Human Immunodeficiency Virus Type 1 Vpr Protein Does Not Modulate Surface Expression of the CD4 Receptor

Abstract

The CD4 receptor is required for the entry of human immunodeficiency virus (HIV) into target cells. It has long been known that Nef, Env, and Vpu participate in the removal of the viral receptor from the cell surface. Recently, it has been proposed that the HIV type 1 (HIV-1) Vpr protein may also play a role in the downmodulation of CD4 from the surfaces of infected cells (L. Conti, B. Varano, M. C. Gauzzi, P. Matarrese, M. Federico, W. Malorani, F. Belardelli, and S. Gessani, J. Virol. 74:10207-10211, 2000). To investigate the possible role of Vpr in the downregulation of the viral receptor Vpr alleles from HIV-1 and simian immunodeficiency virus were transiently expressed in transformed T cells and in 293T fibroblasts, and their ability to modulate surface CD4 was evaluated. All Vpr alleles efficiently arrested cells in the G₂ stage of the cell cycle. However, none of the tested Vpr proteins altered the expression of CD4 on the cell surface. In comparison, HIV-1 Nef efficiently downmodulated surface CD4 in all the experimental settings. Transformed T cells and primary lymphocytes were challenged with wild-type, Nef-defective, and Vpr-defective viruses. A significant reduction in the HIV-induced downmodulation of surface CD4 was observed in viruses lacking Nef. However, Vpr-deletion-containing viruses showed no defect in their ability to remove CD4 from the surfaces of infected cells. Our results indicate that Vpr does not play a role in the HIV-induced downmodulation of the CD4 receptor.

Human immunodeficiency virus type 1 (HIV-1) Vpr is a small protein of 96 amino acids expressed late during infection (8, 9, 33). Vpr shuttles between the nucleus and cytoplasm and is incorporated into virions through a specific interaction with the p55^Gag precursor (29, 36). Several in vitro functions have been attributed to Vpr, and a number of cellular proteins have been found to mediate these effects through direct interactions with the viral product (4, 21, 27). Although the open reading frame (ORF) for Vpr is frequently lost in viruses adapted to passage in tissue culture, Vpr is highly conserved in primary isolates (18, 43), suggesting that some of the in vitro properties assigned to Vpr may be important in vivo. Vpr plays a role in the nuclear transport of the preintegration complex in newly infected nondividing cells, such as macrophages (5, 11, 19, 30). Vpr also arrests infected cells at the G₂ stage of the cell cycle and prevents their progression into mitosis (15, 21, 24, 31). In addition, Vpr has been shown to modulate apoptosis in several in vitro cell systems (3, 12, 16, 23, 38), to enhance the fidelity rate of the viral reverse transcriptase (27), and to transactivate viral and cellular promoters (39). Recently, it was reported that expression of HIV-1 Vpr in a lymphoma T-cell line induces the downmodulation of the CD4 receptor and impairs entry of HIV particles into cells (14). These results led to the suggestion that Vpr could also contribute to the CD4 downmodulation that occurs in HIV-infected cells. We have further explored these observations and investigated the possible role of Vpr in the downmodulation of the CD4 receptor.

To address the effect of the viral protein in surface expression of CD4, we expressed Vpr in transformed T-cell lines. We also analyzed the ability of wild-type and Vpr-defective viruses to downregulate CD4 on the surfaces of these cells, as well as in primary lymphocytes. Our results suggest that Vpr does not play a role in the modulation of surface levels of the CD4 receptor in HIV-infected cells.

To investigate the ability of Vpr to modulate surface CD4, we first performed experiments in which both Vpr (HIV-1 NL4.3) and CD4 were transiently expressed in human embryonic kidney 293T cells. In this setting, Vpr and CD4 were expressed from heterologous cytomegalovirus (CMV) promoters, and analysis by flow cytometry allowed us to evaluate changes in expression of surface CD4 occurring at a posttranscriptional level. Nef, Vpu, and Env are known to achieve their effects at posttranslational steps (2, 7, 22, 28, 32, 34, 42). Cells were transfected with a mixture of Vpr-, CD4-, and green fluorescent protein (GFP)-expressing plasmids and 48 h after transfection were stained with a Cy5-conjugated CD4-specific antibody (Dako) and analyzed by flow cytometry. Figure 1A shows CD4 levels on the surfaces of GFP-positive cells. HIV-1 Nef (NA7 allele) efficiently downmodulated the levels of the viral receptor. However, expression of CD4 remained high in Vpr-transfected cells. Similar results were observed in cells expressing a Vpr protein truncated at position 43. Surface levels of a CD4 protein lacking its cytoplasmic domain were not altered by either Nef or Vpr (data not shown). Expression of Nef and Vpr was confirmed by Western blot analysis (Fig. 1B). Cell lysates from transfected cells were separated on sodium dodecyl sulfate (SDS)-polyacrylamide gels and probed with Nef- or Vpr-specific antiserum (NIH AIDS Research and Reference Reagents Program) after transfer to polyvinylidene difluoride membranes. Vpr and Nef proteins were detected as bands migrating at 14 and 27 kDa, respectively (Fig. 1B, lanes 2 and 8). The small VprΔ43 protein ran off the gel and escaped detection with Vpr-specific antibodies. To ensure that Vpr was functionally active, the DNA content of GFP-positive transfected cells was analyzed by flow cytometry after staining with propidium iodide. As expected, wild-type Vpr arrested cells in the G₂ stage of the cell cycle, whereas neither its truncated version nor Nef was able to block cell cycle progression (Fig. 1C). The above results suggest that neither wild-type Vpr nor mutant proteins unable to arrest cells in G₂ modulate expression of surface CD4 at a posttranscriptional step.

FIG. 1.

Effect of Nef and Vpr on surface levels of CD4 in 293T cells. 293T fibroblasts were transfected with a mixture of DNA containing pCMX-CD4 and pCG-GFP (2 and 1 μg, respectively) together with a Nef or Vpr expression plasmid (20 μg). Forty-eight hours after transfection, the cells were removed from the plate and the levels of surface CD4 were estimated by flow cytometry after staining with a CD4-specific monoclonal antibody. (A) CD4 levels in GFP-positive transfected cells. Filled histograms correspond to cells transfected with pCMH-Vpr, a plasmid expressing wild-type NL4.3 Vpr, and the heavy line shows expression levels in cells transfected with pCMX-Nef. CD4 levels in cells transfected with Vpr truncated at position 43 are shown in a thin-line histogram. Transfection with control plasmid pCMX-PL1 (dashed line) resulted in a CD4 expression pattern identical to VprΔ43. Staining with isotype-matched control antibodies is shown by dotted lines. (B) Expression of Nef and Vpr analyzed by Western blotting. Lysates from 293T-transfected cells were run on SDS-polyacrylamide gel electrophoresis and probed with either Vpr-specific or Nef-specific antisera. The positions of molecular mass markers are indicated on the left. (C) Cells were stained with propidium iodide and analyzed for DNA content 48 h after transfection with NA7 Nef, NL4.3 Vpr, or the truncated NL4.3 Vpr version. Only GFP-positive cells were included in the analysis. The left and right peaks constitute cells in the G₀/G₁ and G₂/M stages, respectively. Cells in the S phase are between the two peaks. ModFitLit software (Verity Software House Inc., Topsham, Maine) was used to calculate the percentages of cells in different stages of the cell cycle. Transfection with control DNA resulted in cell cycle patterns identical to those of Nef-transfected cells.

Vpr has been shown to enhance transcriptional activity from viral and cellular promoters (39). Therefore, it is conceivable that Vpr may modulate CD4 expression at a transcriptional step through the CD4 promoter. To test this hypothesis, we decided to analyze the effect of Vpr expression in T-cell lines in which expression of CD4 is driven by its endogenous promoter. Four HIV-1 alleles (NL4.3, DH12, YU2, and MN) and one simian immunodeficiency virus (SIV) Vpr allele (mac239) were cloned into pCGCG plasmids. In these vectors, a high expression level of the recombinant gene is achieved with a CMV promoter, whereas expression of a GFP reporter gene from the same mRNA is mediated by an internal ribosome entry element (40). After electrotransformation, gating on GFP-positive cells allowed us to monitor subtle variations in the levels of surface CD4 occurring in transfected cells. 293T cells were first transfected with these Vpr constructs to confirm their functional activity. All the HIV-1 Vpr alleles arrested cells at the G₂ stage, as demonstrated by DNA content analysis of transfected 293T cells (data not shown). These plasmids were used to electrotransform SupT1 cells. Forty-eight hours after electrotransformation, cells were stained with a CD4 monoclonal antibody conjugated to Cy5 and analyzed by flow cytometry. Figure 2A shows the fluorescence levels associated with GFP and CD4 expression. As a control, transfection with the plasmid pCG-GFP, which expresses GFP alone, did not alter the surface expression of CD4 (compare CD4 levels in GFP-positive and GFP-negative cells). However, expression of CD4 was reduced 5- to 10-fold in cells transfected with the Nef NA7 allele, compared to GFP-negative cells from the same transfection (Fig. 2C). Similar reductions were observed with other Nef alleles, including NL4.3 (data not shown). None of the tested Vpr alleles decreased the levels of surface CD4 expression. However, comparing CD4 levels in transfected (GFP-positive) and untransfected (GFP-negative) SupT1 cells, a minor but consistent decrease in CD4 expression was observed upon expression of the Vpr alleles (Fig. 2C). A similar reduction was found in GFP-positive cells upon electrotransformation with GFP alone (Fig. 2G), suggesting that this reduction is not Vpr dependent but likely due to the effects of overexpressing heterologous gene products. Similar results were observed in other CD4-positive T-cell lines, including H9 (Fig. 2D) and Jurkat E6 (Fig. 2E), which express medium and low levels of surface CD4, respectively. These experiments were performed with the HIV-1 NL4.3 Vpr allele; however, identical results were obtained with YU2, MN, DH12, and SIVmac239 Vpr alleles (data not shown).

FIG. 2.

Comparison of the effects of Nef and Vpr on the modulation of surface CD4 in transformed T cells. SupT1 (A, C, and G), H9 (B and D), Jurkat E6 (B and E), and Jurkat-T-high CD4 (B and F) cells were electrotransformed with 30 μg of pCDNA6 (mock), pCG-GFP (GFP), pCGCG-Nef NA7, or a pCGCG-Vpr plasmid expressing HIV-1 or SIV Vpr alleles. Forty-eight hours after transfection the surface levels of CD4 were estimated by flow cytometry after staining with a CD4-specific antibody. (A) CD4 surface levels in SupT1 cells are plotted against GFP fluorescence. (B) Similar analyses in Jurkat E6, Jurkat high-CD4, and H9 cells expressing HIV-1 NL4.3 Vpr. mock, mock transfected. (C through F) Filled histograms represent CD4 levels in GFP-negative cells transfected with pCGCG-Vpr NL4.3. Thick and thin lines represent CD4 levels in gated GFP-positive cells transfected with pCGCG-Vpr NL4.3 and pCGCG-Nef NA7, respectively. (G) CD4 levels in SupT1 cells transfected with pCG-GFP (GFP alone). Filled histograms represent CD4 levels in GFP-negative cells, and the solid line shows levels in GFP-positive cells. Staining with isotype-matched antibodies is shown by dotted lines. The isotype control was omitted from the Jurkat E6 histogram. (H) Western blot analysis of SupT1 cells transfected with different Vpr allele constructs. Lysates were probed with an HIV-1 Vpr-specific antiserum.

To rule out the possibility that the levels of Vpr expression achieved in electrotransformed cells were not sufficient to downmodulate the viral receptor, we analyzed Jurkat-T-high CD4 cells. These cells constitutively express the large T antigen of simian virus 40, which enhances transcriptional activity from CMV promoters (14a). Jurkat-T-high CD4 transfected cells consistently achieved levels of GFP expression 10- to 50-fold higher than those in Jurkat E6 or SupT1 cells, as estimated by the mean fluorescence signal derived from GFP, which is transcribed from the CMV promoter that expresses Vpr (Fig. 2B). Despite the higher levels of expression, no changes in the amount of CD4 were observed in Vpr-transfected Jurkat-T-high CD4 cells (Fig. 2F). CD4 levels remained constant even in cells with very high GFP fluorescence (Fig. 2B). Finally, we confirmed expression of the Vpr alleles in transfected cells. Lysates from transfected SupT1 cells were analyzed by Western blotting with Vpr-specific antiserum (Fig. 2H). Expression of all the HIV-1 Vpr alleles was confirmed. SIVmac239 Vpr was not recognized by the HIV-1-specific antiserum. Changes in expression levels among HIV-1 alleles were likely due to differences in the ability of the antiserum to recognize each protein, since similar GFP fluorescence signals were observed by flow cytometry (Fig. 2A). Taken together, the above findings demonstrate that expression of Vpr by itself is not sufficient to downmodulate surface CD4 levels in T cells permissive to HIV infection.

It could be argued that Vpr may require other HIV products to regulate CD4. Alternatively, Vpr expression may occur at higher levels in HIV-infected than in electrotransformed cells. To address the role of Vpr in the context of HIV infection we challenged CEM-GFP cells with either wild-type HIV-1 (NL4.3) or mutant versions lacking either Nef or Vpr. Infection of CEM-GFP cells was monitored by flow cytometry. Expression of Tat in infected cells induces the transcription of the long-terminal-repeat-GFP unit integrated into the chromosomal DNA (17). Cells were infected with 1 μg of p24 virus and 48 h later stained with a CD4-specific monoclonal antibody (OKT4) followed by incubation with a Cy5-conjugated goat anti-mouse antibody (Caltag Laboratories). As shown in Fig. 3A, infection with wild-type HIV caused a drastic reduction (more than fivefold) in the amount of surface CD4 in GFP-positive cells. Infection with Nef-defective viruses did not efficiently reduce surface CD4, confirming a major role of this protein in the HIV-induced downmodulation of the viral receptor 6, 14a, 25). As with the wild type, infection with a Vpr-defective virus efficiently induced the removal of CD4 from the surface of infected cells, demonstrating that unlike Nef, Vpr is not essential for this function. To further confirm these results, we utilized a set of NL4.3-GFP reporter viruses. These viruses contain a copy of the GFP gene inserted in the nef region of the viral genome. To make a wild-type version of the virus, the Nef ORF was restored and its expression driven by an internal ribosome entry element (10). Nef- and Vpr-defective versions were constructed by filling in the XhoI and AflII restriction sites, respectively. Inactivation of the ORFs was confirmed by Western blot analysis of lysates from transfected 293T cells (Fig. 3D). These viruses were used to infect H9 cells and phytohemagglutinin-activated primary lymphocytes (Fig. 3B and C, respectively). In neither of these cases did the absence of a Vpr gene product alter the ability of the virus to downmodulate surface CD4. In fact, infection of primary lymphocytes with Vpr-defective viruses exerted a slightly stronger downmodulation of the receptor than infection with wild-type viruses, as shown by the fact that cells with low and medium levels of GFP expression showed larger reductions in surface CD4 when infected with Vpr-defective viruses (Fig. 3C). It is possible that the absence of Vpr positively affects the expression or function of other CD4 downmodulator products. Taken together, our results strongly argue against a role for Vpr in the HIV-induced downmodulation of CD4.

FIG. 3.

Vpr does not contribute to CD4 downmodulation in HIV-infected cells. (A) CEM-GFP cells were infected with HIV-1 NL4.3 wild-type, Nef-defective (ΔNef), or Vpr-defective (ΔVpr) viruses (1 μg of p24). After 48 h, cells were stained with OKT4 and a Cy5-conjugated goat anti-mouse antibody and analyzed by flow cytometry. CD4 surface levels are plotted against GFP fluorescence. Numbers indicate the mean CD4 fluorescence value in GFP-negative (left) and GFP-positive (right) cells. The right panel shows the fluorescence levels in gated GFP-positive cells only, with the exception of mock-infected cells (filled histograms), where GFP-negative cells are shown. Thick and thin lines represent cells infected with wild-type and Vpr-defective viruses, respectively, and broken lines represent cells infected with Nef-defective viruses. Isotype-matched control antibodies are shown as dotted lines. (B and C) Infections of H9 and PHA-activated peripheral blood mononuclear cells (PBMC), respectively. Infections were performed for 48 h with NL4.3 reporter viruses encoding GFP. (B) Numbers are mean CD4 fluorescence values in GFP-negative (left) and GFP-positive (right) cells. (C) Numbers represent percentages of cells in the CD4-negative GFP-positive quadrant. (D) Lysates of 293T cells transfected with the NL4.3 HIV-GFP proviral constructs used for panels B and C. After separation by SDS-polyacrylamide gel electrophoresis and transfer to polyvinylidene difluoride membranes, samples were probed with Vpr- and Nef-specific antisera. WT, wild type.

In the present study we investigated the role of Vpr in downmodulation of the CD4 receptor. We performed a systematic study to test the ability of different Vpr alleles to alter steady-state levels of surface CD4. Five different Vpr alleles from HIV-1 and SIV were analyzed and compared with Nef for the ability to mediate the removal of CD4 from the cell surface. Our studies included the NL4.3 HIV-1 Vpr allele, previously reported to downmodulate surface CD4 (14), and also the Jurkat E6 T-cell line, used in previous studies. These studies were limited to this particular T-cell line with low levels of expression of surface CD4, where minor variations in CD4 during infection could be erroneously attributed to Vpr. Other T-cell lines with higher levels of surface CD4 may be more appropriate to evaluate Vpr's effects during infection with HIV. We extended our analyses to transformed fibroblasts, different transformed T-cell lines, and primary lymphocytes. None of our experiments showed conclusive evidence for a CD4 downmodulating activity mediated by Vpr. Unlike in the previous study, we utilized a transient system where high-level Vpr expression was achieved by either electroporation or infection of CD4-positive cells. Similar results were observed with longer expression times (up to 7 days after electroporation or infection) (data not shown). In our studies we analyzed four different HIV-1 alleles, including two from laboratory-adapted T-cell-tropic isolates (NL4.3 and MN), one dually tropic isolate (DH12), and one primary isolate with macrophage tropism (YU2) (1, 26, 35, 37). We also extended our studies to the SIVmac239 isolate. Conti et al. utilized a constitutive expression system to analyze Vpr effects (14). In their studies, transfected cells were selected for several days with antibiotics, and single Vpr-expressing clones were isolated for further study. It is conceivable that this procedure might have selected for clones with altered CD4 expression. In our laboratory, transformed T-cell lines undergo significant reductions in surface CD4 after multiple passages. Furthermore, it is striking that single-cell clones expressing the toxic Vpr product could be isolated without interfering with vital functions of the cell. Interestingly, the Vpr-transfected T-cell clones isolated by Conti et al. did not show any delay in their rate of proliferation (13). Our results also argue against a role of virion-incorporated Vpr in CD4 downmodulation, since no defects were documented in infections with virion particles lacking Vpr. It is possible that other CD4 downmodulator genes masked a weak Vpr activity. However, multiple analyses with electrotransformed cells failed to show evidence of such activity. We have to emphasize that our studies cannot rule out the existence of Vpr-induced long-term effects on CD4 expression. The results presented here may not apply to cell lines constitutively expressing Vpr. These effects might not be revealed in infected primary lymphocytes, whose half-life has been estimated to be around 24 h (20, 41), but could play a role during infection of long-lived cells harboring HIV reservoirs, such as macrophages and CD4-positive memory T cells. Additional studies may be required to evaluate in these cell types the role of Vpr in the downmodulation of the viral receptor.