A Mutation in Alpha Helix 3 of CA Renders Human Immunodeficiency Virus Type 1 Cyclosporin A Resistant and Dependent: Rescue by a Second-Site Substitution in a Distal Region of CA

Abstract

The replication of many isolates of human immunodeficiency virus type 1 (HIV-1) is enhanced by binding of the host cell protein cyclophilin A (CypA) to the viral capsid protein (CA). The immunosuppressive drug cyclosporine A (CsA) and its nonimmunosuppressive analogs bind with high affinity to CypA and inhibit HIV-1 replication. Previous studies have identified two mutations, A92E and G94D, in the CypA-binding loop of CA that confer the ability of HIV-1 to replicate in the presence of CsA. Interestingly, CsA stimulates the replication of HIV-1 mutants containing either the A92E or G94D substitution in some human cell lines. Here, we show that substitution of alanine for threonine at position 54 of CA (T54A) also confers HIV-1 resistance to and dependence on CsA. Like the previously identified CsA-resistant/dependent mutants, infection by the T54A mutant was stimulated by CsA in a target cell-specific manner. RNA interference-mediated reduction of CypA expression enhanced the permissiveness of HeLa cells to infection by the T54A mutant. A suppressor mutation, encoding a substitution of threonine for alanine at position 105 of CA (A105T), was identified through adaptation of the T54A mutant virus for growth in CEM cells. A105T rescued the impaired single-cycle infectivity and replication defects of both T54A and A92E mutants. These results indicate that CA determinants outside the CypA-binding loop can modulate the dependence of HIV-1 infection on CypA.

The capsid (CA) protein of human immunodeficiency virus type 1 (HIV-1) consists of two domains, with the N- and C-terminal domains (NTD and CTD) connected by a flexible linker. The NTD contains an N-terminal β-hairpin, seven α-helices, and an extended loop that binds to the cellular protein cyclophilin A (CypA) (20, 29). Cryoelectron microscopy studies have revealed that the NTD forms a hexameric lattice, with the CA CTD making dimeric contacts that connect each ring to its six nearest neighbors (25). Mutagenesis studies have shown that both the NTD and CTD are essential for capsid formation and particle assembly (14, 28, 31). During particle maturation, the CA protein condenses to form a conical core around the ribonucleoprotein complex. Mutations that alter HIV-1 core morphology also reduce infectivity (14, 28, 31, 33, 38). These observations suggest that proper formation of the conical HIV-1 core is essential for the early postentry events in HIV infection.

CypA is a cellular peptidylprolyl isomerase that binds to the HIV-1 CA NTD and is incorporated into virions through interaction with an exposed loop between helices 4 and 5 in HIV-1 and SIVcpz CA proteins (19, 27). Disruption of the CypA-CA interaction, via mutations in CA or addition of cyclosporine A (CsA), inhibits HIV-1 replication (18, 39). An initial model held that incorporation of CypA into HIV-1 particles is necessary for proper uncoating in target cells (26). More recent studies have demonstrated that a requirement for the CypA-CA interaction is manifested following entry of the core into target cells, and incorporation of CypA into virions appears to be biologically irrelevant (21, 35). Despite many years of study, the precise role of CypA in promoting HIV-1 infection remains obscure.

Previous studies have shown that substitutions in the CypA-binding loop, A92E and G94D, confer HIV-1 resistance to CsA (1, 6). Interestingly, infection by these mutants is enhanced by CsA in some cell lines, such as HeLa and H9, but not others, such as HOS and Jurkat (1, 21). Positions 92 and 94 reside in the CypA-binding loop in CA, but these mutations do not affect CypA-CA binding (6), which suggests that CsA resistance is independent of this interaction. The CsA dependence implies that CypA binding to CA has a detrimental effect on infection by these mutants in some cell types.

In the present study, we analyzed in detail the phenotype of a poorly infectious HIV-1 mutant, T54A, encoding a Thr-to-Ala substitution in helix 3 of the CA NTD. We found that the effects of the T54A substitution resemble those of previously characterized CypA-binding loop mutations A92E and G94D.

MATERIALS AND METHODS

Cells and viruses.

293T, HeLa-CD4/LTR-lacZ (HeLa-P4), and HeLa-CypA KD cells (35) were cultured in Dulbecco's modified Eagle medium (Cellgro) supplemented with 10% fetal bovine serum, penicillin (50 IU/ml), and streptomycin (50 μg/ml) at 37°C with 5% CO₂. CEM and H9 cells were cultured in RPMI 1640 medium with the same supplements. The wild-type HIV-1 proviral DNA construct R9, encoding full-length open reading frames for all HIV-1 structural and accessory genes, was used for these studies. The point mutation T54A in the CA region of R9 was previously reported (41) and was the generous gift of Wes Sundquist. In some experiments, HIV-GFP, an envelope-defective HIV-1 reporter virus clone encoding green fluorescent protein (GFP) in place of Nef (22), was employed. Viruses were produced by calcium phosphate transfection of 293T cells (20 μg of plasmid DNA per 2 × 10⁶ cells) as previously described (12). Vesicular stomatitis virus glycoprotein (VSV-G)-pseudotyped reporter virus particles were produced by cotransfection of HIV-GFP plasmid DNA with the VSV-G expression plasmid pHCMV-G (44). One day after transfection, the culture supernatants were harvested and clarified by being passed through 0.45-μm-pore-size filters, and aliquots were frozen at −80°C. The CA contents of the virus stocks were quantified by p24 enzyme-linked immunosorbent assay (ELISA), as previously described (42). The HeLa-P4 cell line, a HeLa cell clone engineered to express CD4 and an integrated long terminal repeat (LTR)-lacZ reporter cassette (11), was used to quantify HIV-1 infectivity, as previously described (11), with the following modifications. HIV-1 stocks were serially diluted in culture medium, and samples (0.125 ml) were used to inoculate HeLa-P4 target cells seeded the day before (20,000 cells per well in 48-well plates). Two hours after inoculation, the cultures were supplemented with additional medium (0.5 ml) and cultured for another 48 h prior to being stained with 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal) to detect infected cells. To determine the number of infected cells per well, individual wells were visualized using a Toshiba Imagemaster CCD Color Camera (model no. IK-6365) equipped with a Navitar Macro Zoom 18- to 108-mm lens, and images were captured with a Power Macintosh 8500 computer. Blue cells were counted using NIH Image 1.62 software in the particle-counting mode. Infections were performed in triplicate, and only values within the linear range of the infection assay (up to 1,000 blue cells per well) were used to calculate infectivity. Infections with HIV-GFP reporter viruses were analyzed by flow cytometric analysis of GFP expression as previously described (16).

Viral-replication assay.

Viral growth was determined by inoculation of cultures of CEM cells (2 × 10⁵ cells in 0.2 ml) or H9 cells with quantities of virus corresponding to 2 ng of p24 antigen. Prior to addition to tissue culture medium, CsA was dissolved in dimethyl sulfoxide to produce a stock concentration of 1 mM. The stock was added to the medium to achieve a final concentration of 2.5 μM or 5 μM when needed. Mock-treated controls in each experiment received the same final concentrations of dimethyl sulfoxide as those in the treated cultures. Every 2 days, samples of culture supernatants (100 μl) were withdrawn and replaced with an equal volume of fresh medium. Reverse transcriptase activity in culture supernatants was quantified as previously described (3).

Virus-cell fusion assay.

The β-lactamase (BlaM)-Vpr HIV-1 fusion assay was performed essentially as previously described (43). Quantities of wild-type and mutant reporter viruses were normalized by p24 content and used to inoculate P4 target cells for 2 h at 37°C. The cells were then loaded with the CCF2-AM fluorogenic substrate overnight at room temperature. The supernatant was removed and replaced with phosphate-buffered saline (PBS). Cellular fluorescence was determined in a microplate fluorometer. The background levels of blue (no virus) and green (no cells or virus) fluorescence were determined at 410 nm and 520 nm, respectively, and were subtracted from the experimental samples. Fluorescence ratios were calculated for each well. For each virus dilution, triplicate determinations were performed, and values typically agreed to within 10%.

Quantitative analysis of HIV-1 reverse transcription in target cells.

HIV-1 reverse transcription in target cells was quantified essentially as described previously (13). One day prior to infection, 100,000 HeLa-P4 cells per well were seeded in 12-well plates. Virus stocks were treated with 20 μg/ml of DNase I and 10 mM MgCl₂ at 37°C for 1 h to remove contaminating plasmid DNA. Inocula were normalized by p24 content to 100 ng per well for HIV-1 in medium containing DEAE-dextran (10 μg/ml). At 8 h postinfection, the cells were washed with 1 ml of PBS and then detached with trypsin. The trypsin was deactivated by the addition of 0.5 ml of complete medium, and the cells were pelleted and washed once with 500 μl of PBS. The cell pellets were resuspended in 200 μl of PBS, and DNA was isolated using a DNeasy kit (QIAGEN) according to the manufacturer's instructions. Viral DNA was quantified by real-time PCR using an MX-3000p thermocycler (Stratagene) utilizing TaqMan chemistry. Reverse transcription products (U5-Gag) were detected using the forward primer MH531 (5′-TGTGTGCCCGTCTGTTGTGT-3′) and the reverse primer MH532 (5′-GAGTCCTGCGTCGAGAGAGC-3′), with the probe LRT-P (5′-6-carboxyfluorescein[FAM]-CAGTGGCGCCCGAACAGGGA-6-carboxytetramethylrhodamine [TAMRA]-3′) as previously described (8). Thermal-cycling conditions were 2 min at 50°C and 10 min at 95°C, followed by 40 cycles of 95°C for 15 s and 60°C for 60 s.

Assay of HIV-1 nuclear targeting.

Assay of HIV-1 nuclear targeting was performed essentially as previously described (13). DNA from acutely infected cells was isolated at 24 h postinfection by using the DNeasy kit (QIAGEN) and analyzed for the presence of two-LTR circles. Two-LTR circle DNA was detected by quantitative real-time PCR utilizing primers and a TaqMan probe specific for the LTR-LTR junction. The forward primer MH535 (5′-ACTAGGGAACCCACTGCTTAAG-3′), the reverse primer MH536 (5′-TCCACAGATCAAGGATATCTTGTC-3′), and the two-LTR probe MH603 (5′-[FAM]-ACACTACTTGAAGCACTCAAGGCAAGCTTT-[TAMRA]-3′) were used. Thermal-cycling conditions were 2 min at 50°C, 10 min at 95°C, and 40 cycles of 95°C for 15 s and 60°C for 90 s.

Isolation of HIV-1 cores.

Cores were isolated from concentrated virions as previously described (24). Briefly, supernatants from transfected 293T cells were filtered to remove cellular debris, and HIV-1 particles were concentrated by ultracentrifugation (100,000 × g for 3 h at 4°C) through a cushion of 20% (wt/vol) sucrose in STE buffer (10 mM Tris-HCl [pH 7.4], 100 mM NaCl, 1 mM EDTA). The viral pellets were resuspended in STE buffer (0.4 ml), and the concentrated virions were subjected to ultracentrifugation (130,000 × g for 16 h at 4°C) through a layer of 1% Triton X-100 into a linear sucrose density gradient (10 ml of STE buffer containing 30 to 70% sucrose). Fractions (1 ml) were collected from the top of the gradient and analyzed for CA content by p24 ELISA. The yield of cores was determined as the CA content (as determined by p24 ELISA) in the peak fractions of cores as a percentage of the total CA content in the gradient.

HIV-1 adaptation studies.

Virus stocks were normalized for reverse transcriptase (RT) activity and used to infect CEM cells. Virus supernatant was harvested at the peak of RT activity and normalized. Two subsequent passages of infection in fresh CEM cells were performed until accelerated growth was observed. Cells were harvested immediately following the peak of RT activity from the third passage. Proviral DNA was purified with the DNeasy kit (QIAGEN), a 900-bp fragment spanning the BssHII and SpeI sites (encoding the matrix protein and most of the NTD of CA) was amplified by PCR, and a BssHII-SpeI restriction fragment was transferred into R9. Mutants A105T and A92E/A105T were created by PCR-based mutagenesis in the R9 background. The A105T single mutant was created by using the sense primer 5′-CAAGATTTAAATACCATGCTAAACACAGT-3′ and the antisense primer 5′-TAGCATGGTATTTAAATCTTGTGGGGTGG-3′ with the R9-T54A/A105T plasmid as a template. A92E/A105T was created by using the sense primer 5′-GGGCCTATTGAACCAGGCCAGATGAGAGA-3′ and the antisense primer 5′-CTGGCCTGGTTCAATAGGCCCTGCATGCA-3′ with the R9-A105T plasmid as a template. The envelope-defective HIV-GFP/T54A reporter virus construct was created by transferring the BssHII-SpeI restriction fragment from R9.T54A into HIV-GFP. The sequences of the replaced regions of all viral constructs were experimentally verified.

Analysis of HIV-1 infection by flow cytometry.

The wild-type HIV-GFP or mutant HIV-GFP/T54A (2) pseudotyped by VSV-G was used to assay infection of HeLa-P4 or HOS cells. Cultures (20,000 cells per well in 12-well plates) were inoculated with various concentrations of reporter viruses in the presence of Polybrene (8 μg/ml) in a total volume of 300 μl. One day later, complete medium (1 ml) was added. Two days after infection, the cells were detached using trypsin and fixed by the addition of an equal volume of PBS containing 4% paraformaldehyde. GFP expression was quantified by flow cytometry by using a FACSCalibur instrument (Becton Dickinson), and the percentage of GFP-expressing cells was quantified with Cellquest software. A minimum of 5,000 cells were analyzed for each sample.

RESULTS

HIV-1 containing the T54A substitution in CA is poorly infectious and is impaired for replication.

A previous study reported the initial characterization of a series of HIV-1 mutants containing different alanine-scanning surface mutations in the N terminus of CA (41). One of these mutants, T54A, in helix 3 of CA, was competent for viral assembly and release but was moderately reduced in infectivity. To verify and further characterize the infectivity impairment associated with this mutant, we generated particles from wild-type and T54A mutant proviruses by transfection of 293T cells and assayed infectivity by titration on HeLa-P4 indicator cells. Similar yields of virus particles were obtained for the wild-type and mutant viruses, as determined by p24 release into the supernatant (data not shown). The specific infectivity of the T54A mutant, determined as the number of infected cells per ng of p24 in the inoculum, was reduced 10-fold relative to the wild type (Fig. 1A). To evaluate the effects of the T54A mutation on the efficiency of HIV-1 replication, we studied the kinetics of viral spread in the CEM human T-cell line. By contrast to the wild type, the T54A mutant particles failed to replicate within the 3-week culture period (Fig. 1B).

FIG. 1.

The HIV-1 T54A mutant is impaired at an early postentry step of infection. (A) Single-cycle infectivity was assayed in HeLa-P4 target cells. Infectivity was determined as the number of infected cells per ng of p24 in the inoculum, and values are expressed as percentage of wild-type (WT) HIV-1 infectivity. The results shown are the mean values of three independent experiments, with error bars representing 1 standard deviation. (B) Replication kinetics of T54A and WT viruses in CEM cells. Cultures were inoculated with equal quantities of WT and T54A mutant viruses. Supernatants were sampled on the days indicated and assayed for RT activity. Shown is a representative growth curve for each virus from duplicate cultures. (C) BlaM reporter assays of HIV-1 fusion. The results are expressed as the extent of conversion of the cell-permeable BlaM substrate CCF2-AM and are representative of two independent experiments. (D and E) Real-time PCR quantitation of HIV-1 DNA synthesis in target cells. Cells were inoculated with DNAse I-treated virus stocks, and the total DNA was harvested at the indicated time postinoculation and analyzed for synthesis of late products of reverse transcription (D) and for two-LTR circular forms (E). The results shown are the mean values of three independent experiments, with error bars representing 1 standard deviation. (F) T54A mutant virions contain unstable capsids. Wild type (WT) and T54A mutant particles were concentrated and subjected to equilibrium ultracentrifugation through a layer of 1% Triton X-100 into a sucrose density gradient. Fractions were collected, and the p24 content was determined by ELISA. The yield of CA present in the fractions corresponding to cores was calculated as a percentage of the total CA quantity in each gradient. The values shown are the means of three independent experiments, with error bars representing 1 standard deviation.

The T54A mutant is impaired at an early postentry step in infection.

Previous studies have demonstrated that HIV-1 particles containing substitutions in CA are frequently impaired in early steps in infection (15, 17, 37). To further define the defect associated with the T54A mutation, we first analyzed entry of the mutant particles using the β-lactamase-based reporter fusion assay (10). Wild-type and T54A mutant particles were generated by cotranfection of 293T cells with proviral DNA and a BlaM-Vpr expression construct. The BlaM-carrying HIV-1 particles were then incubated with HeLa-P4 cells, and fusion was detected by quantifying the conversion of a cell-permeable fluorescent BlaM substrate. The wild-type and T54A mutant viruses exhibited nearly equivalent activities in this assay (Fig. 1C), indicating that the T54A mutation does not impair the efficiency of HIV-1 fusion.

We next asked whether the T54A mutant is competent for reverse transcription in target cells. To quantify the rate and extent of reverse transcription, wild-type and T54A mutant particles were added to cultures of HeLa-P4 cells, and the cells were cultured for various times and harvested. DNA was isolated from cell pellets, and HIV-1 DNA was subsequently quantified by real-time PCR using primers and probes specific for late products of DNA synthesis and the two-LTR circular form, a marker for nuclear import. Relative to wild type HIV-1, the synthesis of late reverse transcripts was only moderately reduced for the T54A mutant (Fig. 1D). Furthermore, the accumulation of two-LTR circles was reduced to a similar extent for the mutant (Fig. 1E). These results indicate that the T54A mutant particles are impaired, albeit only moderately, for reverse transcription in target cells and that the reverse-transcribed products efficiently enter the nucleus.

T54A mutant particles contain unstable capsids.

Previous studies in our laboratory and others have shown that mutations that alter the stability of the HIV-1 capsid are associated with impaired infection, likely due to impaired uncoating in target cells (15, 17, 37). To determine whether the T54A mutation alters the stability of the viral capsid, we quantified the yield of CA protein associated with cores released from wild-type and T54A mutant virions upon treatment with nonionic detergent and equilibrium density gradient sedimentation. For the wild-type virus, approximately 12% of the CA protein was detected in fractions containing HIV-1 cores (Fig. 1F). For the T54A mutant, this value was reduced to approximately 3% of virion-associated CA. The decreased yield of CA in these fractions suggests that the mutant virions contain cores with capsids that are less stable than those of the wild type.

A second-site suppressor mutation, A105T, restores replication and single-cycle infectivity to the T54A mutant.

To further analyze the defect induced by the T54A mutation, we attempted to recover a pseudorevertant by serial passage of the T54A mutant virus in CEM cells. Virus supernatants were harvested immediately following the peak of growth of the T54A mutant virions and were inoculated into fresh cells, and the cultures were maintained until reemergence of HIV-1 in the cultures. To determine whether phenotypic reversion was a result of acquisition of compensatory HIV-1 mutations, DNA was purified from CEM cells harvested near the peak of growth, and a DNA product fragment spanning the matrix protein and most of the N terminus of the CA coding region was generated by PCR, digested with BssHII and SpeI, and inserted into the corresponding region of the wild-type R9 plasmid. Sequencing of several clones identified a mutation at codon 105 resulting in a substitution of Thr for Ala, in addition to the original T54A mutation. To determine whether the A105T substitution was responsible for the accelerated growth kinetics observed upon passage of T54A-derived virus, we assayed replication of the double mutant in CEM cells. The T54A/A105T double mutant exhibited accelerated growth relative to the original T54A mutant yet was significantly delayed relative to the wild type (Fig. 2A). We also tested whether the A105T mutation could relieve the T54A infectivity impairment in the single-cycle infection assay. The results showed that A105T enhanced the infectivity of T54A to nearly that of the wild-type virus (Fig. 2B). As a control, we also constructed the A105T single mutant and tested its infectivity, which was equivalent to that of the wild type (data not shown).

FIG. 2.

Rescue of T54A replication by a second-site mutation. (A) Replication kinetics of wild-type and mutant HIV-1 in CEM cells. Samples were collected on the days indicated and analyzed for exogenous RT activity. Shown are the average RT values obtained from duplicate parallel cultures. (B) Single-cycle infectivity was determined by titration of viruses on HeLa-P4 reporter target cells. The results shown are the mean values of three independent experiments, with error bars representing 1 standard deviation.

Infection by T54A, but not the double mutant T54A/A105T, is enhanced by CsA.

Previous studies identified two mutations, A92E and G94D, which render HIV-1 capable of replicating in the presence of CsA (1). Interestingly, the replication of these mutants is dependent on the presence of CsA in some cell lines, such as H9 (21, 35). An early study reported that adaptation of the A92E mutant for growth in the absence of CsA resulted in the acquisition of additional substitutions, including V86A and A105T (1). Because the A105T mutation restored the ability of the T54A mutant to replicate in T cells, we hypothesized that T54A might also exhibit a CsA-resistant/dependent phenotype. To test this, we first analyzed the effects of addition of CsA on infection by the T54A mutant virus in HeLa-P4 cells. The results showed that infection by the T54A mutant increased by approximately threefold in the presence of CsA. The stimulating effect of CsA on infection was lost upon addition of the second-site compensatory mutation A105T, which rendered the virus CsA sensitive (Fig. 3). Since the previously characterized CsA-resistant mutant A92E exhibits CsA dependence in HeLa or H9 cells (1, 45), we next assessed the effect of CsA on the replication of T54A and T54A/A105T in H9 cells. In H9 cells, the wild-type virus replicated efficiently, with a peak in p24 production at day 8 postinoculation (Fig. 4A). CsA-containing cultures exhibited a moderate reduction in wild-type virus yield in the cultures. By contrast, T54A exhibited only low levels of replication, but its growth was markedly stimulated by CsA, resulting in a peak at day 12 (Fig. 4B). By contrast, the T54A/A105T double mutant replicated moderately efficiently in the absence of CsA, and its replication was partially inhibited by CsA (Fig. 4C). The kinetics of the double mutant resembled that of the wild type, but the virus yield was approximately one-fourth that of the wild type. Collectively, these results indicate that infection and replication of the T54A mutant are stimulated by CsA.

FIG. 3.

Infection by T54A is enhanced by CsA. An infectivity assay was performed in HeLa-P4 cells in the presence or absence of CsA (5 μM) during the infection. The results shown are the mean values from three independent experiments and are expressed as a percentage of the wild-type (WT) infectivity value from cultures lacking CsA.

FIG. 4.

CsA enhances replication of T54A in H9 cells. Replication of wild type (WT) (A), T54A (B), and T54A/A105T (C) HIV-1 in H9 cultures. CsA (2.5 μM) was present in the media of the indicated cultures for the duration of the experiment. Samples were collected on the days indicated and analyzed for RT activity. Shown are the average RT values obtained from duplicate parallel cultures.

A105T complements the CsA-resistant/dependent mutant A92E.

Our results demonstrated that CsA enhances infection of viruses containing mutations in two distinct domains of CA: T54 in helix 3 and A92E in the CypA-binding loop. Our results indicate that addition of the A105T mutation rescues the impaired infectivity of the T54A mutant. The A105T substitution has also been observed in a virus resulting from adaptation of the CsA-resistant/dependent A92E mutant for growth in HeLa-CD4 cells in the absence of CsA (1). An additional mutation, V86A, was also observed in that study; however, it was not determined whether one or both mutations were required to rescue the replication of the A92E mutant virus. To test whether the A105T mutation is sufficient to complement the A92E infectivity defect, we constructed the double mutant A92E/A105T and tested the virus for single-cycle infectivity in HeLa-P4 cells. The double mutant was nearly as infectious as wild-type HIV-1 in the absence of CsA (data not shown). The A105T mutation also rescued the ability of the A92E mutant to replicate in H9 cells in the absence of CsA, and replication of the A92E/A105T virus was not dependent on CsA (Fig. 5). Thus, the A105T mutation enhanced the replication of both T54A and A92E HIV-1 mutants.

FIG. 5.

A105T rescues A92E replication in H9 cells. H9 cultures were inoculated with (A) A92E and (B) the double mutant A92E/A105T. Cultures were maintained with and without CsA at 2.5 μM. Samples were collected on the days indicated and analyzed for RT activity. Shown are the average RT values obtained from duplicate parallel cultures.

The enhancing effect of CsA on infection by T54A mutant HIV-1 is dependent on the target cell.

Previous studies have demonstrated that the ability of CsA to stimulate infection by the A92E mutant depends on the identity of the target cells (21, 35). Our findings that T54A replication is enhanced by CsA, and that the A105T mutation restores replication to T54A and A92E mutant viruses, suggest that these mutants have similar phenotypes. To further test this hypothesis, we assayed infection of different adherent human cells, HeLa-P4 and HOS cells, with VSV-G-pseudotyped reporter viruses. In HeLa-P4 cells, CsA stimulated infection by the T54A mutant virus; by contrast, in HOS cells, a slight inhibition of the mutant was observed in the presence of CsA (Fig. 6). As previously reported (35), infection of HeLa-P4 cells by wild-type HIV-1 was not affected by CsA, whereas infection of HOS cells was inhibited by three- to fourfold (Fig. 6). In additional studies, we confirmed that the infectivity of the A92E mutant virus was enhanced by CsA in HeLa-P4 but not in HOS cells, as previously reported (C. Song and C. Aiken, unpublished data). These results indicate that the infectivity of the T54A mutant, like that of A92E, is enhanced by CsA in a target cell-specific manner.

FIG. 6.

CsA enhances infection by T54A in a target cell-dependent manner. Titration of VSV-G-pseudotyped HIV-GFP in HeLa-P4 or HOS target cells in the presence and absence of CsA (5 μM). Two days later, the percentage of infected cells was determined by flow cytometric analysis of GFP expression. The percentage of infected GFP-positive cells is plotted as a function of the input virus dose.

To determine whether the enhancing effect of CsA on infection by the T54A HIV-1 mutant is mediated through CypA, we assayed infection of HeLa cells in which CypA expression was inhibited by a specific short-hairpin RNA (HeLa-CypA KD). These cells were previously shown to exhibit enhanced permissiveness to CsA-resistant/dependent mutants (21, 35). HeLa-CypA KD cells were approximately 10-fold more permissive to infection by the T54A mutant than control HeLa cells (Fig. 7). Addition of CsA did not further enhance infection in the CypA KD cells. By contrast, the wild-type virus was equally infectious on HeLa and HeLa-CypA KD cells, and CsA had little to no effect on this virus in both cell types. Collectively, these results indicate that HeLa cells restrict infection by the T54A mutant virus by a mechanism that depends on CypA.

FIG. 7.

Reduction of CypA expression enhances cellular permissiveness to infection by T54A mutant virions. VSV-G-pseudotyped wild-type and T54A and A92E mutant viruses encoding GFP were titrated onto control HeLa cells (top) and HeLa-CypA KD cells (bottom) in the presence and absence of 5 μM CsA. Infected cells were quantified by flow cytometry for GFP expression.

DISCUSSION

In this study, we show that a HIV-1 mutant with a substitution of Ala for Thr54 in α-helix 3 of CA is poorly infectious due to an early postentry defect in the virus life cycle. Assays of viral DNA synthesis in target cells revealed a twofold reduction of reverse transcription of the mutant relative to the wild type, with no apparent defect in nuclear entry. The magnitude of the reverse transcription impairment does not seem to account for the 10-fold reduction in infectivity associated with the mutant, suggesting the possibility of an additional integration defect, as recently reported for another CA mutant (13).

Addition of CsA-enhanced infection by the T54A mutant particles in a target cell-dependent manner. This result was unexpected, since the previously identified CsA-resistant/dependent mutants both mapped to the CypA-binding domain of CA, and T54A resides on the outer face of helix 3. We also identified the second-site mutation A105T that complements the T54A infectivity impairment, as well as that of the previously characterized CsA-resistant/dependent mutant A92E. Like the two previously identified CsA dependent/resistant mutants, the enhancing effect of CsA on infection by T54A was observed in HeLa, but not HOS, target cells. CypA-deficient HeLa cells were permissive for infection by the T54A mutant virus, but the mutant virus was still slightly impaired relative to wild-type HIV-1. Collectively, our results indicate that infection by the T54A mutant is restricted at a postentry stage of infection by a CypA-dependent mechanism. Because the addition of CsA did not restore the replication of the mutant to wild-type levels in H9 cells, we cannot exclude the possibility of additional impairments to replication resulting from the mutation. The reduced capsid stability associated with this mutant is consistent with this interpretation.

Inhibition of the CA-CypA interaction by CsA is detrimental to the infectivity of most wild-type HIV-1 isolates, yet CsA promotes infection by the CA mutants A92E, G94D, and T54A in some cells. These observations have suggested the possibility of a host cell factor that inhibits the infection of the mutants in a CypA-dependent manner in target cells (21, 35). The identity of the host factor is unknown, but the CsA-dependent phenotype appears to be independent of TRIM5α-mediated restriction (21, 34). Somewhat paradoxically, the current view holds that CypA protects the wild-type HIV-1 capsid from an unknown restriction factor in human cells (40). Interestingly, a subclone of the human TE671 cell line exhibited loss of TRIM5α-dependent restriction of N-tropic murine leukemia virus (i.e., Ref1 activity), as well as CypA-dependent infection by wild-type HIV-1 (32). The observation that the cell subclone was unaltered in expression of TRIM5α suggests that an unidentified common host factor may be necessary for both apparent restrictions (32, 34).

The molecular basis for the opposing effects of CA-CypA on virus replication in wild-type and CsA-resistant/dependent HIV-1 mutants is currently unknown. The A92E and G94D mutations do not inhibit incorporation of CypA into HIV-1 particles and thus do not appear to block CypA binding to CA. It has been proposed that A92E and G94D mutations perturb the formation of a type II tight turn in the CypA-binding loop, potentially resulting in different effects of the CA-CypA interaction on HIV-1 growth (7). Consistent with this hypothesis, nuclear magnetic resonance structural studies revealed alterations in the type II turn in the G94D mutant CA protein (9). The A105T suppressor identified in our study may compensate for these structural changes in CA proteins or may directly prevent binding of a host cell factor that restricts infection by the CsA-resistant/dependent mutants. Interaction of CA with CypA may also result in conformational changes outside the CypA-binding loop (5). In support of this view, transfer of a small region of the CypA-binding loop of HIV-1 CA to the corresponding region of simian immunodeficiency virus renders this virus CsA dependent (7). Our identification of the T54A mutation, which resides in a CA domain distinct from the CypA-binding loop and renders HIV-1 resistant and dependent on CsA, is consistent with the notion that CypA binding results in allosteric changes in the structure of CA, thereby modulating HIV-1 susceptibility to host cell restriction. Mutations in α-helix 3 have also been reported to reduce the susceptibility of HIV-1 to postentry restriction in simian cells (30). Restriction is dependent on TRIM5α and is modulated by CypA binding (4, 23, 36). Thus, the negative effect of CypA binding on infection by the T54A mutant is reminiscent of TRIM5α-dependent restriction in simian cells, yet restriction of the phenotypically similar A92E mutant is independent of TRIM5α in human cells. A detailed understanding of the mechanism of CsA-dependent infection by HIV-1 mutants in human cells will likely await identification of the putative restriction factor.