM). After 30 minutes of incubation at 37°C proteinase K was removed by phenol extraction (1 × phenol, 1 × phenol/ chloroform/isoamylalcohol, 25:24:1, by vol.). The RNA was ethanol-precipitated and resolved in 12 ml annealing buffer. Extension of the primer and analysis of the cDNA product is described above. The two-step reverse transcription protocol (Figure 6) consists of a first reaction without dGTP and a second reverse transcription without dATP. The radiolabeled dCTP was present only during the first reaction step. The RT enzyme was removed in between the two assays by proteinase K/phenol extraction. The sample was ethanol-precipitated, resuspended in 12 ml annealing buffer and cDNA reinitiated by addition of 6 ml 3 × RT buffer containing 9 mM MgCl2 , 30 mM dithiothreitol, 150 mg actinomycin D per ml, 30 mM dTTP, 30 mM dGTP and 30 mM dCTP, and RT as indicated on top of the panel. After incubation of 25 minutes at 42°C, the sample was precipitated and analyzed as described above.
Acknowledgements We thank Dr D. Stammers for the gift of purified HIV-1 RT enzyme (obtained through the MRC AIDS Reagent Project), Dr K. Peden for the HIV-1 pLAI and HIV-2 ROD infectious clones, and Wim van Est for excellent photography. This study was supported in part by the Netherlands Organization for Scientific Research (NWO) and the Dutch Cancer Society (KWF).
References Aiyar, A., Cobrinik, D., Ge, Z., Kung, H. J. & Leis, J. (1992). Interaction between retroviral U5 RNA and the TYC loop of the tRNA(Trp) primer is required for efficient initiation of reverse transcription. J. Virol. 66, 2464–2472.
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
Aiyar, A., Ge, Z. & Leis, J. (1994). A specific orientation of RNA secondary structures is required for initiation of reverse transcription. J. Virol. 68, 611–618. Arts, E. J. & Wainberg, M. A. (1994). Preferential incorporation of nucleoside analogs after template switching during human immunodeficiency virus reverse transcription. Antimicrob. Agents Chemother. 38, 1008–1016. Arts, E. J., Li, X., Gu, Z., Kleiman, L., Parniak, M. A. & Wainberg, M. A. (1994a). Comparison of deoxyoligonucleotide and tRNA(Lys-3) as primers in an endogenous human immunodeficiency virus-1 in vitro reverse transcription/template-switching reaction. J. Biol. Chem. 269, 14672–14680. Arts, E. J., Mak, J., Kleiman, L. & Wainberg, M. A. (1994b). DNA found in human immunodeficiency virus type 1 particles may not be required for infectivity. J. Gen. Virol. 75, 1605–1613. Arts, E. J., Ghosh, M., Jacques, P. S., Ehresmann, B. & Le Grice, S. F. J. (1996a). Restoration of tRNALys,3primed (−)-strand DNA synthesis to an HIV-1 reverse transcriptase mutant with extended tRNAs. J. Biol. Chem. 271, 9054–9061. Arts, E. J., Marois, J.-P., Gu, Z., Le Grice, S. F. J. & Wainberg, M. A. (1996b). Effects of 3'-deoxynucleoside 5'-triphosphate concentrations on chain termination by nucleoside analogs during human immunodeficiency virus type 1 reverse transcription. J. Virol. 70, 712–720. Back, N. K. T., Nijhuis, M., Keulen, W., Boucher, C. A. B., Oude Essink, B. B., van Kuilenburg, A. B. P., Van Gennip, A. H. & Berkhout, B. (1996). Reduced replication of 3TC-resistant HIV-1 variants in primary cells due to a processivity defect of the reverse transcriptase enzyme. EMBO J. 15. 4040–4049. Baltimore, D. (1970). Viral RNA-dependent DNA polymerase RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature, 226, 1209-1211. Barat, C., Lullien, V., Schatz, O., Keith, G., Mugeyre, M. T., Gru¨ninger-Leitch, F., Barre´-Sinoussi, F., LeGrice, S. F. J. & Darlix, J. L. (1989). HIV-1 reverse transcriptase specifically interacts with the anticodon domain of its cognate primer tRNA. EMBO J. 11, 3279–3285. Barat, C., Le Grice, S. F. & Darlix, J. L. (1991). Interaction of HIV-1 reverse transcriptase with a synthetic form of its replication primer, tRNA(Lys,3). Nucl. Acids Res. 19, 751–757. Barat, C., Schatz, O., Le Grice, S. & Darlix, J. L. (1993). Analysis of the interactions of HIV1 replication primer tRNA(Lys, 3) with nucleocapsid protein and reverse transcriptase. J. Mol. Biol. 231, 185–190. Berkhout, B. (1996). Structure and function of the human immunodeficiency virus leader RNA. In Progress in Nucleic Acid Research and Molecular Biology (Cohn, W. E. & Moldave, K., eds), vol. 54, 1–34. Academic Press, Inc. San Diego. Berkhout, B. & Klaver, B. (1993). In vivo selection of randomly mutated retroviral genomes. Nucl. Acids Res. 21, 5020–5024. Berkhout, B., Oude Essink, B. B. & Schoneveld, I. (1993). In vitro dimerization of HIV-2 leader RNA in the absence of PuGGAPuA motifs. FASEB J. 7, 181–187. Berkhout, B. & Schoneveld, I. (1993). Secondary structure of the HIV-2 leader RNA comprising the tRNAprimer binding site. Nucl. Acids Res. 21, 1171–1178. Cobrinik, D., Aiyar, A., Ge, Z., Katzman, M., Huang, H. & Leis, J. (1991). Overlapping retrovirus U5 sequence
253 elements are required for efficient integration and initiation of reverse transcription. Structure of the retroviral genome. J. Virol. 65, 3864–3872. Coffin, J. (1984). In RNA Tumor Viruses (Weiss, R., Teich, N., Varmus, H. & Coffin, J., eds), 4th edit., pp. 261–368, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Das, A. T., Klaver, B. & Berkhout, B. (1995). Reduced replication of human immunodeficiency virus type 1 mutants that use reverse transcription primers other than the natural tRNA3Lys. J. Virol. 69, 3090–3097. De Rocquigny, H., Gabus, C., Vincent, A., FournieZaluski, M.-C., Roques, B. & Darlix, J.-L. (1992). Viral RNA annealing activities of human immunodeficiency virus type 1 nucleocapsid protein require only peptide domains outside the zinc fingers. Proc. Natl Acad. Sci. USA, 89, 6472–6476. Hu, W.-S. & Temin, H. M. (1990). Genetic consequences of packaging two RNA genomes in one retroviral particle: pseudodiploidy and high rate of genetic recombination. Proc. Natl Acad. Sci. USA, 87, 1556–1560. Huber, H. E., McCoy, J. M., Seehra, J. S. & Richardson, C. C. (1989). Human immunodeficiency virus 1 reverse transcriptase; Template binding, processivity, strand displacement synthesis & template switching. J. Biol. Chem. 264, 4669–4678. Isel, C., Marquet, R., Keith, G., Ehresmann, C. & Ehresmann, B. (1993). Modified nucleotides of tRNALys3 modulate primer/template loop-loop interaction in the initiation complex of HIV-1 reverse transcription. J. Biol. Chem. 268, 25269–25272. Isel, C., Ehresmann, C., Keith, G., Ehresmann, B. & Marquet, R. (1995). Initiation of reverse transcription of HIV-1: secondary structure of the HIV-1 RNA/tRNA3Lys (template/primer). J. Mol. Biol. 247, 236–250. Isel, C., Lanchy, J.-M., Le Grice, S. F. J., Ehresmann, C., Ehresmann, B. & Marquet, R. (1996). Specific initiation and switch to elongation of human immunodeficiency virus type 1 reverse transcriptase require the post-transcriptional modifications of primer tRNALys3. EMBO J. 15, 917–924. Jiang, M., Mak, J., Ladha, A., Cohen, E., Klein, M., Rovinski, B. & Kleiman, L. (1993). Identification of tRNAs incorporated into wild-type and mutant human immunodeficiency virus type 1. J. Virol. 67, 3246–3253. Kaplan, A. H., Krogstad, P., Kempf, D. J., Norbeck, D. W. & Swanstrom, R. (1994). Human immunodeficiency virus type 1 virions composed of unprocessed gag and gag-pol precursors are capable of reverse transcribing viral genomic RNA. Antimicrob. Agents Chemother. 12, 2929–2933. Khan, R. & Giedroc, D. P. (1992). Recombinant human immunodeficiency virus type 1 nucleocapsid (NCp7) protein unwinds tRNA. J. Biol. Chem. 267, 6689–6695. Lapadat-Tapolsky, M., Pernelle, C., Borie, C. & Darlix, J.-L. (1995). Analysis of the nucleic acid annealing activities of nucleocapsid protein from HIV-1. Nucl. Acids Res. 23, 2434–2441. Leis, J., Aiyar, A. & Cobrinik, D. (1993). Regulation of initiation of reverse transcription of retroviruses. In Reverse Transcriptase (Skalka, A. M. & Goff, S. P., eds), pp. 33–48, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Li, X., Mak, J., Arts, E. J., Gu, Z., Kleiman, L., Wainberg, M. A. & Parniak, M. A. (1994). Effects of alterations of primer-binding site sequences on human im-
254
Specific tRNALys3 Priming by HIV-1 Reverse Transcriptase
munodeficiency virus type 1 replication. J. Virol. 68, 6198–6206. Lori, F., Veronese, F. D. M., De Vico, A. L., Lusso, P., Reitz, M. S., Jr & Gallo, R. C. (1992). Viral DNA carried by human immunodeficiency virus type 1 virions. J. Virol. 66, 5067–5074. Mak, J., Jiang, M., Wainberg, M. A., Hammarskjold, M. L., Rekosh, D. & Kleiman, L. (1994). Role of Pr160gagpol in mediating the selective incorporation of tRNA(Lys) into human immunodeficiency virus type 1 particles. J. Virol. 68, 2065–2072. Mely, Y., De Rocquigny, H., Sorinas-Jimeno, M., Keith, G., Roques, B. P., Marquet, R. & Gerard, D. (1995). Binding of the HIV-1 nucleocapsid protein to the primer tRNALys3, in vitro, is essentially not specific. J. Biol. Chem. 270, 1650–1656. Meyerhans, A., Vartanian, J.-P., Hultgren, C., Plikat, U., Karlsson, A., Wang, L., Eriksson, S. & Wain-Hobson, S. (1994). Restriction and enhancement of human immunodeficiency virus type 1 replication by modulation of intracellular deoxynucleoside triphosphate pools. J. Virol. 68, 535–540. Mishima, Y. & Steitz, J. A. (1995). Site-specific crosslinking of 4-thiouridine-modified human tRNALys3 to reverse transcriptase from human immunodeficiency virus type 1. EMBO J. 14, 2679–2687. O’Brien, W. A., Namazi, A., Kalhor, H., Mao, S.-H., Zack, J. A. & Chen, I. S. Y. (1994). Kinetics of human immunodeficiency virus type 1 reverse transcription in blood mononuclear phagocytes are slowed by limitations of nucleotide precursors. J. Virol. 68, 1258–1263. Oude Essink, B. B., Das, A. T. & Berkhout, B. (1995). Structural requirements for the binding of tRNA Lys3 to reverse transcriptase of the human immunodeficiency virus type 1. J. Biol. Chem. 270, 23867–23874. Peden, K. W. C. & Martin, M. A. (1996). Virological and molecular genetic techniques for studies of established H2V isolates. In HIV, a Practical Approach (Karn, J., ed.), pp. 21–25, IRL Press at Oxford University Press, Oxford, New York and Tokyo. Peden, K., Emerman, M. & Montagnier, L. (1991). Changes in growth properties on passage in tissue culture of viruses derived from infectious molecular clones of HIV-1LAI , HIV-1MAL & HIV-1ELI . Virology, 185, 661–672. Reardon, J. E. & Miller, W. H. (1990). Human immunodeficiency virus reverse transcriptase. J. Biol. Chem. 265, 20302–20307. Richter-Cook, N. J., Howard, K. J., Cirino, N. M., Wohrl, B. M. & Le Grice, S. F. (1992). Interaction of tRNA(Lys-3) with multiple forms of human immunodeficiency virus reverse transcriptase. J. Biol. Chem. 267, 15952–15957. Sarih-Cottin, L., Bordier, B., Musier-Forsyth, K. Andreola, M. L., Barr, P. J. & Litvak, S. (1992). Preferential
interaction of human immunodeficiency virus reverse transcriptase with two regions of primer tRNA(Lys) as evidenced by footprinting studies and inhibition with synthetic oligoribonucleotides. J. Mol. Biol. 226, 1–6. Temin, H. M. & Mizutani, S. (1970). Viral RNAdependent DNA polymerase in virions of rous sarcoma virus. Nature, 226, 1211–1213. Trono, D. (1992). Partial reverse transcripts in virions from human immunodeficiency and murine leukemia viruses. J. Virol. 66, 4893–4900. Varmus, H. & Swanstrom, R. (1984). Replication of retroviruses. In RNA Tumor Viruses (Weiss, R., Teich, N., Varmus, H. & Coffin, J., eds), 2nd edit., pp. 369–512, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Wakefield, J. K., Rhim, H. & Morrow, C. D. (1994). Minimal sequence requirements of a functional human immunodeficiency virus type 1 primer binding site. J. Virol. 68, 1605–1614. Wakefield, J. K., Wolf, A. G. & Morrow, C. D. (1995). Human immunodeficiency virus type 1 can use different tRNAs as primers for reverse transcription but selectively maintains a primer binding site complementary to tRNALys3. J. Virol. 69, 6021– 6029. Wakefield, J. K., Kang, S.-M. & Morrow, C. D. (1996). Construction of a type 1 human immunodeficiency virus that maintains a primer binding site complementary to tRNAHis. J. Virol. 70, 966– 975. Wohrl, B. M., Ehresmann, B., Keith, G. & Le Grice, S. F. (1993). Nuclease footprinting of human immunodeficiency virus reverse transcriptase/ tRNA(Lys-3) complexes. J. Biol. Chem. 268, 13617– 13624. Yu, S. F., Baldwin, D. N., Gwynn, S. R., Yendapalli, S. & Linial, M. L. (1996). Human foamy virus replication: a pathway distinct from that of retroviruses and hepadnaviruses. Science, 271, 1579–1582. Zack, J. A., Arrigo, S. J., Weitsman, S. R., Go, A. S., Haislip, A. & Chen, I. S. Y. (1990). HIV-1 entry into quiescent primary lymphocytes: molecular analysis reveals a labile, latent viral structure. Cell, 61, 213–222. Zhang, H., Zhang, Y., Spicer, T. P., Abbott, L. Z., Abbott, M. & Poiesz, B. J. (1993). Reverse transcription takes place within extracellular HIV-1 virions: potential biological significance. AIDS Res. Hum. Retroviruses, 9, 1287–1296. Zhang, H., Dornadula, G., Wu, Y., Havlir, D., Richman, D. D. & Pomerantz, R. J. (1996). Kinetic analysis of intravirion reverse transcription in the blood plasma of human immunodeficiency virus type 1-infected individuals: direct assessment of resistance to reverse transcriptase inhibitors in vivo. J. Virol. 70, 628–634.
Edited by J. Karn (Received 2 May 1996; received in revised form 16 July 1996; accepted 5 September 1996)
Lihat lebih banyak...
Comentários