HIV-2 inhibits HIV-1 gene expression via two independent mechanisms during cellular co-infection

Abstract

Twenty-five years after the first report that HIV-2 infection can reduce HIV-1-associated pathogenesis in dual-infected patients, the mechanisms are still not well understood. We explored these mechanisms in cell culture and showed first that these viruses can co-infect individual cells. Under specific conditions, HIV-2 inhibits HIV-1 through two distinct mechanisms, a broad-spectrum interferon response and an HIV-1-specific inhibition conferred by the HIV-2 TAR. The former could play a prominent role in dually infected individuals, whereas the latter targets HIV-1 promoter activity through competition for HIV-1 Tat binding when the same target cell is dually infected. That mechanism suppresses HIV-1 transcription by stalling RNA polymerase II complexes at the promoter through a minimal inhibitory region within the HIV-2 TAR. This work delineates the sequence of appearance and the modus operandi of each mechanism.

Keywords: HIV-1; HIV-2; HIV-2 TAR; RNA polymerase II; dual infection; interferon response.

Fig 1

Effect of the order of HIV-1 and HIV-2 infection on viral replication in co-infected TZM-bl cells. (A) TZM-bl cells were infected with replication-competent HIV-1 and HIV-2 (multiplicity of infection [MOI] of 2 each virus titrated in TZM-bl cells). In co-infection experiments i, ii, and iii (rows 2, 3, and 4), HIV-1 was added at t₀, t₀, and t₁₂, (zero, zero, and 12 h post-infection by the first virus), whereas HIV-2 was added at t₁₂, t₀, and t₀. All experiments were terminated at t₄₈. Images were acquired by an automated Cytation 5 epifluorescence microscope. Sample images from three independent experiments for each condition are shown. Anti-Vpu (Alexa Fluor 550) and anti-Vpx (Alexa Fluor 488) antibodies are shown in red and green, respectively; DNA was stained with DAPI (blue). Scale bar represents 100 µm. (B) Quantitation of HIV-1 and HIV-2 in co-infected cells. The proportion of Vpu-positive (HIV-1-infected) cells (filled bars) and the proportion of Vpx-positive (HIV-2-infected) cells (empty bars) for dual-infection conditions i, ii, and iii [described in (A)] are shown. Averages from three independent experiments are shown ±standard errors of the means.

Fig 2

Effect of HIV-2 infection on HIV-1 transduction and gene expression. (A) Genome organization of DuoFluo HIV-1 pNL4.3 plasmid used for HIV-1 transduction experiments. mKO2 fluorescent protein expression is under the control of the eIF-1α core promoter and is expressed in transduced cells. eGFP protein expression is under the control of the HIV-1 LTR promoter. (B) Dual-fluorescence HIV-1 transduction outcomes. Schematic indicating possible outcomes of transduction with the HIV-1 reporter virus shown in (A). (C) Dual-infection images in TZM-bl cells. Representative images for each time point of infection (HIV-1 transduction + HIV-2 infection) in TZM-bl cells are shown. Scale bar represents 100 µm. (D) Effects of HIV-1 transduction + HIV-2 infection on eIF1α-driven mKO2 expression in TZM-bl cells. The percentage of HIV-1-transduced cells was determined from the images shown in (C). (E) Effects of HIV-1 transduction + HIV-2 infection on HIV-1 LTR-driven eGFP expression in TZM-bl cells. The percentage of HIV-1-transduced cells with an active provirus was determined from the images shown in (C), n = 3 independent experiments. The Sidak’s (corrected one-way ANOVA test) multiple comparison test was used.

Fig 3

Interferon (IFN) β mRNA analysis by RT-qPCR in TZM-bl cells. Data were plotted as fold-change of IFNβ mRNA relative to the negative control and normalized to GAPDH mRNA. Infected HIV samples were compared to IFNβ-treated control cells. TZM-bl cells were incubated with 1,000 IU of recombinant IFNβ for 24 h before harvesting. n = 3 independent experiments. The Tukey’s (corrected one-way ANOVA test) multiple comparison test was used.

Fig 4

HIV-1/HIV-2 dual infection in Vero cells. (A) Dual-infection images in Vero cells. Images were obtained as described in Fig. 2B and C. Sample images for each time point of infection in Vero cells are shown. Scale bar = 100 µm. (B) Effects of dual transduction on eIF1α-driven mKO2 expression in Vero cells. Quantification was performed as described in Fig. 2D. (C) Effects of dual transduction on HIV-1 LTR-driven eGFP expression in Vero cells. Quantification was performed as described in Fig. 2E. Error bars are standard errors (n = 3 independent experiments). Sidak’s multiple comparison test was used in (B) and (C).

Fig 5

Effects of TAR-2 on HIV-1 expression. (A) TAR-2 RNA expression levels. Full-length pTAR-2 was transfected into TZM-GFP cells at 2 µg for 48 h. Another set of TZM-GFP cells was concurrently infected with HIV-2 at an MOI of 0.8 for 48 h. Transfected and infected cells were harvested, and intra-cellular TAR-2 RNAs were quantified with RT-qPCR using specific primers. (B) Strategy for transfection. (C) Virus release analysis with ELISA p24. Culture supernatants were analyzed with an in-house ELISA p24 technique. Quantifications were performed using a p24 standard curve. For each cell line, the pTZU6 control was set at 100%. pTAR-2 was normalized to it. (D) HIV-1 Gag RNA levels in transfected cells. Transcription from HIV-1 short transcripts per nanogram RNA-transfected pNL4.3 plasmid was measured with RT-qPCR using the primers described in Tables 1 and 2. The quantity of Gag RNA was divided by the number of viral DNA templates to get the ratio of viral Gag RNA to viral DNA templates. (E) HIV-1 short transcription analysis during transfection. HIV-1 short RNA transcripts (first 200 nucleotides from the 5′ end of the virus) were quantified using specific primers. The unpaired t-test was used in (A, D, and E).

Fig 6

TAR-2 effects on HIV-1, MLV, and RSV gene expression. (A) Strategy outline for HIV-1/MLV reporter virus experiments. (B) Assessment of TAR-2 effects on MLV. In the presence of GFP to E2-Crimson MFI HIV-1 and MLV infectivities full-length (FL) pTAR-2 or pTZU6 control, VSV(G)-pseudotyped eGFP expressing HIV-1, and tomato-expressing MLV reporter viruses were produced and used to transduce HEK 293 FT cells. Transduced cells were analyzed with flow cytometry, representative plots are shown. (C) TAR-2 effects on MLV versus HIV-1. The percentage of HIV-1-transduced cells and MLV-transduced cells with reporter viruses made in the presence of the control pTZU6 or the FL pTAR-2 vectors is plotted. (D) Strategy outline for HIV-1/RSV reporter virus experiments. (E) TAR-2 effects on RSV versus HIV-1. Representative plots of flow charts from transfection experiments performed with a dual-reporter eGFP/E2-Crimson RSV vector for 24 h in HEK 293 FT cells. (F) The ratio of the mean fluorescence intensity (MFI) of the eGFP signal to the MFI of the E2-Crimson was calculated. The ratio eGFP MFI to E2-Crimson MFI in the control pTZU6-transfected cells was set at 100%, those of the FL pTAR2-transfected cells were normalized to the pTZU6 control.

Fig 7

Effects of TAR-2 on HIV-1 during infection. (A) TAR-2 RNA expression during HIV-2 infection. TZM-bl cells were infected with replication-competent HIV-2 and incubated for 24, 48, and 72 h prior to harvesting. Infected cells were harvested and processed for RT-qPCR to quantify genomic RNA and total transcripts during infection. (B) HIV-1 total DNA in HIV-infected TZM-bl cells. Infection was performed as described in Materials and Methods, and total viral DNA levels were determined by qPCR. (C and D) Quantification of HIV-1 Gag RNA levels in HIV1-infected TZM-bl cells. Long (C) or short/abortive (D) transcript levels were quantified by RT-qPCR. Averages from three independent experiments are shown with SEM indicated. The Tukey’s multiple comparison test was used in (B), (C), and (D).

Fig 8

TAR-2 blocked HIV-1 transcription elongation during infection. (A) Active P-TEFb occupancy 1 kbp downstream of the HIV-1 TSS during HIV infection. Single HIV1-infected treated or not with DRB (P-TEFb inhibitor), and dual-infected TZM-bl cells were processed with chromatin immunoprecipitation using anti-threonine 186-phosphorylated P-TEFb complex, followed with qPCRs using specific primers (Table 1) spanning the location described above. (B) Ser2-P RNAPII occupancy at 1 kbp downstream of the HIV-1 TSS during HIV infection. Same as (A), except for the antibody used for the immunoprecipitation reaction (anti-serine 2-phosphorylated RNAP II). (C) Ser5-P RNAPII occupancy 1 kbps downstream of the HIV-1 TSS during HIV infection. Same as (A), except for the antibody used for immunoprecipitation (anti-serine 5-phosphorylated RNAP II). (D) Active P-TEFb occupancy at the HIV-1 TSS during HIV infection. Same samples from (A) were amplified with a different set of primers spanning a region encompassing the HIV-1 TSS in the LTR. (E) Ser2-P RNAPII occupancy at the HIV-1 TSS during HIV infection. Samples from (B) were amplified with the set of primers described in (D). (F) Ser5-P RNAPII occupancy at the HIV-1 TSS during HIV infection. Same as (C) were amplified with the set of primers described in (D). The Tukey’s multiple comparison test was used.