The Native Structure of the Human Immunodeficiency Virus Type 1 RNA Genome Is Required for the First Strand Transfer of Reverse Transcription

Transcription

1 VIROLOGY 249, (1998) ARTICLE NO. VY The Native Structure of the Human Immunodeficiency Virus Type 1 RNA Genome Is Required for the First Strand Transfer of Reverse Transcription Ben Berkhout, 1 Atze T. Das, and Jeroen L. B. van Wamel Department of Human Retrovirology, Academic Medical Center, University of Amsterdam, Meibergdreef 15, P.O. Box 22700, 1100 DE Amsterdam, The Netherlands Received December 8, 1997; returned to author for revision January 20, 1998; accepted July 7, 1998 Retroviral particles contain two genomic RNAs of 9 kb that are linked in a noncovalent manner. In vitro studies with purified transcripts have identified particular RNA motifs that contribute to the RNA-dimerization reaction, but the situation may be more complex within virion particles. In this study, we tested whether the primer-binding site (PBS) of the human immunodeficiency virus type 1 (HIV-1) RNA genome and the associated trna Lys3 primer play a role in the process of RNA dimerization. Deletion of the PBS motif did not preclude the formation of RNA dimers within virus particles, indicating that this motif and the trna primer do not participate in the interactions that control RNA packaging and dimerization. Genome dimerization has been proposed to play a role in particular steps of the reverse transcription mechanism. To test this, reverse transcription was performed with the native RNA dimer and the heat-denatured template. These two template forms yielded equivalent levels of minus-strand strong-stop cdna product, which is an early intermediate of reverse transcription. However, melting of the RNA dimer precluded the next step of reverse transcription, in which the minus-strand strong-stop cdna is translocated from the 5 repeat element to the 3 repeat element. The results suggest that the conformation of the dimeric RNA genome facilitates the first strand-transfer reaction of the reverse transcription mechanism Academic Press INTRODUCTION Retroviral particles contain a dimeric genome consisting of two identical, full-length RNA transcripts that are linked by noncovalent bonds (reviewed in Coffin, 1984). The nature of the RNA RNA interaction is poorly understood. Electron microscopic studies with partially denatured dimers demonstrate typical rabbit ear structures, suggesting that the two RNA monomers are joined in a region near their 5 -ends encompassing the untranslated leader region. The dimerization signal was designated the dimer-linkage structure (DLS). Several mechanisms have been proposed for the dimer interaction between two identical-sense RNA genomes. One intriguing model uses a nucleic acid linker that is basepaired to both copies of the RNA genome. This mechanism was originally proposed for the avian sarcoma leukosis virus, with the trna Trp reverse transcription primer as the linker molecule (Haseltine et al., 1977). The trna primer interacts with an 18-bp complementary sequence to the primer-binding site (PBS), but additional basepairing contacts between other trna domains and the viral RNA have been proposed (Aiyar et al., 1992; Berkhout and Schoneveld, 1993; Isel et al., 1995). Thus, one necessary property of a putative linker molecule, which is to interact 1 To whom reprint requests should be addressed. Fax: (31 20) b.berkhout@amc.uva.nl. with the target RNA through multiple interaction sites, has been met. A dimerization assay has been developed for in vitro synthesized transcripts, in which RNA dimers are formed spontaneously under conditions of high ionic strength, without any protein or trna cofactors (Bieth et al., 1990). In agreement with the rabbit ear model, short transcripts encompassing the untranslated leader region of several retroviral species were found to dimerize (Bieth et al., 1990; Darlix et al., 1990, 1992; Awang and Sen, 1993; Marquet et al., 1991, 1994; Berkhout et al., 1993; Roy et al., 1990; Mougel et al., 1993; Tounekti et al., 1992; Sundquist and Heaphy, 1993; Paoletti et al., 1993). Studies with HIV-1 RNA led to the identification of a smaller dimerinitiation site (DIS), consisting of a self-complementary 6-base sequence located in the loop of a hairpin (Skripkin et al., 1994; Laughrea and Jette, 1996, 1994; Paillart et al., 1993, 1996b; Clever et al., 1996). It has been proposed that basepairing between two loops ( loop-loop kissing ) initiates the RNA dimerization process. However, there is accumulating evidence that additional linkages of unknown nature are present in RNA dimers within virions. For instance, RNA dimers with a normal thermal stability were recovered from DIS-mutated virion particles (Haddrick et al., 1996; Berkhout and van Wamel, 1996). By these experiments, we concluded that the DIS motif is not absolutely required for DNA-dimerization, but it is evident that such DIS mutants do exhibit a replication /98 $25.00 Copyright 1998 by Academic Press All rights of reproduction in any form reserved.

2 212 BERKHOUT, DAS, AND VAN WAMEL defect (Haddrick et al., 1996; Berkhout and van Wamel, 1996; Paillart et al., 1996a). Reduced RNA packaging was measured for DIS mutated genomes (Berkhout and van Wamel, 1996; Paillart et al., 1996a), a result that is consistent with the idea that DIS pairing is an early event during virion assembly and perhaps a prerequisite for encapsidation of the viral RNA (Fu and Rein, 1993; Fu et al., 1994). A reverse transcription defect during the second strand transfer of the plus-strand strong-stop cdna has also been reported for such DIS mutants (Paillart et al., 1996a). Additional experiments are required to further elaborate on the mechanism of retroviral RNA dimerization and its role in viral replication. Here, we report efficient RNA dimerization for a PBS-mutated HIV-1 genome that lacks the associated trna Lys3 primer. Furthermore, a normal thermal stability was measured for this RNA dimer, suggesting that the PBS motif and the associated trna molecule are not involved in the dimerization process. We used the wild-type HIV-1 genome, both the native RNA dimer and the heat-denatured form, to prime reverse transcription in vitro from the associated trna molecule on the addition of reverse transcriptase (RT) and dntps. The early reverse transcription product, minus-strand strong-stop cdna, was synthesized with the same efficiency on both the native and denatured template, indicating that RNA template structure and dimerization state do not influence trna-primed initiation of reverse transcription. In contrast, the denatured RNA was a less efficient template for the synthesis of more extended cdna products. This result suggests that the structure of the genomic RNA dimer is required for the first strand-transfer step, in which the nascent strongstop cdna is translocated from the 5 repeat (R) to the 3 R of the HIV-1 genome. RESULTS The PBS and associated trna primer are not involved in HIV-1 RNA dimerization The RNA genome of HIV-1 virions was isolated by phenol extraction and characterized by nondenaturing Northern blot analysis. Consistent with previous studies, a discrete band was observed for the genomic RNA sample (Fig. 1, lane 1). To demonstrate that this RNA form corresponds to the dimeric complex, we incubated the sample for 10 min at 65 C to denature the RNA. This heat treatment converted the RNA sample into a fastermigrating form corresponding to monomeric RNA (Fig. 1, lane 2). However, melting of the RNA dimer yielded only a small amount of intact monomer, and a smear of smaller RNA species was observed. Such a smear of RNA fragments has been observed previously by several workers in studies on HIV-1 and other retroviruses (Fu et al., 1994; Haddrick et al., 1996; Berkhout and van Wamel, FIG. 1. The trna Lys3 primer and viral PBS sequence do not play a role in RNA dimerization. Electrophoretic analysis of HIV-1 genomic RNAs from wild-type and mutant virions. Viruses were produced in infected cultures of the SupT1 T cell line (lanes 1 and 2) or by transiently transfected HeLa cells (lanes 3 8). Viral RNA was phenol extracted from purified virions and subsequently analyzed on a native 0.9% agarose gel. The SupT1 sample was analyzed in the native conformation, and on denaturation at 65 C, all HeLa cell samples were analyzed under native conditions. The RNA was electroblotted from the gel onto a membrane and visualized with a gag probe. Positions of the dimer (d) and monomer (m) RNA are indicted. The HeLa samples were extracted by a protocol that included Bluescript plasmid as carrier, resulting in two weak signals in lanes 3 8 that represent cross-hybridization with the probe. 1996; Goncalves et al., 1996; Fu and Rein, 1993; Darlix et al., 1992). The leader region of retroviral genomes is implicated in RNA dimerization and also contains the PBS motif that is complementary to the trna primer for reverse transcription. A series of HIV-1 constructs with mutations in the PBS site or the gene encoding RT were used to test the role of both the PBS sequence and the trna primer in the dimerization process. Because some of these HIV-1 mutants are replication impaired, we produced the wild-type and mutant virions by transient transfection of the corresponding molecular clones in HeLa cells. To produce virion particles that lack the trna primer, we first tested an RT-deletion mutant because this protein is responsible for selective encapsidation of the trna Lys3 primer, presumably as part of the Gag Pol polyprotein (Mak et al., 1994; Das and Berkhout, 1995). Because the trna binding domain of RT has not been mapped in detail, we introduced a rather extensive deletion. The wild-type virions contained an intact RNA dimer (Fig. 1, lane 3). In contrast, the RT virions yielded no dimeric or monomeric RNA (Fig. 1, lane 4), and little smeared RNA

3 HIV-1 RNA DIMER AND REVERSE TRANSCRIPTION 213 that the PBS motif and the associated trna Lys3 primer do not participate in the process of HIV-1 RNA packaging and dimerization. FIG. 2. Analysis of the thermal stability of the RNA dimer present in wild-type and PBS mutant virions. Aliquots of the virion-extracted RNA were heated for 10 min at different temperatures (40 C, 43 C, 46 C, 48 C, 50 C, 52 C, 54 C, 57 C, and 60 C). The RNA samples were separated on a native gel, blotted, and analyzed with an HIV-1 gag probe. The dimer signal was quantified with a PhosphorImager to calculate the percentage of remaining dimer. The RNA dimer is required for elongation of reverse transcription We tested the capacity of the dimer versus denatured monomer HIV-1 RNA to act as template in reverse transcription. The wild-type RNA genome was extracted from virion particles and dissociation of the dimer was followed on a nondenaturing Northern blot in the range of 27 C to 87 C (Fig. 3A). These templates were compared in trna-extension assays with exogenously added HIV-1 RT and dntps, including radiolabeled - 32 P-dCTP. Reverse transcription is primed by the genome-associated trna Lys3 primer and yields a strong-stop minus-strand cdna product of 257 nucleotides on denaturing gel electrophoresis (Fig. 3B). Quantification of the results indicated that the HIV-1 dimer genome was converted completely to the monomeric form on incubation at 57 C, but this modification did not affect the trna-extension efficiency of the template. Thus, initiation of reverse tran- was observed. Two minor signals below the monomer position represent cross-hybridization of the probe to a carrier plasmid that was added during RNA extraction. Additional experiments revealed incomplete processing of the structural Gag protein in this RT mutant (results not shown), suggesting that the protease activity of the mutant Gag Pol precursor protein was affected, thus producing unwanted side effects on the process of virion assembly/maturation and causing instability of the packaged RNA. Similar results were recently described for a set of HIV-1 mutants with alterations in the RT domain of the Gag Pol polyprotein (Mak et al., 1997). Thus, RT mutants are not appropriate to test for the specific role of the trna primer in genomic RNA dimerization. The alternative approach to produce virions lacking the genome-associated trna primer was to delete the complementary PBS site. Because the RT protein triggers trna Lys3 encapsidation, such a PBS mutant virion is expected to contain normal trna levels, but annealing to the viral RNA is abolished. The RNA genome extracted from the PBS virus was in the dimer conformation (Fig. 1, lane 5). Furthermore, the thermal stability of the PBS RNA dimer was comparable to that of the wild-type RNA genome, with identical melting temperatures of 48 C (Fig. 2). We also tested several HIV-1 mutants with PBS sites complementary to trna primer species not used by natural HIV-1 variants (trna Lys1,2, trna Pro, and trna Trp ). We previously demonstrated binding of these unconventional trna primers to the complementary PBS sites, albeit with reduced efficiency (Das et al., 1995). All these mutant virions contained an RNA genome in the dimer conformation (Fig. 1, lanes 6 8). These results indicate FIG. 3. The RNA dimer is not required for initiation of reverse transcription. The wild-type RNA genome was extracted from SupT1- produced virions and heat-denatured in the C temperature range. These samples were either analyzed on a native agarose gel (A) or used as template for trna extension (B). (A) Position of dimer (d) and monomer (m) RNA. (B) The trna Lys3 primer, which remains associated with the viral RNA on extraction from virions, was extended by the addition of HIV-1 RT and dntps. The 257-nucleotide product, which corresponds to the trna-primed strong-stop minus-strand cdna, was visualized on a 6% polyacrylamide sequencing gel.

4 214 BERKHOUT, DAS, AND VAN WAMEL FIG. 4. The initial steps of reverse transcription. (Top) Schematic of the HIV-1 RNA genome with the trna Lys3 primer basepaired to the PBS site. We boxed the repeat R regions at the extreme 5 - and 3 -ends of HIV-1 RNA. The unique terminal sequences (U5 at the 5 -end, U3 at the 3 -end) are also indicated. Reverse transcription is primed by trna Lys3, and copying of the 5 R region yields the strong-stop cdna intermediate (257 nucleotides in length, including the trna Lys3 primer), which was detected by PCR with the primer set ADR1-CN1. The strong-stop cdna is subsequently transferred to the 3 R region in the first strand-transfer step. Further elongation of reverse transcription generates extended cdna products that were PCR amplified by the primer sets 5 CE-CN1 and NEF.SEQ1-CN1. These primer sets detect cdnas that are 368 and 1,053 nucleotides in length, respectively (including 76 nucleotides of the trna Lys3 primer). scription is not influenced by the multimerization state of the HIV-1 genome. We found that strong-stop cdna synthesis was strongly reduced on templates that were preincubated at 87 C, suggesting that the trna Lys3 primer dissociates from the template at this elevated temperature. Although strong-stop minus-strand cdna is the major product of the trna-extension assay, this molecule is only an intermediate in the process of reverse transcription (see schematic in Fig. 4). Strong-stop cdna contains a copy of the 5 R sequence, and RNase H-mediated degradation of the RNA template will allow this cdna to translocate to the 3 R element, followed by continuation of reverse transcription. Given the considerable amount of strong-stop cdna intermediate produced in the in vitro assay with virion-extracted RNA template, this firststrand transfer step appears to be executed at low efficiency. To detect translocated/extended cdna products, we used a PCR-amplification protocol and two primer sets (Fig. 4, 5 CE-CN1 and NEF.SEQ1-CN1). These PCR reactions allow the detection of translocated cdna products of 368 and 1,053 nucleotides, respectively. As a control, PCR was performed to quantify the amount of strong-stop cdna intermediate (primer set ADR1-CN1). The three PCRs were used to measure the efficiency of reverse transcription along the HIV-1 RNA template, which was either the native RNA dimer or the heatdenatured sample. The different PCR products were observed as ethidium bromide-stained bands in agarose gels (Fig. 5), and we performed Southern blotting to allow accurate quantification of the signals (not shown). Consistent with the trna-extension data of Figure 3B, melting of the RNA dimer at 60 C did not affect strong-stop cdna synthesis, and pretreatment of the RNA template at 90 C did severely abolish strong-stop cdna synthesis because the trna primer was released. Interestingly, synthesis of translocated cdnas was severely reduced with RNA templates treated at 60 C or higher temperatures. The Southern blot was used to quantify the strongstop and extended cdna products, and the ratio of these products was calculated as a relative measure of first strand-transfer efficiency. The transfer efficiency (set at FIG. 5. Analysis of the first strand-transfer on dimeric versus monomeric HIV-1 RNA. Aliquots of the HIV-1 RNA dimers were incubated at increasing temperatures [30 90 C range (top)] and subsequently used as template for reverse transcription. The genome-associated trna Lys3 molecule was used as primer and extended by the addition of HIV-1 RT and dntps. The extent of cdna production was analyzed by PCR with the three primer sets schematically depicted in Figure 4. The PCR products were analyzed on a 1.4% agarose gel. (Right) Size of the DNA fragments (in bp). (Lanes m) DNA molecular weight marker (100-bp ladder). The DNA signals were quantified by Southern blot analysis with an HIV-1 probe directed against the R-U5 region, which detects both early and late reverse transcription products (not shown).

5 HIV-1 RNA DIMER AND REVERSE TRANSCRIPTION % for the untreated RNA template) dropped to 20% on treatment of the RNA sample at 60 C, and 10% transfer was measured for the RNA templates that were preincubated at higher temperatures. This result suggests that the dimer conformation of the genomic RNA template is required for the first strand-transfer step during reverse transcription. An alternative possibility is that the dimer conformation may allow RT to bypass breaks within the RNA template. However, according to this scenario, one would expect a more gradual decline in reverse transcription efficiency along the template. DISCUSSION In this study, we demonstrate that the PBS motif and the genome-associated trna Lys3 primer do not play a role in the process of RNA dimerization. Consistent with this result is the observation that RT-deficient mutants of Rous sarcoma virus, which do not contain a genomeassociated trna primer, do contain a dimeric RNA genome (Sawyer and Hanafusa, 1979; Peters and Hu, 1980). Experiments with a similar RT-deficient HIV-1 variant failed due to instability of the mutant Gag Pol precursor protein (Mak et al. 1997; this study). Although there is accumulating evidence that the processes of genome packaging and dimerization are regulated by multiple motifs within the untranslated HIV-1 leader RNA, the combined results of this dimerization analysis and previous studies that focused on the packaging function (Nagashunmugan et al., 1992; Das et al., 1995; McBride et al., 1997) indicate that the PBS trna duplex is not involved in these processes. We also performed reverse transcription assays with the wild-type HIV-1 RNA genome extracted from virion particles. The associated trna Lys3 primer was elongated to produce the minus-strand strong-stop cdna, and this reaction was equally efficient for the native and denatured RNA template. Thus, it seems that no specific template RNA structure is required for initiation of reverse transcription. Obviously, these experiments do not address the possibility that specific template RNA structures play a role in annealing of the trna primer to the viral genome, which has been proposed based on in vitro experiments (Isel et al., 1995). In contrast to strong-stop cdna synthesis, a dramatic reduction in the subsequent phase of reverse transcription was measured for the denatured RNA template. No extended cdnas were synthesized, indicating that the first strand transfer was abolished. In this reaction, the cdna copy of the 5 -end of the viral RNA genome is transferred to the homologous region located at the 3 -end of the template. This is the first of two cdna transfer steps during reverse transcription (reviewed in Telesnitsky and Goff, 1993, 1997). Because the virus contains two copies of the genomic RNA, this first strand transfer can occur either intramolecularly or intermolecularly. It was initially suggested that the first strand transfer occurs between the two copackaged strands (Panganiban and Fiore, 1988), but more recent studies indicate that there is no bias of interstrand vs intrastrand transfer (Hu and Temin, 1990; Jones et al., 1994; van Wamel and Berkhout, 1998). Formally speaking, even the intramolecular strand transfer is an intermolecular reaction if the template consists of nicked RNA genomes. Regardless of the type of strand transfer, we determined that the native RNA dimer conformation is required for efficient reverse transcription beyond strong-stop cdna synthesis, suggesting that the dimer conformation facilitates the first strand transfer, perhaps by juxtaposing the 5 R and 3 R elements. Interestingly, although it has been suggested previously that virion factors facilitate critical steps of reverse transcription (Borroto-Esoda and Boone, 1991), these effects are usually discussed in the context of protein cofactors. Our experiments suggest that the structure of the RNA dimer is important for the first strand transfer reaction. We will use this in vitro assay system with virion-extracted genomes to further analyze the mechanism of reverse transcription. For instance, studies with mutant genomes are in progress to study in more detail the potential role of higher-order RNA structure in the strand transfer reaction. MATERIALS AND METHODS Mutant and wild-type HIV-1 plasmids The infectious molecular clone plai (Peden et al., 1991) was used for the construction of mutant HIV-1 genomes. The PBS mutations were generated by sitedirected mutagenesis as described previously (Das et al., 1995). The RT construct contains a deletion of an MscI fragment from position 2201 to 4133 (Das and Berkhout, 1995). All mutations were verified by sequence analysis. Cells and DNA transfection SupT1 T cells were cultured in RPMI 1640 medium containing 10% fetal calf serum (FCS) at 37 C and 5% CO 2. These cells were transfected by electroporation as described previously (Back et al., 1996). HeLa cells were grown in Dulbecco s modified Eagle s medium containing 5% FCS and transfected by the DEAE-dextran method (Berkhout et al., 1989). Preparation of virus stocks and RNA isolation The standard protocol for virus production and RNA isolation is as follows. Routinely, 20 ml of culture supernatant of an infected SupT1 culture was harvested at the peak of infection, as indicated by the presence of giant syncytia. Cells were removed by low-speed centrifugation (10 min at 1,500 rpm). The supernatant was filtered through a m-pore filter, and virion particles were

6 216 BERKHOUT, DAS, AND VAN WAMEL pelleted by centrifugation at 25,000 rpm for 30 min at 4 C in a Beckman SW27 rotor. A virus pellet resulting from 20 ml of culture medium was resuspended in 500 l of buffer (50 mm Tris, ph 7.4, 10 mm EDTA, and 100 mm NaCl). A small aliquot of 10 l was removed for CA-p24 ELISA and RT activity assays. The virus stock solution was supplemented with 50 l of a 10% SDS solution and 5 l of trna [50 g/ml Escherichia coli trna (RNase-free); Boehringer-Mannheim] and 5 l of Proteinase K (50 g/ml; Boehringer-Mannheim) and incubated at 37 C for 30 min. The sample was extracted with 0.5 volume of phenol and subjected to phase separation by centrifugation after the addition of a similar amount of chloroform isoamyl alcohol (24:1). The extraction was repeated, and the viral RNA was precipitated from the aqueous layer by the addition of 2.5 volumes of 96% ethanol and 0.1 volume of 3 M sodium acetate (ph 5.2) (30 min at 20 C). The RNA was pelleted (10 min at 14,000 rpm at 4 C) and rinsed with 70% ethanol. The pellet was dried briefly, dissolved in 50 l of TE (10 mm Tris, ph 7.5, 1 mm EDTA), aliquoted, and stored at 80 C. Virus of the replication-impaired RT and PBS mutants was prepared in transiently transfected HeLa cells as described previously (Das et al., 1995). In contrast to the viral RNA extracted from SupT1-produced virions, we added carrier Bluescript plasmid to these samples. All virus samples were analyzed for CA-p24 by ELISA and RT activity by the poly(ra)/oligo(dt) assay as described previously (Back et al., 1996). In vitro manipulation of virion RNA To melt the virion-extracted dimer genome, 1 l ofrna sample was mixed with 9 l of TEN SDS buffer (10 mm Tris [ph7.5], 1 mm EDTA, 1% SDS, and 100 mm NaCl) and incubated for 10 min at 65 C. To obtain a melting curve, RNA samples were treated at increasing temperatures as indicated in individual experiments. For gel analysis, 4 l of sample buffer (15% Ficoll 400, 0.25% bromphenol blue) was added to the samples. The RNA was separated on nondenaturing 0.9% agarose gels in 1 Tris borate EDTA buffer. Electrophoresis was at 100 V for 4 h. The gel was soaked in 3 volumes of 10% formaldehyde at 65 C for 30 min, and the RNA was electroblotted at 30 V (BioRad chamber) in 1 TBE buffer onto a positively charged membrane (Microporous Nylon 66; Boehringer- Mannheim). The damp membrane was UV-crosslinked at 120,000 J/cm 2 (Stratalinker), washed overnight in 2 SSC (1 SSC 0.15 M NaCl plus M sodium citrate), and then prehybridized for 4hat65 C in buffer A (3 SSC, 0.1% SDS, 10 Denhardt s solution, and 50 g of salmon sperm DNA/ml). Hybridization with a denatured HIV probe was done for 16 h at 65 C in buffer B (6 SSC, 0.1% SDS, 10% dextran sulfate, and 50 g of salmon sperm DNA/ml). The 32 P-labeled HIV-1 DNA probe was generated by random-primer labeling (Boehringer-Mannheim) of a BstXI fragment of the gag gene (plai coordinates ). Filters were washed four times in buffer C (1 SSC, 1% SDS) for 15 min at 65 C. The blots were quantified with a PhosphorImager (Molecular Dynamics). Reverse transcription assays with virion-extracted RNA cdna synthesis on the viral RNA template was initiated by the natural trna Lys3 primer that remains annealed to the viral RNA on purification. This trna-extension assay was performed by the addition of dntps and RT enzyme as described previously (Das et al., 1995). The HIV-1 RT enzyme used is a recombinant GST-RT fusion protein purified from E. coli (Oude Essink et al., 1997). The incorporation of [ - 32 P]dCTP results in the production of a labeled 257-nucleotide-long cdna product that is referred to as minus-strand strong-stop cdna, which was visualized on a denaturing 6% polyacrylamide urea sequencing gel. To analyze more extended cdna products, we used an alternative reverse transcription protocol. In brief, we first treated the RNA sample with RNase-free DNase and performed the reverse transcription reaction with unlabeled dntps and HIV-1 RT enzyme. Different cdna products were analyzed by PCR amplification. The primer sets used in this study are schematically depicted in Figure 3. The PCR products were separated on 1.0% agarose gels, and quantification of the products was performed by Southern blotting (Sambrook et al., 1989) with 32 P-labeled probes. Radioactive bands were quantified on a Molecular Dynamics PhosphorImager. ACKNOWLEDGMENTS We thank Wim van Est for assistance in preparing the figures. This research was sponsored by the European Community (Grant ), the Dutch Cancer Society (KWF), and the Dutch AIDS Fund (AIDS Fonds, Amsterdam). REFERENCES Aiyar, A., Cobrinik, D., Ge, Z., Kung, H. J., and 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, Awang, G., and Sen, D. (1993). Mode of dimerization of HIV-1 genomic RNA. Biochemistry 32, 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., and 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, Berkhout, B., Silverman, R. H., and Jeang, K. T. (1989). Tat trans-activates the human immunodeficiency virus through a nascent RNA target. Cell 59, Berkhout, B., Oude Essink, B. B., and Schoneveld, I. (1993). In vitro dimerization of HIV-2 leader RNA in the absence of PuGGAPuA motifs. FASEB J. 7,

7 HIV-1 RNA DIMER AND REVERSE TRANSCRIPTION 217 Berkhout, B., and Schoneveld, I. (1993). Secondary structure of the HIV-2 leader RNA comprising the trna-primer binding site. Nucleic Acids Res. 21, Berkhout, B., and van Wamel, J. L. B. (1996). Role of the DIS hairpin in replication of human immunodeficiency virus type 1. J. Virol. 70, Bieth, E., Gabus, C., and Darlix, J.-L. (1990). A study of the dimer formation of Rous sarcoma virus RNA and its effect on viral protein synthesis in vitro. Nucleic Acids Res. 18, Borroto-Esoda, K., and Boone, L. R. (1991). Equine infectious anemia virus and human immunodeficiency virus DNA synthesis in vitro: characterization of the endogenous reverse transcriptase reaction. J. Virol. 65, Clever, J. L., Wong, M. L., and Parslow, T. G. (1996). Requirements for kissing-loop-mediated dimerization of human immunodeficiency virus RNA. J. Virol. 70, Coffin, J. (1984). Structure of the retroviral genome. In RNA Tumor Viruses, 4th ed., Vol. 1 (R. Weiss, N. Teich, H. Varmus, and J. Coffin, Eds.), pp Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Darlix, J.-L., Gabus, C., Nugeyre, M. T., Clavel, F., and Barre-Sinoussi, F. (1990). cis elements and trans-acting factors involved in the RNA dimerization of the human immunodeficiency virus HIV-1. J. Mol. Biol. 216, Darlix, J. L., Gabus, C., and Allain, B. (1992). Analytical study of avian reticuloendotheliosis virus dimeric RNA generated in vivo and in vitro. J. Virol. 66, Das, A. T., Klaver, B., and Berkhout, B. (1995). Reduced replication of human immunodeficiency virus type 1 mutants that use reverse transcription primers other than the natural trna(3lys). J. Virol. 69, Das, A. T., and Berkhout, B. (1995). Efficient extension of a misaligned trna-primer during replication of the HIV-1 retrovirus. Nucleic Acids Res. 23, Fu, W., Gorelick, R. J., and Rein, A. (1994). Characterization of human immunodeficiency virus type 1 dimeric RNA from wild-type and protease-defective virions. J. Virol. 68, Fu, W., and Rein, A. (1993). Maturation of dimeric viral RNA of Moloney murine leukemia virus. J. Virol. 67, Goncalves, J., Korin, Y., Zack, J., and Gabuzda, D. (1996). Role of Vif in human immunodeficiency virus type 1 reverse transcription. J. virol. 70, Haddrick, M., Lear, A. L., Cann, A. J., and Heaphy, S. (1996). Evidence that a kissing loop structure facilitates genomic RNA dimerisation in HIV-1. J. Mol. Biol. 259, Haseltine, W. A., Maxam, A. M., and Gilbert, W. (1977). Rous sarcoma virus genome is terminally redundant: the 5 sequence. Proc. Natl. Acad. Sci. USA 74, Hu, W.-S., and Temin, H. M. (1990). Retroviral recombination and reverse transcription. Science 250, Isel, C., Ehresmann, C., Keith, G., Ehresmann, B., and Marquet, R. (1995). Initiation of reverse transcription of HIV-1: secondary structure of the HIV-1 RNA/tRNA(3Lys) (template/primer). J. Mol. Biol. 247, Jones, J. S., Allan, R. W., and Temin, H. M. (1994). One retroviral RNA is sufficient for synthesis of viral DNA. J. Virol. 68, Laughrea, M., and Jette, L. (1994). A 19-nucleotide sequence upstream of the 5 major splice donor is part of the dimerization domain of human immunodeficiency virus 1 genomic RNA. Biochemistry 33, Laughrea, M., and Jette, L. (1996). HIV-1 genome dimerization: formation kinetics and thermal stability of dimeric HIV-1LAI RNAs are not improved by the and regions flanking the kissingloop domain. Biochemistry 35, Mak, J., Jiang, M., Wainberg, M. A., Hammarskjold, M. L., Rekosh, D., and Kleiman, L. (1994). Role of Pr160gag-pol in mediating the selective incorporation of trna(lys) into human immunodeficiency virus type 1 particles. J. Virol. 68, Mak, J., Khorchid, A., Cao, Q., Huang, Y., Lowy, I., Parniak, M. A., Prasad, V. R., Wainberg, M. A., and Kleiman, L. (1997). Effects of mutations in Pr160gag-pol upon trnalys3 and Pr160gag-pol incorporation into HIV-1. J. Mol. Biol. 265, Marquet, R., Baudin, F., Gabus, C., Darlix, J. L., Mougel, M., Ehresmann, C., and Ehresmann, B. (1991). Dimerization of human immunodeficiency virus (type 1) RNA: stimulation by cations and possible mechanism. Nucleic Acids Res. 19, Marquet, R., Paillart, J. C., Skripkin, E., Ehresmann, C., and Ehresmann, B. (1994). Dimerization of human immunodeficiency virus type 1 RNA involves sequences located upstream of the splice donor site. Nucleic Acids Res. 22, McBride, M. S., Schwartz, M. D., and Panganiban, A. T. (1997). Efficient encapsidation of human immunodeficiency virus type 1 vectors and further characterization of cis elements required for encapsidation. J. Virol. 71, Mougel, M., Tounekti, N., Darlix, J. L., Paoletti, J., Ehresmann, B., and Ehresmann, C. (1993). Conformational analysis of the 5 leader and the gag initiation site of Mo-MuLV RNA and allosteric transitions induced by dimerization. Nucleic Acids Res. 21, Nagashunmugam, T., Velpandi, A., Goldsmith, C. S., Zaki, S. R., Kalyanaraman, V. S., and Srinivasan, A. (1992). Mutation in the primer binding site of the type 1 human immunodeficiency virus genome affects virus production and infectivity. Proc. Natl. Acad. Sci. USA 89, Oude Essink, B. B., Back, N. K. T., and Berkhout, B. (1997). Increased polymerase fidelity of the 3TC-resistant variants of HIV-1 reverse transcriptase. Nucleic Acids Res. 25, Paillart, J.-C., Berthoux, L., Ottmann, M., Darlix, J.-L., Marquet, R., Ehresmann, B., and Ehresmann, C. (1996a). A dual role of the putative RNA dimerization initiation site of human immunodeficiency virus type 1 in genomic RNA packaging and proviral DNA synthesis. J. Virol. 70, Paillart, J.-C., Skripkin, E., Ehresmann, B., Ehresmann, C., and Marquet, R. (1996b). A loop-loop kissing complex is the essential part of the dimer linkage of genomic HIV-1 RNA. Proc. Natl. Acad. Sci. USA 93, Paillart, J. C., Marquet, R., Skripkin, E., Ehresmann, B., and Ehresmann, C. (1994). Mutational analysis of the bipartite dimer linkage structure of human immunodeficiency virus type 1 genomic RNA. J. Biol. Chem. 269, Panganiban, A. T., and Fiore, D. (1988). Ordered interstrand and intrastrand DNA transfer during reverse transcription. Science 241, Paoletti, J., Mougel, M., Tounekti, N., Girard, P. M., Ehresmann, C., and Ehresmann, B. (1993). Spontaneous dimerization of retroviral Mo- MuLV RNA. Biochimie (Paris) 75, Peden, K., Emerman, M., and Montagnier, L. (1991). Changes in growth properties on passage in tissue culture of viruses derived from infectious molecular clones of HIV-1 LAI, HIV-1 MAL, and HIV-1 ELI. Virology 185, Peters, G. G., and Hu, J. (1980). Reverse transcriptase as the major determinant for selective packaging of trnas into avian sarcoma virus particles. J. Virol. 36, Roy, C., Tounekti, N., Mougel, M., Darlix, J. L., Paoletti, C., Ehresmann, C., Ehresmann, B., and Peoletti, J. (1990). An analytical study of the dimerization of in vitro generated RNA of Moloney murine leukemia virus MoMuLV. Nucleic Acids Res. 18, Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Sawyer, R. C., and Hanafusa, H. (1979). Comparison of the small RNAs of polymerase-deficient and polymerase-positive Rous sarcoma virus and another species of avian retrovirus. J. Virol. 29,

8 218 BERKHOUT, DAS, AND VAN WAMEL Skripkin, E., Paillart, J. C., Marquet, R., Ehresmann, B., and Ehresmann, C. (1994). Identification of the primary site of the human immunodeficiency virus type 1 RNA dimerization in vitro. Proc. Natl. Acad. Sci. USA 91, Sundquist, W. I., and Heaphy, S. (1993). Evidence for interstrand quadruplex formation in the dimerization of human immunodeficiency virus 1 genomic RNA. Proc. Natl. Acad. Sci. USA 90, Telesnitsky, A., and Goff, S. P. (1993). Strong-stop strand transfer during reverse transcription. In Reverse Transcriptase, 4th Ed. (A. M. Skalka and S. P. Goff, Eds.), pp Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Telesnitsky, A., and Goff, S. P. (1997). Reverse transcriptase and the generation of retroviral DNA. In Retroviruses (J. M. Coffin, S. H. Hughes, and H. E. Varmus, Eds.), pp Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Tounekti, N., Mougel, M., Roy, C., Marquet, R., Darlix, J. L., Paoletti, J., Ehresmann, B., and Ehresmann, C. (1992). Effect of dimerization on the conformation of the encapsidation Psi domain of Moloney murine leukemia virus RNA. J. Mol. Biol. 223, van Wamel, J. L. B., and Berkhout, B. (1998). The first strand transfer during HIV-1 reverse transcription can occur either intramolecularly or intermolecularly. Virology 244,