Co-infection generates HIV-1 virions pseudotyped by the HTLV-1 Env

We next asked if co-infection of a cell by HIV-1 and HTLV-1 could lead to pseudotyping of the former by the latter envelope protein. Because an HIV-1 virion carrying an HTLV-1 envelope could allow the broader HIV-1 tropism, we first examined HIV-1 pseudotyping by the HTLV-1 envelope (Fig 2). Accordingly, we prepared VSV-G pseudotyped HIV-1 particles by cotransfecting a VSV-G expression vector and an envelope-deleted NL4-3 ΔEnv GFP (Zhang et al., 2004) molecular clone into 293T cells. Supernatant containing VSV-G pseudotyped HIV-1 virions was harvested from the 293T cells 48 hours after transfection and filtered; the filtrate was used to infect either Jurkat or MT2 cells. Three days later, the infected Jurkat or MT2 cells were separately co-cultured with the HIV-1 indicator cell line TZMbl. We expected that the envelope-deleted NL4-3 ΔEnv GFP genome would not produce infectious particles in Jurkat cells because these cells express no retroviral Env protein. On the other hand because MT2 cells do express HTLV-1 Env protein (Harada et al., 1985). Should this envelope protein be competent for pseudotyping HIV-1, then HTLV-1 Env pseudotyped HIV-1 particles should be produced. In that setting, HTLV-1 Env pseudotyped HIV-1 virions could be detected by an infection assay using TZMbl cells. Indeed, β-galactosidase staining of a co-culture of NL4-3 ΔEnv GFP/MT2 with TZMbl cells shows positive staining. The control co-culture of NL4-3 ΔEnv GFP/Jurkat with TZMbl does not (Fig 2A, B).

Open in a separate window

Figure 2

HIV-1 may be pseudotyped by HTLV-I Env

Jurkat or MT2 cells were infected with mock supernatant or NL4-3 ΔEnv pseudotyped with VSV-G. Three days post infection, cells were co-cultured with the HIV-1 indicator cell line TZMbl. One day after co-culture, infection with HIV-1 was detected by b β-galactosidase staining of the TZMbl cells. A) Representative micrographs of the four conditions. B) Triplicate experiments were performed to determine the number of β-galactosidase positive TZMbl cells following co-culture. C) Infection of MT2 cells with mock or NL4-3 ΔEnv pseudotyped with VSV-G was performed as above. Co-culture was performed in the presence of increasing concentrations of anti-anti-HIV-1 (VRC01) or anti-HTLV-1 neutralizing antibodies. ** p-value ≤ 0.01, *** p-value ≤ 0.001

To further verify that the results from co-culture of NL4-3 ΔEnv GFP/MT2 with TZMbl cells were due to pseudotyping of NL4-3 ΔEnv GFP with the HTLV-1 Env from MT2, we repeated the co-culturing assay in the presence of neutralizing antibodies specific for HIV-1 or HTLV-1 envelope proteins (Fig 2C). In the presence of anti-HTLV-1 antibody, a dose-dependent inhibition of TZMbl reporter readout was seen, while in the presence of anti-HIV-1 antibody a small change could be observed only at the highest antibody concentration. These results support the interpretation that infectious NL4-3 ΔEnv GFP virions arise after pseudotyping by HTLV-1 Env protein.

Construction of a chimeric HTLV-1 genome that expresses HIV-1 Envelope

We took a different approach to investigate whether HTLV-1 could be pseudotyped with HIV-1 Env. Using the HTLV-1 ACH molecular clone (Kimata et al., 1994), we created a chimeric HTLV-1 genome in which the native Env gene was replaced with the HIV-1 Env gene from the NL4-3 molecular clone (Fig. 3). We reasoned that if HIV-1 Env could complement HTLV-1 Env function, then this chimeric HTLV-1 genome should be replication competent. In our chimera, we modeled our cloning strategy after that performed by Delebecque et al. in which the HTLV-I Env in a molecular clone was replaced by a heterologous Env gene from MuMLV (Delebecque et al., 2005; Delebecque et al., 2002). In order not to perturb the splicing and processing of the HTLV-I genome, the signal peptide from the HTLV-1 Env gene was fused to the NL4-3 HIV-1 Env resulting in a modified Env whereby the signal peptide of HIV-1 Env was replaced with the HTLV-I signal peptide. This chimeric Env was then inserted back into the ACH molecular clone using the SphI and NsiI restriction sites (Fig 3A). Although this construct will not express HIV-1 Rev, the HTLV-1 Rex coding sequence remains un-perturbed. Rex has been shown to be capable of stabilizing HIV-1 Env, complementing Rev function and will allow expression in this chimera (Hanly et al., 1989).

Open in a separate window

Open in a separate window

Figure 3

Construction and characterization of a chimeric Env and chimeric HTLV-1 molecular clone

A) Signal peptide from HTLV-I Env (top panel) was fused to the HIV-1 Env sequence (middle panel) to create a chimeric Env (bottom panel) which was then transferred back into the pACH molecular clone by way of the SphI and NsiI sites. This process results in the preservation of HTLV-I splicing sequences and perturbs no viral genes apart from the Env. B) Western blot analysis of expression of HIV-1 Env after transfection of seeded 2×10⁵ TZMbl cells with 0, 0.25, 0.5 or 1 microgram of pACH or pACH-Chimera. Arrows indicate gp160 and gp120. Extracts were used to blot for β-actin to insure equal loading. C) 5×10⁴ Jurkat T-cells were seeded in each well of a 96 well plate and infected with ACH or ACH-Chimera in the presence of αHIV-1 nAb (1:100) or αHTLV-1 nAb (1:200). Eight days post-infection RNA was isolated from the cells and used to determine relative infection by qPCR for HTLV-1 Gag. * p-value ≤ 0.05, ** p-value ≤ 0.01, *** p-value ≤ 0.001 D) 6×10⁶ Jurkat T-cells were infected with equal amounts of ACH or ACH-Chimera molecular clones as determined by p19 Elisa. RNA was extracted from infected cells at days 4, 6, 8, 10, 12 and 14 post infection. Quantitative RT-PCR was used to determine the expression of HTLV-1 Gag mRNA as compared to cellular GAPDH mRNA. E) ACH or Chimera infected Jurkat T-cells from day 4 of infection were exposed to equal quantities of NL4-3 HIV-1 molecular clone. Supernatant was harvested at days 2, 4, 6 and 8 post infection. HIV-1 replication was determined by measuring infectious units per ml of culture media in TZMbl cells.

We next determined if the pACH-Chimera proviral genome could express HIV-1 Env. TZMbl cells were transfected separately with increasing amounts of pACH or pACH-Chimera, or with a pCMV 4-3 Env expression vector as control (Fig 3B). Western blotting using human HIV-1 hyper-immune serum showed that increasing the transfection of pACH-Chimera led to increasing expression of Env protein in the TZMbl cells (Fig. 3B, lanes 6–8) which co-migrated with envelope protein produced from transfected pCMV 4-3 Env expression vector (Fig. 3B, lane 1). In contrast the transfection of the parental pACH genome produced no HIV-1 Env protein (Fig. 3B, lanes 3–5).

To test the infectivity of our ACH-Chimera virus, we generated viral stocks by transfecting 293T cells with 10 µg of pACH with 1 µg pCTax (CMV promoter-driven Tax) or 10 µg pACH-Chimera with 1 µg pCTax. Seventy-two hours after transfection supernatants were harvested, filtered, and HTLV-1 virus production was determined by HTLV-1 p19 Gag Elisa (Zeptometrix, Buffalo, NY). To characterize the ACH and ACH-Chimera viruses, we asked if we could distinguish between the two viruses based on their envelope proteins. Thus, we challenged ACH-virus or ACH-Chimera particles with either HIV-1 or HTLV-1 neutralizing antibodies (Fig 3C). Accordingly, Jurkat T-cells were infected with p19 Gag-equivalent amounts of either ACH WT or ACH Chimera in the presence of neutralizing antibodies specific for HIV-1 or HTLV-1. Eight days post infection the level of cell associated viral RNA was determined by quantitative RT-PCR of total cellular RNA, and we observed that the ACH-virus was sensitive to HTLV-1 neutralizing antibody while the ACH-Chimera was sensitive to HIV-1 neutralizing antibody, consistent with the expression and function of HIV-1 Env expressed from the latter genome (Fig. 3C).

We next determined the ability of the ACH-Chimera virus to replicate in tissue culture. Jurkat T-cells were infected with ACH or ACH-Chimera virus, normalized by HTLV-1 p19 Gag measurements. RNA was extracted from infected cells at 4, 6, 8, 10, 12 and 14 days post infection, and HTLV-1 viral RNA was quantified by q-RT-PCR (Fig 3D). The results showed that both ACH and ACH-Chimera were replication competent with the latter replicating approximately 0.5 to 1 log less well than the wild type ACH.

Flow cytometry

Infection of J-LTR-G cells was followed by flow cytometry. Cells were stained using anti-gp46 antibody (Abcam) for one hour, washed and then stained with anti-mouse AlexaFluor647 (Invitrogen). Staining for cell surface gp46 and GFP expression was detected using a Becton Dickinson Fortessa LSR.

Fluorescent Microscopy

Jurkat cells were affixed to a positively charged glass slide through use of a CytoSpin and fixed with 1% paraformaldehyde for 10 minutes. Cells were then washed, blocked in PBS with 1% BSA and 0.1% NP40 and stained with goat anti-gp120 (1:500) (AbCam) or mouse anti-Tax hybridoma supernatant (1:200). After another series of washes, the cells were stained with anti-goat antibody conjugated to AlexaFluor594 (1:1000) (Invitrogen) and anti-mouse AlexaFluor488 (1:1000) (Invitrogen). Cells were then washed, stained with DAPI and imaged use a fluorescent microscope.

Infection and neutralization assays

TZMbl cells were used to determine infection via HIV-1 (either WT or pseudotyped). Cells were incubated with cell free virus or co-cultured with source cells for twenty-four hours. After incubation with virus, the cells were fixed and stained for expression of β-galactosidase fluorescently using the ImaGeneRed LacZ kit (Invitrogen). For neutralization assays, cells were incubated with anti-HIV-1 VRC01 or anti-HTLV-1 antiserum (AIDS Reagent Program) concomitantly with infecting virus. Final dilutions of antibody were 1:100, 1:200 or 1:1000 as indicated.

Viral infections

HIV-1 viral stocks were generated by transfecting the proviral plasmid pNL4-into 293T cells. Seventy-two hours after transfection, the culture supernatant was clarified by centrifugation and filtered using a 0.22 μM filter to remove remaining cells. Viral stocks were quantified using reverse transcriptase assay and titered on TZMbl cells. Infection of TZMbl cells was determined by β-galactosidase staining and replication in Jurkat T-cells was followed by reverse transcriptase assay.

HTLV-1 viral stocks were generated by transfecting 293T with pACH proviral plasmid and pCMV-Tax at a ratio of 10:1. Seventy-two hours after transfection the culture supernatant was clarified by centrifugation, filtered using a 0.22 μM filter to remove remaining cells and concentrated by ultracentrifugation. Viral stocks were quantified using p19 Elisa (Zeptometrix). For cell free infection, 6×10⁶ Jurkat cells were incubated with HTLV-1 viral stock containing 10ng p19 in 2 ml media for 16 hours. Infected cells were then washed and re-suspended in media at 1×10⁶ cells/ml. Infection was followed by extracting cell associated RNA from the cells every other day and virus replication was detected by qPCR for HTLV-1 Gag

Construction of chimeric HTLV-1

The leader peptide of HTLV-1 Env was PCR amplified from the X1MT HTLV-1 molecular clone using primers X1MT 5226 F and HTLV Env ExtR. The HTLV Env Ext R primer contains sequence matching the desired junction with the HIV-1 envelope and the KpnI site found in the HIV-1 Envelope. Amplified HTLV-1 leader peptide plus adapter was digested with SphI and KpnI and ligated into the SphI and KpnI sites of pUC19 to created pUC-3’HTLVEnv.

A fragment of the HIV-1 Env was amplified from pNL4-3 using HIV Env F and HIV 8785 EcoRI R primers. The resulting PCR fragment was digested with KpnI and EcoRI and ligated into the KpnI and EcoRI sites of pUC-3’HTLVEnv. The complete envelope containing fragment was then amplified by PCR using primers Chim PstI and Chim R (below). Fragment was digested using SphI and PstI and ligated into pACH using the SphI and NsiI sites.