HIV-1 auxiliary proteins do not influence cell-to-cell infection of macrophages by cell-cell fusion with infected T cells

To investigate whether Nef, Vpr, Vif and Vpu modulated viral cell-to-cell transfer from infected CD4 T cells to macrophages, mutated viruses respectively deleted of each viral protein were generated. These viruses were used to infect Jurkat cells serving as virus-donor T cells (Fig 1A and 1B). As schematized in Fig 1C, these infected Jurkat cells were then used for analyzing virus cell-to-cell transfer in MDMs by coculture for 24 h, as described previously [40,43]. After elimination of T cells, Gag + MDMs were quantified by flow cytometry just after coculture (D0), and 4 (D4) and 6 (D6) days after coculture (Fig 1D). Surprisingly, no differences in virus cell-to-cell transfer and spreading to MDMs were observed between WT and deleted viruses. Similarly, no differences were evidenced in virus production quantified in the MDM cell-culture supernatant (Fig 1E, left panel, and S1A and S1B Fig). By contrast, virus spreading and production were affected when MDMs were infected with deleted cell-free viruses (Fig 1E, right panel, and S1C and S1D Fig). The exception was the Nef-deleted viruses, inducing a slight but not significant decrease in virus production by cell-free infected MDMs compared to WT viruses (Fig 1E).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Influence of HIV-1 auxiliary proteins in virus cell-to-cell transfer from infected CD4 T cells by cell-cell fusion with macrophages.

(A and B) Jurkat cells were infected with the WT HIV-1 NLAD8 or the mutated viruses deleted of each viral protein, and then analyzed 36 h later by flow cytometry after intracellular Gag staining (A) or by Western blotting (B) on cell lysates using antibodies targeting each viral protein, anti-Gag (p24) and anti-actin antibodies. In (A), the results are the means of 6 independent experiments, and are expressed as the percentages of Gag+ Jurkat relative to the percentage determined with the WT virus (100%). Each dot corresponds to an individual experiment. (C) Experimental design for analysis of HIV-1 cell-to-cell transfer from infected CD4 T cells to MDMs (created with Biorender). (D) Jurkat T cells infected with the WT or mutated viruses were cocultured for 24 h with MDMs. After T cell elimination, MDMs were analyzed immediately or cultured for 4 or 6 additional days before quantification of the percentages of Gag + MDMs by flow cytometry. Results are the means of at least 3 independent experiments performed with MDMs from at least 3 different blood donors, and are expressed as the percentages of Gag + MDMs relative to that determined by coculture of MDMs with Jurkat cells infected with WT viruses (100%). Each dot corresponds to an individual donor. (E) MDMs were infected by coculture with infected Jurkat T cells (left panel) or exposed to cell-free viruses (right panel) for 24 h. After elimination of T cells or cell-free viruses, cell-culture supernatant from MDMs were collected 6 days later, and the viral production was quantified (p24). The results were expressed as the means of p24 production quantified from MDM supernatants of at least 5 independent experiments from at least 5 different donors. NI, MDMs cocultured with noninfected Jurkat cells. Each dot corresponds to an individual donor. (F and G) Jurkat cells infected with WT or mutated NLAD8 viruses were cocultured for 24 h with MDMs. After T cell elimination, MDMs were stained immediately after coculture or 4 and 6 days later, with anti-Gag (green) antibodies, phalloidin (F-actin, red), and the nuclei with Dapi (blue), before observation by confocal microscopy. The total number of nuclei (Dapi+) per Gag + MDM was quantified on at least 50 cells. In F), representative images are shown. In G), results are expressed as the means of nucleus number per Gag + MGC. NI, MDMs cocultured with noninfected Jurkat cells. Error bars represent 1 standard error of the mean (SEM). Statistical significance was determined using the Anova test, and P values were obtained by Dunnett’s post-test correction (ns, P > 0.05; ***, P < 0.001; ****P < 0.0001).

https://doi.org/10.1371/journal.ppat.1013130.g001

Since we previously showed that HIV-1 cell-to-cell transfer is mainly related to cell-cell fusion of infected T cells with macrophages [43], we evaluated whether cell fusion was affected when T cells were infected with deleted viruses. Infected Jurkat cells were cocultured for 24 h with MDMs before visualization of Gag+ cells by confocal microscopy, just after coculture, or 4 and 6 days after T cell elimination. All Gag + MDMs showed a diffuse Gag staining, and were multinucleated (Figs 1F, 1G and S2). These results indicate that deletion of Vif, Vpr, Nef and Vpu has no influence on HIV-1 cell-to-cell infection of macrophages by initial cell-cell fusion with infected T cells.

HIV-1 cell-to-cell transfer in macrophages is not influenced by SERINC5

A main function of Nef is to counteract SERINC proteins restricting HIV-1 entry [45,46]. To investigate the role of SERINC5, the main SERINC member restricting HIV-1, in the cell-cell fusion with infected T cells for virus cell-to-cell dissemination in macrophages, we used SERINC5-KO Jurkat clones generated via CRISPR/Cas9-assisted gene editing [47]. Since SERINC5 restriction is also dependent of the viral envelope, the 89.6 HIV-1 strain, known as very sensitive to SERINC5 [48], was included. As shown in S3 Fig, deletion of Nef in the NLAD8 or 89.6 strains had no effect on virus cell-to cell infection of macrophages by initial cell-cell fusion with infected T cells. Parental or SERINC5-KO Jurkat cells (B1, B5, and F6 clones) were then infected with WT or Nef-deleted NLAD8 and 89.6 viruses (S4A and S4B Fig), and then cocultured for 24 h with MDMs (Figs 2 and S4C–S4L). Virus cell-to-cell transfer and MGC formation were analyzed by flow cytometry and confocal microscopy after coculture, or 6 days after T cell elimination. By flow cytometry, no differences in virus cell-to-cell infection of MDMs were observed when SERINC5-KO Jurkat cells were compared to the parental cells, even when the SERINC5-KO clones were infected with the Nef-deleted 89.6 strain (Fig 2A and 2B). Accordingly, parental and KO-SERINC5 Jurkat cells infected with Nef-deleted (Fig 2C–2J) or WT (S4C and S4L Fig) viruses equally formed Gag + MGCs. These results indicate that SERINC5 suppression in virus-donor T cells does not impact cell-cell fusion for HIV-1 infection of MDMs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Infection of MDMs by cell-cell fusion with Serinc5-deleted infected T cells.

Parental (mock) or SERINC5-KO Jurkat cells (B1, B5, and F6 clones) infected with Nef-deleted NLAD8 or 89.6 viruses were cocultured for 24 h with MDMs. After elimination of T cells, MDMs were analyzed immediately or cultured for 6 additional days. (A-B) Flow cytometry analysis. Results are the means of at least 5 independent experiments performed with MDMs from at least 5 different donors, and are expressed as the percentages of Gag + MDMs relative to that determined by coculture of MDMs with the infected parental Jurkat cells (100%). (C-J) Fluorescent confocal microscopy analysis, performed as described in Fig 1. The number of nuclei (Dapi+) per Gag + MDM was quantified on at least 50 cells. In C, E, G and I), representative images are shown. In D, F, H and J), results are expressed as the percentage of Gag + MDMs with 2, 3, 4 or more than 4 nuclei quantified from a representative experiment (right panels). In left panels), the results are expressed as the means of nuclei per Gag + MGC, and represent the means of at least 4 independent experiments performed with MDMs of 4 different donors. Error bars represent 1 SEM. Statistical significance was determined using the Anova test, and P values were obtained by Dunnett’s post-test correction (ns, P > 0.05).

https://doi.org/10.1371/journal.ppat.1013130.g002

HIV-1 cell-to-cell spreading in macrophages and MGC formation are not affected by interferon-alpha and escape APOBEC3G restriction

Since expression of some host restriction factors is induced by IFN-I [37,49], we next investigated the susceptibility to IFNα of HIV-1 cell-to-cell macrophage infection. MDMs were pretreated with increasing concentrations of IFNα, and then cocultured with primary autologous infected CD4 + T cells as previously. MDMs were then analyzed by flow cytometry and confocal microscopy, just after coculture or 6 days later (Fig 3). Surprisingly, macrophage infection by virus cell-to-cell transfer was not affected by IFNα, even at the highest concentration (10,000 Unit/mL) (Fig 3A). In contrast, cell-free MDM infection was strongly inhibited by 1,000 Unit/mL of IFNα or lower concentration (Figs 3B, S5A and S5E). Cell-free infection of primary CD4 T cells was also affected by IFNα (Fig 3C). By contrast, cell-cell fusion between infected T cells and MDMs, as well as MGC formation were not affected by IFN-I (Figs 3D–3F and S5I–S5M) compared to cell-free virus infection (S5B–S5D and S5F–S5M Fig). Accordingly, virus production assessed 6 days after coculture was moderately affected by treatment with 1,000 Unit/mL of IFNα (Fig 3G, left panel), whereas cell-free virus MDM infection was totally inhibited (Fig 3G, right panel). These findings demonstrate that HIV-1 cell-to-cell spreading in macrophages through cell-cell fusion is not sensitive to IFN-I. Since HIV-1 restriction is largely governed by IFN-I-stimulated cellular factors such as APOBEC3 cytidine deaminases [49], we confirmed the stimulation of APOBEC3G and 3A expression in MDMs treated with 1,000 Unit/mL of IFNα (Fig 3H). Similarly, expression of the IFN-stimulated Mx2 protein, a factor inhibiting viral DNA nuclear translocation [50], was also upregulated in IFNα-treated MDMs (Fig 3I).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Sensitivity to type I IFN and role of APOBEC3G in HIV-1 cell-to-cell infection of macrophages by cell-cell fusion with infected T cells.

(A) Infected primary CD4 T cells were cocultured for 6 h with autologous MDMs (CTC) pretreated overnight with the indicated concentration of IFNα2b. After T cell elimination, MDMs were stained for intracellular Gag and analyzed by flow cytometry just after the 6 h of coculture (left panel) or cultured for 6 additional days (right panel) with or without IFNα. Results are the means of at least 2 independent experiments performed with MDMs from at least 2 different donors, and are expressed as the percentages of Gag + MDMs relative to that determined by coculture of MDMs with infected CD4 T cells without IFNα (100%). Each dot corresponds to an individual donor. (B and C) MDMs (B) or primary CD4 T cells (C) were infected with cell-free (CF) viruses with or without IFNα (1,000 U/mL), and analyzed by flow cytometry after intracellular Gag staining 6 days later. Results are the means of at least 4 independent experiments performed with MDMs and CD4 T cells from at least 4 different donors, and are expressed as the percentages of Gag + MDMs or CD4 T cells relative to that determined by cell-free infection without IFNα (100%). (D-F) MDMs were inoculated with cell-free viruses or cocultured (CTC) with autologous infected T cells for 6 h without (mock in D) or with the indicated concentration of IFNα. After elimination of T cells or cell-free viruses, MDMs were analyzed by fluorescence microscopy just after coculture, or cultured for 6 additional days. The number of nuclei per Gag + MDM was quantified on at least 50 cells. Representative images are shown in (D). NI, MDMs were cocultured with noninfected CD4 T cells. In E and F), results are expressed as the means of nucleus number per Gag + MGC, and represent the means of at least 4 independent experiments performed with MDMs of 4 different donors. (G) MDMs were cocultured with infected CD4 T cells (left panel) or infected by cell-free viruses (right panel), for 6 h without (mock) or with IFNα (1,000 U/mL). After elimination of viruses or T cells, virus production was analyzed 6 days later from MDM supernatants from 5 independent experiments performed with MDMs of 5 different donors. Results are expressed as the percentages of p24 production relative to that determined without IFNα (mock, 100%). Each dot corresponds to an individual donor. Error bars represent 1 SEM. Statistical significance was determined using the Mann-Whitney U-test (*, P < 0.05; ****, P < 0.0001). (H and I) MDMs were cultured overnight without (mock) or with IFNα (1,000 U/mL), and then analyzed by Western blotting with antibodies against APOBEC3G and 3A (H), or MX2 (I). Representative experiments are shown in left panels. Right panels correspond to the quantification of the band intensities, and the signals measured in IFN-treated MDMs were normalized to the signal obtained for actin. Results are expressed as the percentages of the signal intensity measured with IFNα relative to the signal measured without (mock, 100%), and are means of at least 5 independent experiments performed with MDMs of at least 5 different donors. Each dot corresponds to an individual donor. Error bars represent 1 SEM. Statistical significance was determined using the Mann-Whitney U-test (*, P < 0.05). (J and K), Western blot analysis of lysates of siRNA-transfected MDMs (control siCtrl or siA3G) using anti-APOBEC3G and -3A, and anti-Actin antibodies. A representative experiment is shown in J). In K), results correspond to the quantification of the band intensities, and the signals measured were normalized to the signal obtained for actin. Results are expressed as the percentages of the signal intensity measured in MDMs treated or not with IFNα and transfected with siRNA relative to the signal measured in IFN-treated MDMs transfected with siCtrl (100%). Results are the means of 5 independent experiments performed with MDMs of 5 different donors. Each dot corresponds to an individual donor. Bars represent 1 SEM. Statistical significance was determined using the Mann-Whitney U-test (ns, P > 0.05; *, P < 0.05; ***, P < 0.001****, P < 0.0001). (L and M) After coculture of MDMs with ∆Vif (L) or WT-infected (M) CD4 T cells, MDMs were analyzed after coculture (6 h) or 6 days later by flow cytometry after intracellular Gag staining, and the results are expressed as the percentages of Gag+ cells related to the percentage measured in IFN-treated MDMs transfected with siCtrl (100%). Results are the means of 3 independent experiments performed with MDMs of 3 different donors. Each dot corresponds to an individual donor. Error bars represent 1 SEM. Statistical significance was determined using the Mann-Whitney U-test (ns, P > 0.05). (N-P) After 6 h of coculture of siRNA-transfected MDMs with ∆Vif- or WT-infected CD4 T cells, MDMs were analyzed by fluorescence microscopy, and the number of nuclei per Gag + MDM was quantified on at least 50 cells. Representative images are shown in (N). In O and P), results are expressed as the levels of MDM infection (infection index, right panels), or as the percentage of Gag + MDMs with 2, 3, 4 or more than 4 nuclei (left panels), quantified from a representative experiment of 3 independent experiments performed with MDMs of 3 different donors.

https://doi.org/10.1371/journal.ppat.1013130.g003

Since APOBEC3G is a potent restriction factor for HIV-1 replication in primary cells [49,51,52], we evaluated APOBEC3G restriction in HIV-1 cell-to-cell spreading from infected primary CD4 T cells to IFNα-stimulated macrophages. Differentiated MDMs were transfected with siRNA targeting APOBEC3G (siA3G), before pretreatment with 1,000 Unit/ml of IFNα, and then cocultured with infected T cells as schematized on S6A Fig. As evidenced in Fig 3J and 3K, the IFNα-induced APOBEC3G upregulation was reduced by siA3G, leading to expression equivalent to the unstimulated MDMs transfected with siCtrl (Fig 3J and 3K, left panel). By contrast, IFNα-induced APOBEC3A upregulation was not affected by siA3G (Fig 3J and 3K, right panel). APOBEC3G-depleted MDMs were then pretreated with IFNα, and cocultured with primary T cells infected with ΔVif or WT viruses. No differences in virus cell-to-cell infection of IFNα-stimulated MDMs were observed in APOBEC3G-depleted MDMs analyzed by flow cytometry, immediately after coculture or 6 days later (Fig 3L and 3M). Similarly, no differences in MGC formation were observed with IFNα-stimulated MDMs transfected with siA3G or siCtrl, either cocultured with WT or ΔVif infected T cells (Figs 3N–3P, S6B and S6C). These results show that cell-to-cell infection of IFNα-stimulated macrophages by formation of MGCs is not restricted by APOBEC3G. This finding also suggests that other IFN-stimulated APOBEC3 proteins do not affect cell-to-cell infection of macrophages by cell-cell fusion.

The T cell-derived nucleus of MGCs is transcriptionally active

To characterize the mechanisms responsible for bypassing the cellular restriction factors independently of the viral auxiliary proteins, as well as the resistance of MGC formation to IFN-I, we investigated the maintenance of the T-cell nucleus containing proviral DNA. Therefore, we first used the Jurkat-LTR-GFP cell-line, expressing GFP under the control of the viral promoter, as virus-donor T cells. About 80% of infected Gag+ Jurkat-LTR-GFP cells were co-expressing GFP (S7A Fig). We then cocultured infected Jurkat-LTR-GFP cells with MDMs for 24 h and the MGC formed expressing Gag were monitored for GFP expression (S7B Fig). As shown in Fig 4A, Gag + MGCs expressing GFP could be detected even 20 days after the initial coculture of infected T cells with MDMs. A large majority of the GFP+ cells were Gag+ (75–95%, Fig 4B) and contained several nuclei (Fig 4C and 4D). Representative experiments of the proportion of GFP + , Gag + and double Gag + /GFP+ cells quantified 12 and 20 days after coculture, are shown in S7C–S7F and S7G–S7J Fig, respectively. 12 days post-coculture, around 50% of the Gag + MGCs were GFP+ (S7E Fig), and these double Gag + /GFP+ cells were multinucleated (S7E and S7F Fig, right bars). Around 35% of the Gag + MGCs were GFP- (S7F Fig), and could correspond to MDMs infected with cell-free viruses released by the Gag + MGCs during the 12 days of culture [40]. In agreement, around 40% of the Gag + /GFP- cells were mononucleated (S7E Fig). Similar results were obtained on infected MGCs co-expressing Gag and GFP 20 days after the initial coculture (S7G–S7J Fig). Since GFP was expressed only from the integrated LTR-GFP of T cell nuclei, these results show that these nuclei remain transcriptionally active in most MGCs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Maintenance of transcriptionally-active T cell nuclei in MGCs formed upon cell-cell fusion.

Jurkat-LTR-GFP (A-D) or purified CD4 + T (E and F) cells were infected with HIV-1 NLAD8, and then cocultured with MDMs. After elimination of T cells, MDMs were analyzed immediately (24 h, or day 0) or cultured for different period of time (until day 12 or 20) before analysis by fluorescence microscopy after staining with anti-Gag (yellow brown) antibodies, phalloidin (F-actin, red), GFP or Gag (green), and Dapi (Nucleus, blue). The total number of nuclei (Dapi+) per Gag + MDM was quantified on at least 50 cells. (A and E) Representative images are shown. In B) Results correspond to the percentage of Gag+ cells among GFP + MDMs. Each dot corresponds to an individual donor. In C) Results are expressed as the percentage of Gag + /GFP + MDMs with 2, 3, 4 or more than 4 nuclei quantified from a representative experiment. In D) Results are expressed as the means of the nucleus number per Gag + /GFP + MDM, and represent the means of at least 4 independent experiments performed with MDMs of 4 different donors. Error bars represent 1 SEM. (E and F) Fluorescence microscopy analysis of the results obtained from a representative experiment performed on MDMs, just after 24 h coculture with primary CD4 T cells, or 6 or 12 days after the initial coculture and elimination of T cells. Cells were then stained with anti-CD3, anti-Gag, and DAPI (nuclei) and analyzed by confocal microscopy. Representative images are shown in E). In F), intracellular CD3 mean fluorescence intensities (MFI) were quantified as indicated in Materials and Methods. Each dot corresponds to 1 cell, and at least 30 cells were analyzed for each condition. Horizontal bars represent means + /- 1 SEM. Statistical significance was determined with the Mann-Whitney U-test (****, P < 0.0001).

https://doi.org/10.1371/journal.ppat.1013130.g004

To further confirm the transcriptional activity of the T cell nuclei in MGCs, we used purified primary CD4 T cells as virus-donor T cells. As previously, infected primary T cells were cocultured with MDMs for 24 h, and expression of the specific CD3 T cell marker was followed after T cell elimination, just after the coculture, and 6 and 12 days later. As shown in Fig 4E and 4F, the CD3 T cell specific marker was initially detected just after coculture, but also 12 days later, confirming that the T cell nuclei remain transcriptionally active in most MGCs.

HIV-1 cell-to-cell spread in macrophages is resistant to antiretroviral drugs

Since the T cell nucleus remains transcriptionally active in MGCs, we investigated whether this mode of infection might also impact on the susceptibility of antiretroviral drugs targeting early steps of virus replication. MDMs were pretreated with antiretroviral drugs before coculture with infected Jurkat cells, including zidovudine (AZT) and nevirapine (NVP) as nucleoside and non-nucleoside reverse-transcriptase inhibitors (NRTI and NNRTI, respectively), raltegravir (RAL) as integrase inhibitor (INSTI), and PF74 as a capsid-targeting compound [53,54]. After coculture, MDM infection was first analyzed by flow cytometry. As shown in Fig 5A (left panel), there was no effect of antiviral drugs on virus transfer when Gag + MDMs were analyzed just after coculture. Interestingly, only a partial decrease of Gag + MDMs, statistically significant compared to non-infected cells (NI), was observed when infection was analyzed 4 days later (central panel). By contrast, a total inhibition was observed when MDMs were infected by cell-free viruses (right panel). Similarly, when analyzed by fluorescence microscopy (Figs 5B–5D, S8A and S8B), MGCs were formed just after the coculture and 4 days later even in the presence of antiretroviral drugs. By contrast, a complete inhibition of the low level of MGCs formed was observed when MDMs pretreated with antiviral drugs were infected by cell-free viruses, and almost all the MDMs were mononucleated (Figs 5B and S8C–S8E). Accordingly, viral production was partially resistant to antiviral drugs when MDMs were infected by virus cell-to-cell transfer (Fig 5E), whereas it was abolished by cell-free infection (Fig 5F). These results confirm that bypassing the early steps of the viral life cycle through transfer of viral material and infected T cell nuclei, result in a significant resistance to antiretroviral drugs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Sensitivity to antiretroviral drugs of HIV-1 cell-free and cell-to-cell infection of macrophages.

(A) MDMs pretreated with AZT, NVP, RAL or PF74 were infected with cell-free viruses (right panel) or cocultured for 24 h with infected Jurkat cells and analyzed just after coculture (left panel) or 4 days later (middle panel) by flow cytometry after intracellular Gag staining. Results are the means of at least 5 independent experiments performed with MDMs from at least 5 different donors, and are expressed as the percentages of Gag + MDMs relative to that determined without drugs (mock, 100%). Each dot corresponds to an individual donor. Error bars represent 1 SEM. Statistical significance was determined using the Anova test, and P values were obtained by Dunnett’s post-test correction (ns, P > 0.05; *, P < 0.05; ***, P < 0.001; ****, P < 0.0001). (B-D) MDMs pretreated with PF74, AZT, NVP, and RAL or without (mock) were inoculated with cell-free viruses (CF) or cocultured (CTC) with infected Jurkat cells for 24 h. After T cell elimination, MDMs were analyzed by fluorescence microscopy just after the coculture, or cultured for 4 additional days with or without antiretroviral drugs. The total number of nuclei per Gag + MDM was quantified on at least 100 cells. Representative images are shown in (B). NI, MDMs cocultured for 6 h with noninfected CD4 T cells. In C), results are expressed as the percentage of Gag + MDMs with 2, 3, 4 or more than 4 nuclei quantified from a representative experiment. In D), results are expressed as the means of nucleus number per Gag + MGC, and represent the means of at least 4 independent experiments performed with MDMs of 4 different donors. (E and F) MDMs pretreated or not with antiretroviral drugs were cocultured with infected CD4 T cells (E, CTC) for 24 h or infected by cell-free viruses (F, CF). After elimination of the virus inoculum or T cells, virus production was analyzed 4 days later from MDM supernatants of 6 independent experiments performed with MDMs of 6 different donors. Results are expressed as the percentages of p24 production relative to that determined without drugs (mock, 100%). Each dot corresponds to an individual donor. Error bars represent 1 SEM. Statistical significance was determined using the Anova test, and P values were obtained by Dunnett’s post-test correction (ns, P > 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001).

https://doi.org/10.1371/journal.ppat.1013130.g005

Both lymphoid and myeloid nuclei of MGCs contain HIV-1 DNA

Since some T cell nuclei remain transcriptionally active in the Gag-positive MGCs, we questioned whether copies of viral DNA were present only in T cell nuclei or might be also found in myeloid nuclei. DNA-FISH experiments were performed to visualize viral DNA in MGC nuclei. As schematized in Fig 6A, Jurkat cells derived from a male with a T cell leukemia [55], or primary CD4 T cells purified from a male blood donor, were used as virus-donor T cells as previously, while MDMs differentiated from female donor monocytes were used as macrophage targets. To distinguish between lymphoid and macrophage nuclei, a probe targeting the KDMC5 X chromosome gene was employed [56]. Jurkat or primary T cells were infected and then cocultured for 6 or 24 h with MDMs, and DNA-FISH was performed just after cocultures or 4 days later. As expected, when female MDMs were cocultured with non-infected Jurkat cells, mononucleated non-infected MDMs exhibited 2 copies of the KDMC5 gene (Fig 6B and 6D, NI). By contrast, MGCs formed upon coculture with infected Jurkat cells or male primary T cells containing a single X chromosome (T) could be discriminated from the myeloid nuclei (M) (Fig 6B and 6D). When MDMs were stained with viral DNA probes generated from the full length proviral NLAD8 strain, almost all MGC T cell nuclei contained, as expected, fluorescent dots of HIV-1 DNA as soon as 6 h of coculture (S9A Fig). Interestingly, viral DNA dots were also detected in some myeloid nuclei as soon as 6 h of coculture (Fig 6B–6E), suggesting that viral DNA present in the cytoplasm of infected T cell can be translocated into the macrophage nuclei. Then, the number of myeloid nuclei containing viral DNA significantly increased after 24 h of coculture or 4 days later (Fig 6C and 6E). Pre-treatment of MDMs with antiretroviral drugs such PF74, AZT, NVP and RAL, resulted in a slight (at 6 h) or a significant (at 24 h of coculture and 4 days later) decrease of myeloid nuclei containing viral DNA (Fig 6F and 6G), showing the specificity of the viral DNA detection. These observations and findings show that both T cell and myeloid nuclei of MGCs can contain viral DNA copies, and can explain the high virus-production of MGCs.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. FISH analysis of HIV-1 DNA in lymphoid and myeloid nuclei of MGCs.

(A) Schematic representation of the experimental design for detection of viral DNA in MGC nuclei (created with Biorender). (B-E) MDMs were cocultured for 6 or 24 h with non-infected- or NLAD8-infected Jurkat (in panels B and C) or primary T (in panels D and E) cells, or cultured after elimination of T cells for 4 additional days before DNA-FISH analysis. MDMs were fixed, permeabilized, stained with Dapi, and then incubated after RNAse treatment with specific probes for detection of the KDM5C gene located on the X chromosome together with the HIV-NLAD8 probe for detection of the proviral DNA. Representative images of non-infected mononucleated (left images, NI) and infected multinucleated MDMs (6 or 24 h after coculture, or cultured for 4 additional days), are shown. Enlargements from the right images marked with a red rectangle are shown. The outlines of nuclei are indicated by dotted lines; T and M correspond to nuclei of T cell and myeloid origins containing a single and two X chromosomes, respectively. In C and E), quantification of the percentage of MGCs with MDM nuclei containing HIV-positive dots after 6 or 24 h of coculture, or 4 days later. The number of HIV-positive M-nuclei was quantified in at least 30 MGCs. (F and G) MDMs pretreated or not with antiretroviral drugs (PF74, AZT, NVP or RAL) were cocultured for 6 or 24 h with NLAD8-infected Jurkat (panel F) or primary CD4 T (panel G) cells, or cultured for 4 additional days in the presence of the drugs before DNA-FISH analysis. After elimination of T cells, MDMs were fixed, permeabilized, stained with Dapi, and then incubated with probes for detection of the KDM5C gene together with the HIV-NLAD8 probe. Representative images of infected multinucleated MDMs after 6 or 24 h of coculture, or 4 days later are shown in S9B and S9C Fig. Results are expressed as the percentage of MGCs HIV + M-nuclei containing HIV-positive dots after 6 or 24 h of coculture, or 4 days later, in non-treated (mock) or treated MGCs. The number of HIV-positive M-nuclei was quantified in at least 30 MGCs.

https://doi.org/10.1371/journal.ppat.1013130.g006

Ethics statement

Blood samples from anonymous healthy donors were purchased at “Etablissement Français du Sang” (EFS) (French National Blood Agency). Donors provided written informed consent to EFS at the time of blood collection (agreement #18/EFS/030). Samples were used in accordance with legal and ethical conditions we previously detailed in our previous work [42].

Plasmids and reagents

The proviral pNLAD8 and p89.6 plasmids were obtained from the AIDS Research and Reference Reagent Program, NIAID. The Nef-deleted p89.6 ∆Nef was obtained from Bernard Lagane (Inserm, Toulouse, France), and the pVSVg plasmid encoding the VSV-G envelope glycoprotein has been described [42]. The pNLAD8 plasmids deleted from each of the HIV-1 auxiliary proteins (∆Nef, ∆Vif, ∆Vpr, and ∆Vpx) were constructed from the wild-type pNLAD8 plasmid. Briefly, pNLAD8∆Vif was constructed through insertion of the AgeI-PfIMI fragment of the pNL4.3Vif-Vpr- plasmid (obtained from the AIDS Research and Reference Reagent Program) in the pNLAD8 previously digested by AgeI and PflMI. The pNLAD8∆Vpr was constructed through insertion of the PflMI-EcoRI fragment of the pNL4.3Vif-Vpr- plasmid in the pNLAD8 previously digested by PfIMI and EcoRI. The pNLAD8∆Vpu was constructed through insertion of the EcoRI-BamHI fragment of the pNL4.3Vpu- plasmid (obtained from the AIDS Research and Reference Reagent Program) in the pNLAD8 previously digested by EcoRI and BamHI. The pNLAD8∆Nef was construct by digestion of the pNLAD8 with XhoI, filled with Klenow before religation. The SERINC5-KO Jurkat cell lines (B1, B5, and F6 clones) were obtained from Christine Goffinet (Liverpool School of Tropical Medicine, United Kingdom), and have been described and characterized previously [47]. The following antibodies were used: RD1- or fluorescein isothiocyanate (FITC)-conjugated anti-Gag (clone KC57, Beckman Coulter); Alexa Fluor 647- or Fluor 555-conjugated phalloidin (Life Technologies); while anti-Vpr, -Vif, -Nef and -Vpu, as well as anti-APOBEC3G/A (anti-Apo17) antibodies were obtained from the AIDS Research and Reference Reagent Program; anti-Mx2 was purchased from Novus Bio; and anti-b-actin was purchased from Invitrogen [33]. The antiretroviral AZT, NVP, RAL and PU74 compounds were obtained from the AIDS Research and Reference Reagent Program, while IFNα2a was purchased from Miltenyi Biotech.

Cell culture

HEK293T and Jurkat cell lines were obtained from the ATCC, and were maintained in culture as described [42]. Monocytes were purified from blood of healthy donors using a CD14-positive selection kit (CD14 microbeads; Miltenyi) according to manufacturer’s guidelines. Monocytes were then differentiated into macrophages for 8 days in culture medium supplemented with 25 ng/ml of macrophage colony-stimulating factor (M-CSF) (Miltenyi) as described [43]. Human primary CD4 + T cells were isolated and maintained in culture as described [43].

Virus production and titration

Wild-type or deleted HIV-1 NLAD8 and 89.6 strains were produced in HEK293T cells as previously described [42]. The amounts of viral p24 produced were determined by enzyme-linked immunosorbent assay (ELISA; Innogenetics) as recommended by the manufacturer. Viral titers were determined using Jurkat cells and flow cytometry as described previously [36]. For production of viruses dedicated to the cell-free infection experiments, viral stocks were used at a multiplicity of infection (MOI) of 0.5 to infect Jurkat cells for 16 h. After washing, cells were cultured for another 24 h, and the amount of Gag p24 produced in the cell-culture supernatant was determined by ELISA as described [42].

Cell-free and cell-to-cell viral transfer and dissemination in MDMs

For cell-free infection of MDMs or primary CD4 T cells, viruses produced in infected Jurkat T cells were used at a concentration of 250 ng/mL of viral p24 for 10⁶ MDMs as described [43]. After 24 h of infection and elimination of the viral inoculum, the level of Gag + MDMs was analyzed by flow cytometry as previously described [42], just after the 24 h of infection or after culture for 4 or 6 additional days. For cell-to-cell virus infection assays of MDMs from infected Jurkat or primary purified CD4 T cell, cells were carried out as described previously [33,40]. Briefly, Jurkat and primary CD4 T cells were infected with virus stocks produced from HEK 293T cells at a MOI of 0.5 for 16 h, and then for another 24 h after an intermediate washing step. Where indicated, MDMs were pretreated before coculture for 1 h with antiretroviral drugs (AZT, NVP, RAL, and PF74) at the concentration of 10 ∆M, or overnight with IFNα2b at the indicated concentration (ranging from 100 to 10,000 U/mL), and then cocultured at a 2:1 cell ratio, without or with either antiretroviral drugs or IFN2αb, with infected T cells for 6 or 24h as described [42]. T cells were eliminated by washes in PBS and 4 mM EDTA-containing PBS [40]. The percentages of Gag + MDMs were analyzed by flow cytometry using the KC57 anti-Gag mAb (1/500) just after the coculture or after 4 or 6 additional days in culture [40]. Cell-culture supernatants were also harvested for quantification of the virus production by p24 ELISA as described [33,40]. Similarly, MDMs depleted of APOBEC3G after transfection of specific siRNA (Dharmacon) as described [33] were cocultured with primary CD4 T cells, and then analyzed by flow cytometry or for viral production in the cell-culture supernatant as described above.

Fluorescence microscopy

To visualize virus cell-to-cell transfer by cell-cell fusion, infected Jurkat or primary CD4 T cells were cocultured for 6 or 24 h with MDMs plated onto coverslips as described [40,42]. Where indicated, MDMs were pretreated before coculture with IFNα2b (ranging from 100 to 10,000 U/mL) or antiretroviral drugs (AZT, NVP, RAL, or PF74 at the concentration of 10 ∆M) as described above. After elimination of T cells, MDMs were directly analyzed after the coculture or cultured for different period of time, before fixation, blockage and staining with Dapi as described [42]. MDMs were then permeabilized and stained using KC57 FITC-conjugated anti-Gag antibody (1/200 dilution), and phalloidin-Alexa Fluor 647 (Molecular Probes), diluted in permeabilization buffer for 1 h. Coverslips were then washed with PBS and mounted on slides using 10 μl of Fluoromount (Sigma). Images were acquired on a spinning disk (CSU-X1M1; Yokogawa)-equipped inverted microscope (DMI6000; Leica) and were then processed using Fiji software (ImageJ; NIH). Quantitative image analysis and determination of Dapi+ nuclei were analyzed from images of at least 100 cells followed by processing using Fiji as described [33,40]. Quantitative CD3 expression was analyzed form images on at least 50 cells after staining with FITC-conjugated KC57 anti-Gag, phalloidin-Alexa Fluor 647 and PE-conjugated anti-human CD3 (Biolegend) using Fiji software by defining a region of interest using the F-actin staining and measuring the whole fluorescence intensity of the CD3 marker in Gag+ cells, with respect to noninfected cells as described [33,40].

Western blotting

For analysis of expression of auxiliary proteins in infected T cells, and APOBEC3G/A and MX2 in MDMs, cells (10⁶) were lysed in 100 μl of reducing Laemmli sample buffer (2x) supplemented with PhosSTOP (phosphatase inhibitor cocktail tablets; Sigma-Aldrich) and cOmplete (EDTA-free protease inhibitor cocktail; Sigma-Aldrich), and then boiled at 96°C for 10 min. Cell lysates were then resolved by SDS-PAGE as described [33], and analyzed by Western blotting using anti-Vif, anti-Nef, anti-Vpu, anti-Vpr, anti-HIV-1 p24 (Abcam; ab9071), anti-APOBEC3G/A (1/1000) and Mx2 (1/1000), and anti-β-actin (1/1000) antibodies.

DNA-FISH analysis

After 24 h of coculture of NLAD8-infected Jurkat cells with female MDMs pretreated or not with antiretroviral drugs (AZT, NVP, and PF74) at the concentration of 10 mM, T cells were eliminated, and MDMs were analyzed by DNA-FISH performed as previously described [56]. Briefly, the HIV-1 NLAD8 (plasmid pNLAD8) and X-linked KDM5C gene (BAC clone RP11-236P24) probes were labeled by Nick translation (Vysis kit) with Aminoallyl-dUTP-ATTO-550 and Aminoallyl-dUTP-ATTO-488 respectively (Jena Bioscience). For one coverslip, around 200 ng of each probe were prepared separately: HIV-1 probe was precipitated, resuspended in 10 μl of hybridization buffer (50% Formamide, 20% Dextran sulfate, 2X SSC, 2 mg BSA) and denatured at 75°C for 7 min while KDM5C probe was precipitated with 1μg of human Cot-1 DNA, resuspended in 10 μl of hybridization buffer, denatured at 75°C for 7 min and incubated for 45 min at 37°C to anneal all the repetitive sequences. The two probes were then pooled just prior to overnight hybridization with the coverslips. MDMs cultured on coverslips were fixed in 2% paraformaldehyde for 15 min, permeabilized in 1X PBS/0.5% Triton X-100 on ice for 7 min, and progressively dehydrated in ethanol. Coverslips were then incubated in 0.1 μg/μl RNase A (Thermofisher) in 1X PBS for 1h at 37°C. After two washes in 1X PBS, cells were then incubated in ice-cold 0.7%-Triton/0.1M HCl for 10 min on ice, rinsed twice in 2X SSC and denatured in 50% Formamide/2X SSC for 30 min at 80°C. After 3 washes in ice-cold 2X SSC, cells were then hybridized with probes overnight at 37°C in a dark and humid chamber. After 3 washes in 50% formamide/2X SSC, 3 washes in 2X SSC at 42°C, and DAPI counterstaining, coverslips were mounted on slides and visualized under fluorescence microscope. Images were acquired on a spinning disk (CSU-X1M1; Yokogawa)-equipped inverted microscope (DMI6000; Leica) and were then processed using Fiji software (ImageJ; NIH). Quantitative image analysis and determination of HIV+ nuclei were analyzed from images of at least 30 MGCs followed by processing using Fiji and Procreate illustrate specific signals found in nuclei of representative MGCs.

Statistical analysis

Statistics and curve fitting were performed using the GraphPad prism software. As mentioned in the figure legends, when comparisons were performed between more than 2 groups, ANOVA tests were used for statistical analyses, and P values were obtained by Dunnett’s post-tests correction to compare every mean to the control group mean. For all our experiments, graphics represent the results of at least three independent experiments (n = 3). Since each experiment was conducted using primary MDMs from different donors, the results for each experiment were normalized to the control group.