Human Immunodeficiency Virus Type 1 Nef Increases the ... - NCBI

3 downloads 0 Views 311KB Size Report
We have analyzed the replication of Nef and Nef isogenic human immunodeficiency virus in CEM,. HUT78, MT4 lymphoid, and U937 monocytic cell lines.
JOURNAL OF VIROLOGY, July 1995, p. 4053–4059 0022-538X/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 69, No. 7

Human Immunodeficiency Virus Type 1 Nef Increases the Efficiency of Reverse Transcription in the Infected Cell ´ CHAL, OLIVIER DANOS, OLIVIER SCHWARTZ,* VALERIE MARE

AND

JEAN-MICHEL HEARD

Laboratoire Re´trovirus et Transfert Ge´ne´tique, Unite´ Recherche Associe´e Centre National de la Recherche Scientifique 1157, Institut Pasteur, 75724 Paris Cedex 15, France Received 23 December 1994/Accepted 5 April 1995

We have analyzed the replication of Nef1 and Nef2 isogenic human immunodeficiency virus in CEM, HUT78, MT4 lymphoid, and U937 monocytic cell lines. At each passage of infected cells, we have assessed the relative infectivity of the virus particles released in culture media by measuring the number of infectious units per nanogram of p24 protein. Values appeared to be 3- to 10-fold higher for the Nef1 virus than for the Nef2 virus. The positive effect of Nef was observed regardless of the cell line, the multiplicity of infection, and the number of virus replication cycles achieved. We showed, by using cells expressing glycosylphosphatidylinositollinked CD4, that the enhancement of virion infectivity could be dissociated from the down-regulation of cell surface CD4 also induced by Nef. The gp120-to-p24 ratio and the RNA content of virus particles produced in the presence or in the absence of Nef were equivalent. Virions bound to cell surface CD4 receptors with equal efficiencies. Equivalent reverse transcriptase activities were measured both on exogenous substrate and on particle genomic RNAs. In contrast, reverse transcription in infected cells generated 5- to 10-fold less DNA when the virions were produced in the absence of Nef, indicating that these particles performed reverse transcription in a suboptimal environment. These data suggest that the expression of Nef in virus-producing cells is required for efficient processing of the early stages of virus replication in target cells, including the internalization in an appropriate cell compartment and the uncoating of the particle. The nef gene is unique to primate lentiviruses, human immunodeficiency virus (HIV), and simian immunodeficiency virus (for reviews, see references 10 and 22). It is expressed early in the viral cycle and encodes a 27- to 30-kDa myristoylated protein which is predominantly localized in the cytoplasm of infected cells and associated with membranes (12, 18). In simian immunodeficiency virus-infected rhesus monkeys, the expression of Nef is necessary for maintaining high virus loads and for inducing AIDS (19). The best-documented biological activity of Nef is to stimulate the internalization of cell surface CD4, its accumulation into early endosomes, and its degradation (2, 14, 16, 26, 27). Several groups have observed a delayed and reduced production of virus by cells infected with nef mutants in comparison with wild-type HIV, indicating that Nef facilitates virus replication in cultured cells (7, 11, 20, 23, 28, 32). This effect was much pronounced in quiescent primary lymphocytes activated upon infection (23, 28) but has also been observed in immortalized T-cell lines infected with low viral inputs (7). The observation that, for equivalent amounts of p24, wild-type HIV particles infected five to six times more cells than Nef2 mutants suggests a role of Nef in virion infectivity (7, 23). The increased infectivity of viral particles observed in the presence of Nef seems to be independent of CD4 down-regulation, as Nef1 viruses produced in transfected COS cells lacking CD4 were also more infectious (23). Using HeLa-CD4 bgal as indicator target cells, Miller et al. have shown that Nef-enhanced virion infectivity can be revealed within a single cycle of virus replication, suggesting that the mechanism supporting this effect operates in target cells between CD4 binding and Tat expression (23). Miller et al. have also recently shown that Nef

increases particle infectivity independently of gp160 or viral entry (24). Since Nef has not been detected in virus particles (4, 12), it is likely that the enhanced infection of target cells is determined in virus-producing cells. We have analyzed the infectivity of particles from Nef1 and Nef2 isogenic HIV produced in a variety of cell lines. For each viral supernatant, we have defined a relative particle infectivity index (RPI) as the number of infectious units per nanogram of p24 protein. Whatever the cell line in which viruses were produced and the number of virus replication cycles achieved, RPIs were 3- to 10-fold higher in the presence of Nef. The absence of Nef in the producer cells was not associated with any significant differences in the amounts of Gag and Env proteins, reverse transcriptase (RT) activity, and viral RNA incorporated into the particles. The capacity of virions to bind surface CD4 on target cells was equivalent for HIV and HIVDnef. In contrast, despite equivalent RT activity of the virions, proviral DNA synthesis in target cells was 5- to 10-fold less efficient for HIVDnef than for HIV. Overall, our data show that Nef acts at a very early step of the virus replication cycle, after the binding of the virion to its receptor and before the completion of reverse transcription. MATERIALS AND METHODS Cells, viruses, and reagents. CEM, HUT78, HSB2-GPiCD4, and MT4 lymphocytic cells and U937 monocytic cells were grown in RPMI medium supplemented with 10% fetal calf serum, antibiotics, and 2 mM glutamine. The HSB2-GPiCD4 cell line is a derivative of HSB2 T cells which express glycosylphosphatidylinositol (GPi)-linked CD4 (a kind gift of Robert W. Finberg, Dana-Farber Cancer Institute, Boston, Mass. [21]). P4 cells (8) were grown in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal calf serum, antibiotics, and 2 mM glutamine. HIV was produced from a transfected pNL432 plasmid containing full-length HIV type 1 (HIV-1) DNA (1). HIVDnef was produced from the pNL432-Xho plasmid, constructed by inserting a frameshift mutation of four bases at the unique XhoI site, with the consequence that the nef gene from pNL432-Xho had the coding capacity for only the first 35 N-terminal amino acid residues of the protein. pNL432 and pNL432-Xho plasmids were a kind gift of Franc¸oise Bachelerie (Institut Pasteur).

* Corresponding author. Mailing address: Laboratoire Re´trovirus et Transfert Genetique, URA CNRS 1157, Institut Pasteur, 25 rue du Dr Roux, 75724 Paris Cedex 15, France. Phone: 33 (1) 45 68 82 46. Fax: 33 (1) 45 68 88 85. Electronic mail address: [email protected]. 4053

4054

SCHWARTZ ET AL.

Infection of cell lines. Two protocols were used for viral infections. First, 10 mg of pNL432 or pNL432-Xho was transfected into HeLa cells by calcium phosphate coprecipitation. CEM and HUT78 cells were infected in a 48-h cocultivation period with transfected HeLa cells. At this time, lymphoid cells were harvested and suspended at each passage at a concentration of 3 3 105 viable cells per ml in fresh medium. Second, virus stocks from HIV- and HIVDnef-infected HUT78 cells were prepared by centrifugation of cell supernatants to remove debris, followed by filtration through a 0.45-mm-pore-size Millipore filter. The titers of the stocks were determined (p24 contents and PFU) and the stocks were stored at 2808C. Infections of HSB2-GPiCD4, MT4, and U937 cells were done by incubating the cells with viral supernatants previously adjusted for identical numbers of PFU. Cells were incubated for 2 h with HIV or HIVDnef at the same multiplicity of infection (0.1 PFU per cell) in the presence of 5 mg of Polybrene per ml, and the samples were diluted 10 times in fresh medium. After overnight infection, and at each cell passage, the cells were washed and resuspended in culture medium at a concentration of 3 3 105 viable cells per ml. Virus stocks from infected MT4 and U937 cells were prepared, and the titers were determined as described for HUT78 cells. Titration of HIV. Virus stocks were assessed for p24 concentration by using a commercial antigen capture enzyme-linked immunosorbent assay (ELISA) (HIV-1 p24 Core Profile ELISA; Du Pont de Nemours-NEN), according to the manufacturer’s instructions. The infectivity of viral supernatants was determined by using P4 cells (8). These HeLa-CD4 cells carry the bacterial lacZ gene under the control of the HIV-1 long terminal repeat (LTR), and cytoplasmic accumulation of b-galactosidase is strictly dependent on the presence of Tat. One day before infection, 1.2 3 105 cells were plated in each well of a 24-well plate. The following day, the medium was replaced with 500 ml of medium containing serial dilutions of the virus stock (from 100 to 0.1 ml) in the presence of 20 mg of DEAE-dextran per ml. After 3 h of incubation with the viral samples, the medium was completed to 1 ml with fresh medium. After a 24-h incubation at 378C, cells were fixed, and b-galactosidase activity was revealed with 5-bromo4-chloro-3-indolyl-b-D-galactopyranoside (X-Gal) (8). Blue cell foci were scored by using a binocular microscope. Experiments were performed in triplicate or in duplicate, and variations for each sample were less than 10%. A background level of 5 to 15 blue cells was detected in wells of uninfected cells and was subtracted from the number of blue cells scored for experimental points. This assay was reproducible and generated linear values when serial virus dilutions were used. Immunoprecipitations. After starvation in methionine- and cysteine-free medium, cells (107 per sample) were metabolically labeled for 16 h with 35STranslabel (100 mCi/ml in 4 ml [final volume]; Amersham). Cells and supernatants were separated by a 5-min centrifugation at 1,200 3 g. Cells were lysed in 13 lysis buffer (1% Nonidet P-40 [NP-40], 150 mM NaCl, 20 mM Tris-HCl [pH 8], 2 mM EDTA, 1 mM phenylmethylsulfonyl fluoride [PMSF]), and supernatants were divided into two parts. One milliliter was lysed by using 103 lysis buffer in order to obtain a final concentration of 13 lysis buffer. The other part (3 ml) was subjected to ultracentrifugation for 15 min at 60,000 rpm in a Beckman TL100 centrifuge, and pellets were resuspended in 1 ml of 13 lysis buffer. Cell extracts (corresponding to 2 3 106 cells for each experimental point) were clarified by centrifugation and precleared overnight with protein A-Sepharose (Sigma). Lysates were then incubated with a serum from an HIV-1-seropositive subject at a final dilution of 1:100 for 2 h, and immune complexes were precipitated with protein A-Sepharose. After washing in a solution containing 20 mM Tris (pH 7.4), 0.1% deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 0.1% Triton X-100, and 150 mM NaCl, immunoprecipitates were subjected to SDS– 10% polyacrylamide gels electrophoresis (PAGE) followed by fluorography. Gels were revealed by autoradiography or analyzed by using a PhosphorImager (model 425F; Molecular Dynamics), and signal quantification was done with ImageQuant software version 3.1 (Molecular Dynamics). Quantification of viral RNA. Viral RNA from supernatants or pelleted viruses was extracted by incubating samples (750 ml) for 45 min at 378C with 90 ml of RNA lysis buffer (20 mM EDTA [pH 7.4], 0.8 mg of tRNA per ml, 4.2% SDS, 4.2 mg of proteinase K per ml). The samples were then extracted sequentially in phenol, phenol-chloroform, and chloroform and precipitated by the addition of 0.1 volume of 5 M NaCl and 2.5 volumes of 100% ethanol. Samples were washed once in 100% ethanol, resuspended in 50 ml of H2O, serially diluted, incubated for 15 min at 658C with denaturating solution (20 ml of 203 MOPS [morpholinepropanesulfonic acid], 35 ml of formaldehyde, 10 ml of formamide), and kept on ice. One volume of ice-cold 203 SSC (13 SSC is 0.15 M NaCl plus 0.015 M sodium citrate) was added before the solution was blotted onto nylon N1 membranes (Amersham) by using a dot blot apparatus. The membranes were washed twice in 103 SSC, dried, and fixed on a UV transilluminator (Stratagene Stratalinker). Prehybridization was performed for 1 h at 658C in 7% SDS–0.5 M phosphate buffer (pH 7). A denatured HIV DNA probe (1.9-kb MscI fragment of the pol gene of pNL43) was hybridized overnight at 658C. Membranes were washed three times at room temperature in 23 SSC–0.1% SDS and twice in 0.13 SSC–0.1% SDS at 658C and autoradiographed for 3 h. A markedly reduced signal was detected when RNA was similarly prepared from a mutant HIV with a deletion in its packaging signal (9), indicating that this assay allowed the specific detection of incorporated viral RNA. No signal was detected when samples were hybridized with an irrelevant probe.

J. VIROL. Exogenous RT activity. RT activity was measured by incubating 10 ml of cell-free supernatant or viral pellet at 378C for 1 h with 50 ml of RT cocktail containing 0.05% NP-40, 50 mM Tris [pH 7.8], 7.5 mM KCl, 2 mM dithiothreitol, 5 mM MgCl2, 5 mg of poly(rA) per ml, 1.6 mg of oligo(dT) per ml, [a-32P]dTTP (specific activity, .400 Ci/mmol; final concentration, 10 mCi/ml; Amersham). The samples were then filtered on DE 81 paper sheets (Whatman) by using a dot blot apparatus and washed six times in 23 SSC. The sheets were then dried and autoradiographed for 3 h. Endogenous RT assay. An endogenous RT assay was performed by using a modification of a procedure that avoided virion concentration and purification (5). To 500 ml of virus sample (corresponding to a 24-h virus harvest in culture medium), 50 ml of 103 TME {500 mM Tris HCl [pH 8.1], 30 mM MgCl2, 30 mM EGTA [ethylene glycol-bis-(b-aminoethyl ether)-N,N,N9,N9,-tetraacetic acid], 25 ml of 203 deoxynucleoside triphosphates [final concentration, 100 or 10 mM each], 25 ml of 203 NP-40 [final concentration, varying from 0 to 1%]} was added. The reaction mixtures were incubated for 2 h at 418C, and the reactions were stopped by the addition of an equal volume of stop buffer (final concentrations, 0.5% SDS, 25 mM EDTA, 100 mM NaCl). Proteinase K was added to 100 mg/ml, and the samples were incubated for 1 h at 568C and then subjected to extractions with phenol and phenol-chloroform-isoamyl alcohol (25:24:1). DNA products were precipitated with ethanol and dissolved in Tris-EDTA. Samples were run in neutral 0.9% gels after alkaline hydrolysis in 200 mM NaOH in order to remove viral RNA and to denature the DNA. After transfer to nylon membranes, the samples were hybridized with a 32P-labeled HIV DNA probe (1.9-kb MscI fragment from the pol region of pNL43). Virus binding assay. CD41 cells were incubated with viral samples (supernatants or viral pellets) for 60 min at 48C, and surface-bound gp120 levels were measured with the 110.H anti-gp120 monoclonal antibody (a kind gift of Fran¸cois Traincart, Hybridolab, Institut Pasteur) by using a FACScan cytofluorimeter (Becton Dickinson) as previously described (25). In this assay, the signal fluorescence intensity is proportional to the amounts of cell-bound soluble gp120 or virus particles. Analysis of viral DNA synthesis in infected cells. CEM target cells (25 3 106 per sample) were infected by cocultivation with chronically HIV- and HIVDnefinfected HUT78 or U937 cells (5 3 106 per sample). MT4 target cells (12 3 106 per sample) were infected with cell-free HIV and HIVDnef (from acutely infected MT4 cells) adjusted for an equivalent p24 concentration (24 mg in a final volume of 35 ml), in the presence of 20 mg of DEAE-dextran per ml. Twenty-four hours after infection, low-molecular-weight DNA was prepared by Hirt extraction (17), EcoRI digested, and subjected to Southern blot analysis. DNA from approximately 4 3 106 target cells was analyzed in each lane. The 32P-labeled DNA probe used for hybridization was the 1.9-kb MscI fragment from the pol region of pNL43. Total amounts of low-molecular-weight DNA were detected by using the mitochondrial gene cytochrome b as a probe. Gels were visualized by autoradiography or with a PhosphorImager.

RESULTS Replication of HIV and HIVDnef in various immortalized cell lines. In the first experiment, two viruses derived from the molecular clone pNL432 were produced in HUT78 cells, a nef mutant virus (HIVDnef) carrying a frameshift mutation in the nef gene (see Materials and Methods) and the wild-type virus (HIV). HUT78 cells were infected by cocultivation with HeLa cells transfected with viral DNAs and monitored for virus production by measuring p24 in culture supernatants (Fig. 1, left column). HIV and HIVDnef produced equivalent amounts of p24 with the same kinetics. Similar virus-induced cytopathic effects, evidenced by the presence of numerous syncytia and cell agregates at days 12 to 16 postinfection, were observed for HIV and HIVDnef. These data agreed with previous observations that Nef is dispensable for HIV replication in T-cell lines (20, 29). We determined the number of infectious units (PFU) in the supernatants by using the P4 (HeLa-CD4 LTR lacZ) indicator cell line (see Materials and Methods) (Fig. 1, central column), and the RPI was defined as the number of PFU per nanogram of p24 protein (Fig. 1, right column). The RPI varied during the period of virus production. With HIV, it peaked at 700 PFU/ng of p24 at the time of maximal p24 production, before the occurrence of an important cytopathic effect (day 10 postinfection). It is likely that the time of maximal RPI corresponds to an optimal incorporation of viral proteins and RNA in the virions, associated with a low virusinduced cytopathic effect. RPI was 3- to 10-fold lower for HIVDnef than for HIV and never rose above 110 PFU/ng of

VOL. 69, 1995

FIG. 1. Replication of HIV and HIVDnef. HUT78 cells were infected with HIV and HIVDnef by a 48-h cocultivation with HeLa cells transfected with the pNL43 and pNL43-Xho plasmids, respectively, and then were split at 3 3 105 per ml every 3 to 4 days. Supernatants were collected at each cell passage and analyzed for the p24 concentration by ELISA (left column) and for infectious titers by using the P4 (HeLa-CD4 LTR lacZ) indicator cell line (middle column). Particle infectivity (right column) was calculated as the number of PFU per nanogram of p24. U937, MT4, and HSB2-GPiCD4 cells were infected overnight by incubation with cell-free supernatants from HIV- or HIVDnef-infected HUT78 cells that were previously adjusted for PFU contents (multiplicity of infection, 0.1). Viral replication was monitored every 3 to 4 days. HSB2-GPiCD4 cells were derived from HSB2 lymphoid cells (21) and express a GPi-linked CD4 molecule that was not down-regulated by Nef. Data were obtained from at least two independent experiments, except for infections of HSB2-GPiCD4 cells, which were analyzed once.

p24. The same effect was observed after the viruses were pelleted by ultracentrifugation, indicating that the difference of infectivities between the HIV and HIVDnef was not due to the presence of soluble viral or cellular compounds in cell supernatants (data not shown). Experiments were also performed with MT4 lymphoblastoid and U937 monocytoid cell lines (Fig. 1). Cells were infected with supernatants from HIV- and HIVDnef-infected HUT78 cells, by using equivalent viral inputs (0.1 PFU per cell). The infection of MT4 cells with HIV and HIVDnef was very efficient, producing 4,000 ng of p24 per ml at day 4. It was associated with a massive cytopathic effect at day 5 (95% cell mortality). In U937 cells, the peak of p24 production was observed on day 9. As seen before with HUT78 cells, no sig-

HIV-1 Nef INCREASES EFFICIENCY OF RT

4055

nificant difference in p24 production was observed between HIV- and HIVDnef-infected cells. However, with both cell lines, RPIs were 3- to 12-fold higher in the supernatants of HIV-infected cells. Similar data were obtained with CEM cells (data not shown). These experiments showed that the enhanced infectivity of Nef-encoding viruses was independent of the nature of the virus-producing cell. Infected HUT78 and U937 cells were maintained in culture for at least 8 weeks, exhibiting little or no cytopathic effect. Cells chronically infected with HIV and HIVDnef produced equivalent amounts of p24 (50 to 200 ng of p24 per ml of supernatant). RPIs were 3- to 10-fold lower for HIVDnef than for wild-type viruses during the observation period (data not shown). To determine whether Nef-induced CD4 down-regulation and Nef-associated enhanced infectivity of virus particles could be dissociated, we examined the replication of HIV and HIVDnef in HSB2 cells expressing the GPiCD4 cell surface molecule. This receptor associates the extracellular domain of CD4 and a GPi membrane anchor region (21). Because the cytoplasmic tail of CD4 is necessary for Nef-induced internalization (2, 3), GPiCD4 was not expected to be down-regulated in Nef-expressing cells. This was verified by expressing Nef from a vaccinia virus vector (16) in HSB2-GPiCD4 cells (data not shown). As observed in HUT78, MT4, U937, and CEM cells, HIV produced in HSB2-GPiCD4 cells showed an RPI 4to 13-fold higher than that of HIVDnef (Fig. 1). This experiment indicated that Nef-associated enhanced infectivity of virus particles did not require CD4 down-regulation. Analysis of virion structural components. In an attempt to elucidate the basis of Nef-enhanced virion infectivity, we determined the viral protein and RNA content of HIV or HIVDnef particles. We first examined the synthesis of viral proteins in virus-producing cells. U937 cells chronically infected with HIV or HIVDnef were labeled for 16 hours with [35S]methionine, and cell lysates, culture supernatants, and pelleted virions were immunoprecipitated with a polyclonal human anti-HIV-1 serum and analyzed by SDS-PAGE (Fig. 2a). In cell lysates, the profile of viral proteins was similar for HIV and HIVDnef, which contained similar amounts of gp160/ gp120 envelope glycoproteins and p24 proteins. These data showed that Nef did not significantly alter the synthesis of the main structural components of virus particles. As expected, the 27-kDa Nef-specific signal protein was absent in HIVDnefinfected cells. Cell supernatants mostly contained the envelope gp120 protein and the p24 capsid protein. PhosphorImager quantification gave equivalent gp120/p24 ratios in HIV- and HIVDnef-infected cell supernatants (data not shown). Pelleted particles were also analyzed for envelope glycoproteins and capsid proteins. The gp120 signal was reduced in pelleted virions, because the soluble fraction was eliminated during ultracentrifugation. Equivalent gp120/p24 ratios were observed in pelleted HIV and HIVDnef virions, indicating that Nef expression did not affect the coating of virus particles with envelope glycoproteins. Similar analysis of chronically infected HUT78 cells and during the acute infection of MT4 cells led to the same conclusions (data not shown). RNAs were extracted from virus particles produced in HIVand HIVDnef-infected HUT78 cells. Culture supernatants and viral pellets were adjusted for equivalent p24 contents. RPIs were 826 and 86 PFU/ng of p24 for HIV and HIVDnef, respectively. After blotting and hybridization with an HIV-specific probe, equivalent signals were detected with RNAs extracted from HIV and HIVDnef particles (Fig. 2b). Similar results were obtained with viruses collected during acute infection of MT4 cells (data not shown). These data indicated

4056

SCHWARTZ ET AL.

J. VIROL.

In the endogenous RT assay, we measured in vitro the capacity of virion-associated RT to synthesize DNA from virionassociated tRNA and genomic RNA. HIV and HIVDnef supernatants from acutely infected MT4 cells (Fig. 3b) or from HUT78 cells (data not shown) were adjusted for equivalent amounts of p24 and were incubated for 2 h at 418C in the presence of various concentrations of deoxynucleoside triphosphates and detergent. Viral DNA was extracted, denatured with NaOH, and analyzed by Southern blotting and hybridization with a labeled probe corresponding to the pol region of pNL43 (Fig. 3b). Optimal DNA synthesis was obtained in the presence of 0.01% NP-40 and 100 mM deoxynucleoside triphosphates. The main reaction product was a 9-kb species, which was approximately 0.7 kb shorter than full-length viral DNA obtained from Hirt supernatants of freshly infected cells (Fig. 3, lane labeled LMW HIV DNA). As previously described for Moloney murine leukemia virus (15), this 9-kb product is likely an almost complete minus DNA strand, in which the 39 extremity corresponds to the primer binding site. As previously reported (5), subgenomic bands were also observed, with sizes and proportions varying with experimental conditions. This assay behaved linearly as the signal was 10-fold lower with 100 ng than with 1,000 ng of p24 HIV. Equivalent signal intensities were detected for HIV and HIVDnef virions, indicating that these viruses had the same capacities to synthesize minus-strand viral DNA in vitro. Analysis of virion-associated RT activity with both exogeFIG. 2. Protein and RNA contents of HIV and HIVDnef virions. (a) Immunoprecipitation of viral proteins. Noninfected (lanes 0) and chronically HIV- and HIVDnef-infected U937 cells were labeled with [35S]methionine and [35S]cysteine for 16 h. Cell lysates, supernatants, and virus particles pelleted by ultracentrifugatrion were immunoprecipitated with a human polyclonal anti-HIV-1 serum and analyzed by SDS-PAGE (gp41 was not recognized by this serum). Molecular weight markers (in kilodaltons) are indicated on the right. (b) RNA content of HIV and HIVDnef particles. Supernatants and pelleted virus particles from HIV- and HIVDnef-infected HUT78 cells were adjusted for equivalent amounts of p24 before RNA extraction. Serial dilutions of the samples were dot blotted onto a nylon membrane, which was then hybridized with an HIV probe, washed, and autoradiographed. One experiment representative of three is shown.

that Nef did not modify the incorporation of viral genomic RNA into virus particles. Viral DNA corresponding to partial and heterogenous reverse transcripts can be detected in virions (31). We examined whether the levels of virion-associated viral DNA differed between HIV and HIVDnef. DNA was extracted from virions purified by ultracentrifugation, and PCR amplification of viral DNA was performed as described previously (31). Equivalent levels of viral transcripts were detected in HIV and HIVDnef virions, indicating that the expression of Nef in producing cells does not modify the amount of viral DNA present in particles (data not shown). Analysis of virion-associated RT activity. The RT activity present in HIV and HIVDnef virions was determined by using either an exogenous or endogenous template. In the exogenous assay, polymerization initiated from an oligo(dT) primer, with synthetic poly(rA) as the template. Supernatants collected 3 days after infection of MT4 cells, in which RPIs were 283 and 46 PFU/ng of p24 for HIV and HIVDnef, respectively, were adjusted for equivalent amounts of p24, and virus particles were pelleted. Equivalent amounts of 32P-labeled poly(dT) were synthesized in the presence of HIV and HIVDnef supernatants and viral pellets (Fig. 3a). Similar results were obtained with viruses produced in chronically infected HUT78 or U937 cells (data not shown).

FIG. 3. Analysis of virion-associated RT activity. (a) Exogenous RT activity. Supernatants and pelleted virus particles from HIV- and HIVDnef-infected MT4 cells (day 3 postinfection) were adjusted for equivalent amounts of p24 and assayed for exogenous RT activity by using oligo(dT) as a primer and poly(rA) as a template for the enzyme. Serial dilutions of the viral samples were performed to verify the linearity of the assay. One experiment representative of two is shown. (b) Endogenous RT assay. Supernatants from MT4 cells acutely infected with HIV and HIVDnef were adjusted for the indicated p24 amounts and incubated for 2 h at 418C with the indicated concentrations of deoxynucleoside triphosphates (dNTPs) and NP-40, in the presence of MgCl2 and EGTA. DNA was then extracted, and the samples were denatured in 200 mM NaOH before electrophoresis, Southern blotting, and hybridization with a 32P labeled 1.9-kb MscI fragment from the pol region of pNL43. Nondigested Hirt supernatant from a 24-h infection of MT4 cells with HIV was also analyzed in order to visualize full-length viral DNA (lane LMW [low-molecular-weight] HIV DNA). Molecular mass markers (in kilobases) are indicated on the left.

VOL. 69, 1995

HIV-1 Nef INCREASES EFFICIENCY OF RT

4057

FIG. 4. Cell surface binding of HIV and HIVDnef virions. MT4 cells were incubated for 1 h at 148C with either supernatants or pelleted virus particles from HIVand HIVDnef-infected MT4 cells that were adjusted for equivalent p24 contents (600 ng per sample). The cells were then washed, and gp120 bound to the surface was measured by cytofluorometry with the 110.H anti-gp120 monoclonal antibody and a phycoerythrin-labeled goat anti-mouse antibody. Cells which had not been incubated with virus were similarly stained and used as a negative control (CTRL curves). One experiment representative of three is shown.

nous and endogenous templates strongly suggested that Nef did not modify the capacity of virus particles to initiate and perform reverse transcription. Binding of HIV and HIVDnef virions. The interactions of HIV and HIVDnef virions with cell surface CD4 were examined by quantifying gp120 molecules bound on target cells by cytofluorometry (25). Supernatants and viral pellets from HIVand HIVDnef-infected MT4 cells in which the RPIs were 283 and 46 PFU/ng of p24, respectively (Fig. 4), or from HUT78 cells (data not shown) were adjusted for equivalent amounts of p24 (600 ng per sample) and incubated with noninfected MT4 cells for 60 min at 48C. Surface-bound gp120 molecules were then revealed with anti-gp120 monoclonal antibodies (Fig. 4). Equivalent fluorescence intensities were measured for both HIV and HIVDnef supernatants or viral pellets. Similar results were obtained when CEM cells were used as targets (data not shown). These data showed that the expression of Nef in virusproducing cells did not modify the ability of HIV particles to bind CD4 receptors on target cells. Synthesis of viral DNA in infected cells. We have examined the reverse transcription and the nuclear import of preintegrative viral DNA in infected cells. CEM cells were infected by cocultivation with HUT78 or U937 cells chronically infected with HIV and HIVDnef, and MT4 cells were infected by incubation with HIV or HIVDnef cell-free supernatants adjusted for equivalent p24 contents. Nonintegrated viral DNA was extracted by the Hirt procedure and was analyzed 24 h after infection, before the occurrence of secondary cycles of virus replication. Viral DNA molecules undergoing circularization after their transport to the nucleus were used as a marker to monitor the nuclear import of the preintegrative complex (6). Low-molecular-weight DNA was analyzed by Southern blot with a probe complementary to the HIV pol region (Fig. 5). Samples were digested with EcoRI, which produced diagnostic fragments with sizes of 5.7 and 9.1 kb from linear DNA (L) and DNA containing one LTR circle (C), respectively. No signal was detected when target cells were treated with zidovudine, indicating that the hybridizing DNA was actually de novo synthesized during the infection period (data not shown). The overall amounts of viral DNA synthesized during RT were 5to 10-fold higher for HIV than for HIVDnef. This was observed in both CEM and MT4 cells infected either with cellfree supernatants or by cocultivation. This experiment showed that the expression of Nef in virus-producing cells resulted in larger amounts of viral DNA accumulating in target cells.

Since the ratios of circular to total (circular plus linear) forms of viral DNA were equivalent for HIV and HIVDnef (40%), we concluded that Nef did not affect the nuclear import of preintegration complexes. DISCUSSION We have measured the infectivity of Nef-encoding and Nefdeficient HIV particles produced in different immortalized cell lines. Our results showed that the expression of Nef in virusproducing cells is associated with an enhanced capacity of the virions to infect target cells. Analysis of HIV and HIVDNef particles revealed no obvious difference in the structural compositions of the virions and in their RT activities. However, the reverse transcription process in target cells appeared markedly

FIG. 5. Synthesis of viral DNA in infected cells. (a) CEM cells were infected by cocultivation with either HIV or HIVDnef chronically infected HUT78 or U937 cells (at a ratio of five target cells for one chronically infected cell). (b) MT4 cells were infected with equal p24 amounts of HIV and HIVDnef from acutely infected MT4 cells. Twenty-four hours following infection, low-molecular-weight DNA was prepared by Hirt extraction, EcoRI digested, and analyzed by Southern blot with a 32P-labeled 1.9-kb MscI fragment from the pol region of pNL43 as a probe. The DNA content from approximatively 4 3 106 target cells was analyzed in each lane. EcoRI digestion of unintegrated linear HIV DNA generates a 5.7-kb fragment (L) corresponding to the 59 half of the genome, which is recognized by the MscI fragment. Circular molecules containing one LTR generate a 9.1-kb fragment (C). The ratios of circular/total (circular plus linear) viral DNA species measured by using a PhosphorImager were 40% for HIV and HIVDnef. The total amount of cellular low-molecular-weight DNA present in each sample was measured by hybridizing the same blots with the mitochondrial gene cytochrome b (cyt. b) as a probe (lower panels).

4058

SCHWARTZ ET AL.

less efficient for HIVDnef, whereas the nuclear import of viral DNA products was not impaired. The analysis of the virion infectivity was performed by calculating an RPI as the ratio of productive infectious events to the amount of p24 present in each viral sample. The productive infectious events were determined by using the P4 cell line in which the expression of a b-galactosidase gene depends on the transactivation of the HIV LTR by Tat (8). X-Gal staining of P4 cells 24 h after incubation with a Tat-encoding HIV stock allows a precise quantification of the number of virus particles having completed a single cycle of replication. RPIs were 3- to 10-fold higher for HIV than for HIVDnef regardless of the lymphocytic or monocytic immortalized cell line used to produce the viruses. Similar results were obtained for viruses produced during acute or chronic infection. These data showed that the positive effect of Nef on virus particle infectivity is not restricted to primary T cells and does not depend on the type of producer cells or the number of virus replication cycles achieved. It was previously observed that the Nef-positive effect on virus replication was restricted to primary lymphocytes or infection of immortalized cells with low viral inputs (7, 11, 23, 28). It is likely that the high efficiency of laboratory-adapted HIV strains to propagate in T-cell lines has masked weak differences in particle infectivities. The use of the P4 cell line, in which single hit infection events are investigated, and the calculation of the RPI allowed us to reveal differences in a broad spectrum of infection events and to show that enhanced infectivity is a general feature of HIV particles synthesized in the presence of Nef. Since Nef modifies the intracellular trafficking of the HIV receptor, CD4 (2, 14, 16, 26, 27), it was tempting to associate its effect on particle infectivity with its capacity to stimulate CD4 endocytosis. We examined this issue in virus-producing cells exhibiting a modified CD4 molecule (GpiCD4) (21) that was not internalized by Nef. The RPIs of HIV and HIVDnef produced in GpiCD4 cells differed to the same extent as those of viruses produced in CD41 cells, confirming that the expression of CD4 is not required for the higher infectivity associated with Nef (23). We could detect no obvious difference in the compositions of HIV particles produced in the presence and in the absence of Nef. Equivalent ratios of capsid to envelope proteins were found, and the RNA contents were not different. We investigated the reverse transcription capacity of HIV and HIVDnef particles by using both exogenous and endogenous substrates. This latter assay explores not only the catalytic activity of the polymerase and RNase H complex but also the presence of a functional tRNA primer and the capacity of the encapsidated genome to serve as a substrate. Both assays gave equivalent results with HIV and HIVDnef particles. We also examined the capacity of the virions produced in the presence or in the absence of Nef to bind CD4, the cellular receptor for entry of HIV particles. No differences in CD4 binding were detected between HIV and HIVDnef, indicating that Nef did not affect the recognition of target cells by HIV particles. This was consistent with the detection of similar amounts of gp120 in virions. Whereas we failed to detect structural differences between HIV and HIVDnef particles, the behavior of these viruses differed after entry into target cells, where the synthesis of viral DNA appeared more efficient for the wild type than for the nef mutant virus. This difference alone was sufficient to account for the enhanced infectivity of the wild-type virus. It was restricted to the reverse transcription process itself and did not influence the ability of synthesized DNAs to be transported to the nucleus. The altered reverse transcription of HIVDnef in target

J. VIROL.

cells contrasted with the capacity of HIVDnef particles to perform reverse transcription efficiently when assayed in a cellfree environment. This observation strongly suggests that the intracellular environment in which HIVDnef performs the reverse transcription process differs from the environment found by HIV particles. Suboptimal conditions for reverse transcription of the HIVDnef genome may result either from inappropriate uncoating of the particle or from internalization in a cellular compartment where the catalytic activity of RT is altered or where the neosynthesized DNA is rapidly degraded. These different mechanisms may not be exclusive. Whatever they are, our data indicate that they operate in the target cell at one or several of the early stages of the viral cycle. How the expression of Nef in virus-producing cells modifies the behavior of the particles after entry into the target cell (24) is a provocative question. One possibility is that a low number of membrane-associated Nef molecules are incorporated into virus particles during the budding process. Until now, Nef has not been recognized as a component of the virion (4, 12). It might be worthwhile to reexamine this point by using moresensitive experimental approaches. It is also possible that Nef alters the incorporation of cellular components into particles, leading to a modified behavior in target cells (4, 13, 30). Alternatively, Nef could operate by imprinting special features (by posttranslational modification) of one or several of the viral components of the particle. ACKNOWLEDGMENTS We thank Franc¸ois Clavel for discussions and for critical reading of the manuscript, Franc¸oise Bachelerie for the gift of the pNL43 and pNL43-Xho plasmids, Robert Finberg for the gift of the HSB2GPiCD4 cells, and Franc¸ois Traincart for the gift of the 110.H monoclonal antibody. O.S. is a fellow of the association Pasteur-CANAM. This work was supported by grants from the Agence Nationale de Recherches sur le SIDA (ANRS) and the Pasteur Institute. REFERENCES 1. Adachi, A., H. E. Gendelman, S. Koenig, T. Folks, R. Willey, A. Rabson, and M. A. Martin. 1986. Production of acquired immunodeficiency syndromeassociated retrovirus in human and nonhuman cells transfected with an infectious molecular clone. J. Virol. 59:284–291. 2. Aiken, C., J. Konner, N. R. Landau, M. E. Lenburg, and D. Trono. 1994. Nef induces CD4 endocytosis: requirement for a critical dileucine motif in the membrane-proximal CD4 cytoplasmic domain. Cell 76:853–864. 3. Anderson, S. J., M. Lenburg, N. R. Landau, and J. V. Garcia. 1994. The cytoplasmic domain of CD4 is sufficient for its down-regulation from the cell surface by human immunodeficiency virus type 1 Nef. J. Virol. 68:3092–3101. 4. Arthur, L. O., J. W. Bess, R. C. Sowder II, R. E. Benveniste, D. L. Mann, J. C. Chermann, and L. E. Henderson. 1992. Cellular proteins to immunodeficiency viruses: implications for pathogenesis and vaccines. Science 258:1935– 1938. 5. Borroto-Esoda, K., and L. Boone. 1991. Equine infectious anemia virus and human immunodeficiency virus DNA synthesis in vitro: characterization of the endogenous reverse transcriptase reaction. J. Virol. 65:1952–1959. 6. Brown, P. O., B. Bowerman, H. E. Varmus, and J. M. Bishop. 1987. Correct integration of retroviral DNA in vitro. Cell 49:347–356. 7. Chowers, M. Y., C. A. Spina, T. J. Kwoh, N. J. Fitch, D. D. Richman, and J. C. Guatelli. 1994. Optimal infectivity in vitro of human immunodeficiency virus type 1 requires an intact nef gene. J. Virol. 68:2906–2914. 8. Clavel, F., and P. Charneau. 1994. Fusion from without directed by human immunodeficiency virus particles. J. Virol. 68:1179–1185. 9. Clavel, F., and J. M. Orenstein. 1990. A mutant of human immunodeficiency virus with reduced RNA packaging and abnormal particle morphology. J. Virol. 64:5230–5234. 10. Cullen, B. R. 1994. The role of Nef in the replication cycle of the human and simian immunodeficiency viruses. Virology 205:1–6. 11. de Ronde, A., B. Klaver, W. Keulen, L. Smit, and J. Goudsmit. 1992. Natural HIV-1 Nef accelerates virus replication in primary human lymphocytes. Virology 188:391–395. 12. Franchini, G., M. Robert-Guroff, J. Ghrayeb, N. Chang, and F. Wong-Staal. 1986. Cytoplasmic localisation of the HTLV III 39 orf in cultured T cells. Virology 155:593–599.

VOL. 69, 1995 13. Franke, E. K., H. En Hui Yuan, and J. Luban. 1994. Specific incorporation of cyclophilin A into HIV-1 virions. Nature (London) 372:359–362. 14. Garcia, J. V., and A. D. Miller. 1991. Serine phosphorylation-independent downregulation of cell-surface CD4 by nef. Nature (London) 350:508–511. 15. Gilboa, E., S. W. Mitra, S. Goff, and D. Baltimore. 1979. A detailed model of reverse transcription and tests of crucial aspects. Cell 18:93–100. 16. Guy, B., M. P. Kieny, Y. Rivie`re, C. Le Peuch, K. Dott, M. Girard, L. Montagnier, and J. P. Lecoq. 1987. HIV F/39orf encodes a phosphorylated GTP-binding protein resembling an oncogene product. Nature (London) 330:266–269. 17. Hirt, B. 1967. Selective extraction of polyoma DNA from infected mouse cell cultures. J. Mol. Biol. 26:365–369. 18. Kaminchik, J., N. Bashan, A. Itach, N. Sarver, M. Gorecki, and A. Panet. 1991. Genetic characterization of human immunodeficiency virus type 1 nef gene products translated in vitro and expressed in mammalian cells. J. Virol. 65:583–588. 19. Kestler, H. W., D. J. Ringler, K. Mori, D. L. Panicali, P. K. Sehgal, M. D. Daniel, and R. C. Desrosiers. 1991. Importance of the nef gene for maintenance of high virus loads and for development of AIDS. Cell 65:651–622. 20. Kim, S., K. Ikeuchi, R. Byrn, J. Groopman, and D. Baltimore. 1989. Lack of a negative influence on viral growth by the nef gene of human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. USA 86:9544–9548. 21. Marshall, W. L., E. S. Mittler, P. Avery, J. P. Lawrence, and R. W. Finberg. 1994. Glycosylphosphatidylinositol-anchored CD4 supports human immunodeficiency virus type 1 replication, but not cytopathic effect, in T-cell transfectants. J. Virol. 68:4039–4042. 22. Miller, M. D., M. B. Feinberg, and W. C. Greene. 1994. The HIV-1 nef gene acts as a positive viral infectivity factor. Trends Microbiol. 2:294–297. 23. Miller, M. D., M. T. Warmerdam, I. Gaston, W. C. Greene, and M. B. Feinberg. 1994. The human immunodeficiency virus-1 nef gene product: a positive factor for viral infection and replication in primary lymphocytes and

HIV-1 Nef INCREASES EFFICIENCY OF RT

4059

macrophages. J. Exp. Med. 179:101–113. 24. Miller, M. D., M. T. Warmerdam, K. A. Page, M. B. Feinberg, and W. C. Greene. 1995. Expression of the human immunodeficiency virus type 1 (HIV-1) nef gene during HIV-1 production increases progeny particle infectivity independently of gp160 or viral entry. J. Virol. 69:579–584. 25. Schwartz, O., M. Alizon, J. M. Heard, and O. Danos. 1994. Impairment of T cell receptor-dependent stimulation in CD41 lymphocytes after contact with membrane-bound HIV-1 envelope glycoprotein. Virology 198:360–365. 26. Schwartz, O., A. Dautry-Varsat, B. Goud, V. Mare´chal, A. Subtil, J.-M. Heard, and O. Danos. 1995. Human immunodeficiency virus type 1 Nef induces accumulation of CD4 in early endosomes. J. Virol. 69:528–533. 27. Schwartz, O., Y. Rivie`re, J.-M. Heard, and O. Danos. 1993. Reduced cell surface expression of processed human immunodeficiency virus type 1 envelope glycoprotein in the presence of Nef. J. Virol. 67:3274–3280. 28. Spina, C. A., T. J. Kwoh, M. Y. Chowers, J. C. Guatelli, and D. D. Richman. 1994. The importance of nef in the induction of human immunodeficiency virus type 1 replication from primary quiescent CD4 lymphocytes. J. Exp. Med. 179:115–123. 29. Terwilliger, E., J. G. Sodroski, C. A. Rosen, and W. A. Haseltine. 1986. Effects of mutations within the 39 orf open reading frame region of human T-cell lymphotrophic virus type III (HTLV-III/LAV) on replication and cytopathogenicity. J. Virol. 60:754–760. 30. Thali, M., A. Bukovsky, E. Kondo, B. Rosenwirth, C. T. Walsh, J. Sodroski, and H. G. Go ¨ttlinger. 1994. Functional association of cyclophilin A with HIV-1 virions. Nature (London) 372:363–365. 31. Trono, D. 1992. Partial reverse transcripts in virions from human immunodeficiency and murine leukemia viruses. J. Virol. 66:4893–4900. 32. Zazopoulos, E., and W. A. Haseltine. 1992. Mutational analysis of the human immunodeficiency virus type 1 Eli Nef function. Proc. Natl. Acad. Sci. USA 89:6634–6638.