Prior evaluating the potential formation of pseuotyped particles, we first analyzed at the cellular level the co-expression of gB and gp160, or gD and gp160, in HIV-1 infected P4P cells superinfected by HSV-2, and in HIV-1-infected P4P cells transfected by plasmids coding for gB or gD, by using confocal microscopy. Our observations confirm that P4P cells exposed to both HIV-1 and HSV-2 can be co-infected and coreplicate the two viruses, since we can detect in the same cell glycoproteins from both viruses. HIV-1 P4P cells transfected with plasmids coding for gB or gD also express both the HSV-2 glycoprotein and gp160 (Figure 1). In addition, as shown in Figure 1, the superposition of fluorescence signals indicates that gB and gD may co-localize with HIV-1 gp160 in both co-infected and transfected P4P cells. The pattern of staining in single infections with either HIV-1 or HSV-2 was the same as that observed in co-infected cells (not shown).

Prior HSV-1 infection of CEM cells has been shown to result in a delay of HIV-1 production, due to cell growth inhibition and apoptosis, while HSV-1 superinfection of HIV-infected CEM cells promoted HIV expression 17. We have therefore focused our analysis on this latter condition.

CEM cells infected by HIV-1NDK strain were superinfected by HSV-1 or HSV-2 at a multiplicity of infection (MOI) of 0.01, 0.1 or 1 pfu/cell for 72 h. Every each 24 hours, culture supernatant was collected for viral quantifications by real time PCR, and after washing cells were reincubated with fresh medium. Our results show that HSV-1 and HSV-2 did not modify the production of HIV-1 RNA, and that HIV-1 did not influence either HSV-1 or HSV-2 replication in CEM cells (Figures 2 and 3). The cell free supernatants of dually infected CEM cells were then used for the infection of Vero cells. In the first 48 h, cytopathic effect related to HSV-2 was observed by optical microscopy (not shown). No HIV-1 p24 antigen was detected in the supernatant of Vero cells at 24 h or 48 h, and DNA of the Vero cell pellet contained no HIV-1 proviral DNA as detected by real time PCR (not shown). The HSV-1 and HIV-1- co-infected CEM cells and HSV-2- and HIV-1- co-infected CEM cells did not lead to the production of HIV-1 virions able to infect cells that are usually non-permissive.

Intestinal HT29 cells are epithelial cells susceptible to infection by both R5 and X4 HIV-1 strains 18. HT29 cells were infected with HIV-1NDK for 48 h before HSV-2 superinfection at MOIs of 0.01, 0.1 and 1. HIV-1 production was assessed at 24 and 48 h following the infection. At 24 h, the medium was removed for viral quantification and was replaced with fresh medium for the 48 hour time-point. The results depicted in Figure 4 show that HIV-1 production by HT29 cells decreased significantly following HSV-2 superinfection at a MOI of 1 pfu after 24 h (P = 0.04). The influence of HSV-2 at a MOI of 1 pfu on the decrease of HIV replication was more pronounced 24 hours later. In contrast, lower HSV-2 inoculum (0,1 and 0.01 pfu) did not seem to inhibit HIV-1 production in a 48 h delay. Supernatants of HIV-1 and HSV-2 dually infected HT29 cells collected at 24 and 48 h did not lead to HIV-1 infection of Vero cells as determined by detection of proviral DNA, while HSV-2 replication was confirmed by the detection of typical cytopathic effect (not shown).

P4P cells, derived from the endometrial epithelium, were also infected in the same conditions as those used for HT29 cells. The rate of HIV-1 infection of P4P cells was around 50%, as determined by the ratio of HIV-1 DNA and albumin levels, suggesting that any viral superinfection would probably produce dually infected cells. As shown in Figure 5, superinfection by HSV-2 did not significantly influence HIV-1 production in P4P cells. The culture supernatants collected 24 and 48 h after HSV-2 superinfection were used to infect Vero cells. After 48 h of infection, no HIV-1 proviral DNA was detected in the cell pellet, while HSV-2 production was confirmed by the detection of HSV-2 DNA in the supernatant (not shown). Co-infected P4P cells did not produce HIV-1 particles able to infect Vero cells. To check the infectiousness of HIV particles produced in the supernatant of co-infected P4P cells, we tested 24 and 48 h culture supernatants from dually HIV-1 and HSV-2 infected P4P cells on uninfected P4P cells for HIV-1 replication. As expected, these supernatants induced HIV productive infection at 24 hours (not shown), thus confirming that supernatants recovered after co-infection of P4P cells contained infectious HIV-1 particles.

Similar results were observed for both epithelial cell lines with R5-tropic strains HIV Ba-L and HIV JR-CSF (not shown). Thus co-infection of intestinal or endometrial epithelial cells with HIV-1 and HSV-2 did not lead to the production of HIV-1 virions with the ability to infect cells naturally non-permissive for HIV-1.

In order to avoid the necrosis and apoptosis of epithelial cells consequent to HSV-2 infection, we transfected HIV-1 NDK-infected P4P cells with plasmids coding for gB and gD, which are major glycoproteins involved in fusion of the HSV-2 envelope with the plasma membrane. Culture supernatants obtained 48 h following the transfection were used to infect Vero cells for 24 h. Cell DNA was then extracted, and tested for HIV-1 DNA. Proviral HIV-1 DNA was never detected, suggesting that the expression of gB and gD in HIV-1 infected P4P cells did not modify the phenotype of HIV-1 viral particles such that they could infect Vero cells.

In order to assess weteher the colocalization of HSV-2 and HIV glycoproteins can lead to the incorporation of the former into the HIV-1 envelope, we carried out a viral capture assay. Supernatants of HSV-2 and HIV-1 dually infected P4P and CEM cells and of P4P cells infected by HIV-1 then transfected with plasmids coding for gB or gD were first incubated with mouse antibody to gB or gD and biotin-labeled anti-mouse antibody and second with streptavidin microbeads to capture viral particles bearing gB or gD glycoproteins. The positive control capture of HIV-1 particles was assessed by anti-human CD44. Bound viruses were further tested for the detection of HIV-1 RNA and HSV-2 DNA. Capture assays with gB or gD specific antibodies were able to bind HSV-2 virions (Figure 6). No HIV-1 RNA was detected, either from dually infected cells or from gB or gD transfected HIV-1 positive cells, suggesting that HIV-1 particles did not acquire gB or gD glycoproteins during budding (Figure 6).

P4P cells were kindly provided by Florence Damond. P4P cells are adherent HeLa cells transfected by plasmids coding for CD4 and CCR5 (Barin et al, 2004) and allow infection by both X4 and R5 viruses. P4P cells were grown in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 500 mg/mL G418, and 100 mg/mL hygromycin B and antibiotics (Penicillin 100 U/ml and streptomycin 100 μg/ml (PS)). Vero cells (African green monkey kidney cells) were grown in DMEM supplemented with 10% FBS and PS. CEM-SS cells were grown in RPMI-1640 medium supplemented with 10% FBS and PS.

HSV-2 strain VR734 was obtained from the American Type Culture Collection (ATCC) (Manassas, VA), HSV-1 MacIntyre Strain was purchased from Advanced Biotechnologies Inc (Columbia, MA), and propagation and titer determination were carried out on Vero cells.

Primary R5-tropic HIV Ba-L, HIV JR-CSF, and X4-tropic HIV NDK were a gift from Pr. F. Barre'-Sinoussi (Institut Pasteur, Paris, France). HIV Ba-L and HIV JR-CSF were produced in primary macrophages; HIV NDK was produced on IL-2-activated peripheral blood lymphocytes. Viral titre was measured by HIV-1 p24 antigen quantification.

Mouse anti-gB and anti-gD monoclonal antibodies (Mabs) were obtained from Serotec (Oxford, UK) and Advanced Biotechnologies Incorporated, respectively. Purified human polyclonal anti-gp160 IgG was described previously 39. FITC-labeled F(ab')2 goat anti-human IgG, FITC-labeled F(ab')2 rabbit anti-human IgG and TRITC-labelled rabbit anti-mouse IgG were purchased from Jackson Immunoresearch (Cambridgeshire, UK) and Molecular Probes (Cergy Pontoise, France), respectively.

Plasmids pMM245 and pMM346, coding for HSV-2 gB and gD respectively, were described previously 40.

P4P and HT29 cells in 12-well plates were infected with 5 ng/ml p24 HIV-1BaL, HIV-1JR-CSF, and HIV-1NDK for 3 h at 37°C and CEM cells were infected with 1 ng/105 cells of p24 HIV-1BaL, HIV-1JR-CSF, and HIV-1NDK. The cells were washed, and were then either cultured for 48 h before superinfection with 1, 0.1 or 0.01 plaque forming unit (pfu)/cell of HSV-2 or transfected with the plasmids pMM245 and pMM346 and cultured for 48 additional hours.

Vero cells in 24-well plates were incubated for 3 h at 37°C with culture supernatants of CEM cells, HT29 cells and P4P cells. After extensive washing, Vero cells were cultured for 24 to 48 h. Supernatants were frozen at -80°C until subsequent analyses and cells were then washed, before the nucleic acid extraction of the cell pellet for the detection of proviral DNA.

The transfection assay was performed using 1 μg of plasmid with the transfection agent Fugene^® (Roche Applied Science, Meylan, France). The protein expression was assessed 48 h following the transfection by intracellular staining as described below.

Cells were washed, fixed with paraformaldehyde (4% in phosphate-buffered saline [PBS]) for 10 min at 4°C before permeabilization in PBS with 0.5% Triton X-100 and 10% sucrose at room temperature. After several washes with PBS, cells were incubated for 30 min with human anti-gp160 IgG, or with mouse anti-gB or anti-gD Mabs diluted in PBS with 1% bovine serum albumin. FITC-labeled F(ab')2 rabbit anti-human IgG and TRITC-labelled rabbit anti-mouse IgG were further added at a dilution of 1/50 for 30 min at room temperature. The coverslips were mounted in Mowiol (Sigma, St. Louis, Mo.) and observed by confocal microscopy. Observations were made by sequential acquisition with a Zeiss LSM510 System, mounted on an Axiovert 100 M optical microscope (Carl Zeiss AG, Oberkochen, Germany), using a planapochromat x63, 1.4 numerical aperture oil immersion objective. Optical sections were acquired, each one with an image resolution of 512 × 512 pixels.

To assess the incorporation of HSV-2 envelope glycoproteins into the HIV-1 membrane, we carried out a viral capture assay to bind viruses bearing gB or gD as follows. HIV-1, HSV-2, and mixture of HIV-1 and HSV-2 virions, produced in gB- or gD-expressing P4P cells (200 μl), were incubated first with 1 μg of mouse anti-gB or mouse anti-gD and second with 1 μg of anti-mouse biotinylated antibody (Sigma-Aldrich, Lyon, France) for 30 min at room temperature to bind viruses bearing HSV-2 glycoproteins. Similarly, to bind HIV-1 paricles, supernatants were incubated with mouse anti -human CD44 (BD Biosciences, Le Pont de Claix, France) and with 1 μg of anti-mouse biotinylated antibody (Sigma-Aldrich) as described 41, since CD44 has been demonstrated to be the most effective host cell marker for the capture of HIV-1 from patient samples and culture-derived virus, independent of the origin of the virus 41. To each sample were added 25 μl μMACS™ Streptavidin MicroBeads (Miltenyi Biotec, Auburn, CA) and the binding reaction was incubated an additional 10 min at room temperature. Antibody-bound viruses were then captured by magnetic separation according to the manufacturer's protocol. Briefly, μMACS columns were placed in the μMACS™ magnetic separator attached to a magnetic MultiStand. Each column was prepared by prewetting with 100 μl protein equilibration buffer and rinsing twice with 100 μl PBS containing 2% fetal bovine serum (PBS/FBS). The entire volume of the virus capture reaction mixture (200 μl) was then applied to the column and allowed to drain completely. The columns were washed six times with 200 μl volumes of PBS/FBS. Fifty microliters of an appropriate lysis buffer (see below) were then added to the column, allowed to stand for 5 min at room temperature, and then followed by an additional 150 μl of the same buffer. The eluates were then processed for nucleic acid purification as follows. Nonspecific viral binding was assessed by adding anti-mouse biotinylated antibody to virus before capture (negative control).

Nucleic acids were extracted with an extraction protocol on silica column system (High Pure Viral Nucleic Acid kit, Roche Diagnostics, Mannheim, Germany) according to the manufacturer's recommendations. Total nucleic acids were eluted in 100 μl of RNAse- and DNAse-free water.

HIV-1 RNA and DNA quantification PCR were carried out by real time using the forward NEC152 (5'-GCCTCAATAAAGCTTGCCTTGA-3') and reverse NEC131 (5'-GGCGCCACTGCTAGAGATTTT-3') primers and the exonuclease probe (5'-FAM-AAGTAGTGTGTGCCCGTCTGTTRTKTGACT-TAMRA-3') in the long terminal repeat gene 42 on the Light cycler instrument 1.5 (Roche Diagnostics) as described 43. For HIV RNA quantification, the positive control consisted of culture supernatant of HIV-1 subtype A strain. For HIV DNA quantification, the positive control consisted of DNA purified from T lymphoblastoid cell line 8E5 (NIH, ATCC 8993), containing a single copy of HIV-1 LAV DNA provirus per cell. The level of albumin DNA copies, quantified by real time PCR as described previously 44, was used as an endogenous reference to normalize the variations in cell number or DNA extraction. HIV-1 RNA was expressed in copies/ml, and HIV-1 DNA proviral load as the number of HIV-1 DNA copies per 10⁶ cells.

HSV-2 DNA was detected and quantified by real time PCR, using HSV pol F (5' GCTCGAGTGCGAAAAAACGTTC 3') and HSV pol R (5' TGCGGTTGATAAACGCGCAGT 3') as primers, and HSV-2 FLU (5' GCGCACCAGATCCACGCCCTTGATGAGC-FLUOR) and HSV-2 LCR (5' LC-Red 640-CTTGCCCCCGCAGATGACGCC-phos) as fluorescence labelled probes, as previously described 45. The positive control consisted of the 324-bp pcDNA3.1/HisC-ul30hsv plasmid containing the targeted nucleotides 2660 to 2963 of the UL30 gene 43. A standard graph of the Ct values was obtained from serial dilutions (10 to 10⁷ copies) of the plasmid.

Levels of HIV-1 p24 antigen released in culture supernatants were determined by a commercial enzyme-linked immunosorbent assay (ELISA) (INNOTEST^® HIV Antigen mAb, Innogenetics, Lille, France), with a cutoff value of 1 pg/ml.