a. Effect of drugs targeting endosomal TLR signaling or HIV reverse transcription in PBMCs. PBMCs were cocultivated with HIV-infected MT4C5 cells, exposed to FLUAV, or to CpG (a TLR9 agonist) during 12 h, in the absence or presence of the indicated drugs. NVP (25 µM) is a reverse transcriptase inhibitor. Bafilomycin A1 (Bafilo, 125 nM) is an inhibitor of vesicular acidification. A151 (5 µg/mL) is a TLR antagonist. Nevirapin (NVP 25 µM) is a reverse transcriptase inhibitor. IFN levels are expressed as the % of the signal obtained without drugs. Mean+sd of 3 independent experiments is shown. *p<0.05 (Kruskal-Wallis). b-d. Viral proteins required for the recognition of HIV-expressing cells by PBMCs. b. Schematic representation of the experimental system. HeLa cells are transfected with various HIV mutants, cocultivated with PBMCs, and levels of IFN released in supernatants are measured 22 h later. c. Defective viruses trigger IFN production by PBMCs. HeLa cells were transfected with the indicated proviral mutants. Supernatants were harvested and analyzed for the presence of infectious virus, after normalization for Gag p24 levels. Env-deleted (ΔEnv), non-fusogenic Env (HIV F522Y), reverse transcriptase (RT-), integrase (IN), protease (PR), Rnase H (RH), and nucleocapsid (NC44) mutants are not or poorly infectious (left panel). Levels of IFN in supernatants, after coculture of tranfected HeLa cells with PBMCs. IFN levels are expressed as a percentage of the signal obtained with WT HIV (right panel). d. Expression of Env alone does not trigger IFN production. HeLa-Env cells (stably expressing a functional Env glycoprotein complex) were either not infected (NI) or infected with ΔEnv or WT viruses (pseudotyped with VSV), then processed and analyzed as in c. c-d: Mean+sd of 2-3 independent experiments is shown. *p<0.05 (Kruskal-Wallis).

a. Activation of Gen2.2 requires a direct contact with HIV-infected cells. Shaking cocultures of Gen2.2 and HIV-1-infected MT4C5 (left panel), or separating infected MT4C5 cells and target Gen2.2 in a transwell chamber (right panel) impairs IFN production, whereas recognition of FLUAV by Gen2.2 is not affected. Mean+sd of 3 independent experiments is shown b. Inhibition of IFN production by Bafilomycin A1. Gen2.2 cells were cocultivated with HIV-infected MT4C5 cells, exposed to FLUAV, or to CpG (a TLR9 agonist) during 24 h, in the absence or presence of Bafilomycin A1 (Bafilo 125 nM). IFN levels are expressed as the % of the signal obtained without drugs. c. Role of TLR7. Gen2.2 cells were transduced with lentiviral vectors expressing shRNAs against TLR7 (shTLR7) or an irrelevant target (shCTRL). Levels of TLR7 mRNA in transduced cells were measured by RT-PCR. Data are normalized to GAPDH mRNA and expressed as relative levels of mRNA compared to shCTRL cells. Mean+sd of 3 independent experiments are shown (left panel). IFN production in TLR7-silenced and control Gen2.2 cells, cocultivated with HIV-infected MT4C5 or stimulated with FLUAV or CpG for 40 h. IFN levels are expressed as a percentage of the signal obtained with shCTRL Gen2.2 cells.

a. Schematic representation of the experimental system. HIV-1 infected T cells are mixed with PBMCs, and levels of IFN released in supernatants are measured 12-24h later. b. IFN release in cocultures of HIV-infected T cells with PBMCs. MT4C5 lymphoblastoid T cells (left panel) or primary CD4+ T lymphocytes (right panel) were used as donors. Whole PBMCs, PBMCs depleted of pDCs (PBMCs-pDCs), and purified pDCs where used as target cells. Targets were either left non stimulated (NS), cocultivated with non-infected (NI) or HIV-infected (NL4-3 strain) MT4C5 cells (MT4C5-HIV), exposed to cell-free HIV virions (free HIV) or Influenza virus (FLUAV). IFN production in supernatants was measured in a bioassay, after 12 h of culture. Donor to target cell ratio were 1∶2 for PBMCs and PBMCs-pDCs, and 2∶1 for pDCs. c. Dose response analysis of IFN release. Whole PBMCs were incubated with MT4C5 cells displaying increasing % of infected (Gag+) cells. IFN production was measured after 24 h of coculture (orange histrograms). The levels of living MT4C5 cells at the beginning of the coculture, determined by flow cytometry (forward and side scatters gating), are shown in blue. d. Sensing of cells infected with various HIV isolates by PBMCs. MT4C5, infected with NL4-3 (NL), NLAD8 (AD8) and with two primary isolates (BON and 132W) were cocultivated with PBMCs and IFN was measured 24 h later. With all isolates, about 25% of infected MT4C5 cells expressed Gag before coculture. e. Role of HIV envelope glycoproteins. MT4C5, infected with wild-type (HIV), Env-deleted (ΔEnv), or a non-fusogenic Env mutant (HIVF522Y), and with similar levels of Gag+ cells, were cocultivated with PBMCs and IFN was measured 12 h later (left panel). IFN levels are expressed as the % of the signal obtained with FLUAV. MT4C5 viability was similar with WT, ΔEnv, and F522Y HIV at the beginning of the coculture. The efficiency of HIV capture by PBMCs was assessed by measuring the % of Gag+ PBMCs by flow cytometry (right panel). Mean+sd of 3-5 independent experiments is shown, * p<0.05 (Kruskal-Wallis).

a. Gen2.2 cells express HIV receptors and are sensitive to HIV infection. Surface expression levels of CD4, CXCR4 and CCR5 are shown on the upper left panel. Gen2.2 cells were then exposed to HIV particles, NL4-3 strain, at two doses (0.5 and 5 ng/0.1 ml Gag p24/10⁶ cells). Viral replication was assessed by following the appearance of Gag+ cells (upper right panel) and Gag p24 release in supernatants (lower left panel). IFN production was measured in the same supernatants (lower right panels). Data are representative of 3 independent experiments. b. Schematic representation of the coculture system. HIV-1 infected T cells are mixed with Gen2.2 cells and levels of IFN released in supernatants are measured 12–22 h later. c. Sensing of HIV-infected cells by Gen2.2 and role of viral envelope glycoproteins. MT4C5, infected with wild-type (HIV), Env-deleted (ΔEnv), or a non-fusogenic Env mutant (HIVF522Y), and with similar levels of Gag+ cells, were cocultivated with PBMCs and IFN was measured 22 h later (left panel). When stated, HIV-infected primary CD4+ cells were used as donors, instead of MT4C5 cells. Cell-free HIV virions (500 ng/ml p24) do not stimulate Gen2.2 cells. The efficiency of HIV capture by Gen2.2 was assessed by measuring the % of Gag+ cells by flow cytometry (right panel). Mean+sd of 10 independent experiments is shown, *p<0.05 (Kruskal-Wallis). d. Visualization of HIV capture by Gen2.2 cells. Gen2.2 cells were labeled with CFSE (green) and cocultured with MT4C5 infected with wild-type (HIV), ΔEnv and F522Y mutants. Cells were harvested after 1 h and stained for HIV Gag (red). Representative fields are shown.

a. Role of CD4 and CXCR4 in 293T cells. Activity of the IFNβ promoter in parental 293T cells, and in 293T cells expressing CXCR4 (293T CXCR4), or CD4 and CXCR4 (293T-4X4). These cells were transfected with IFNβ-luciferase reporter plasmid, and cocultivated for 16 h with MT4C5 cells expressing wild-type, ΔEnv, or F552Y HIV. The fold induction of luciferase activity, compared to non-stimulated cells is shown. The paramyxovirus Sendai virus (SeV) was used as a positive control (right panel). b. Role of IRF3. 293T-4X4 cells were transfected with an irrelevant (CTRL) or an anti IRF3 siRNA. 48 h later, cells were cocultivated with HIV-infected MT4C5 cells and assayed for IFN-β promoter activity. Left panel: IRF3 and NFkB levels, assessed by western blot, in control (CTRL) and silenced 293T-4X4 cells. Right panel: IFN-β promoter activity is expressed as a percentage of the signal obtained with control cells (CTRL). SeV, which signals through IRF3, was used as a control. Mean+sd of 3 independent experiments is shown. *p<0.05 (Kruskal-Wallis). c–d. Automated quantification of IRF3 nuclear translocation. 293T-4X4 cells were cocultivated with non-infected (NI) or HIV-infected Far-red dye stained MT4C5 cells, exposed to SeV, or left non-stimulated (NS) for 16 h. Cells were then stained with anti-IRF3 Abs and nuclei were visualized with DAPI. Automated quantification of IRF3 nuclear translocation on gated 293T-4X4 (Far-red dye negative) cells was performed using the ImageStream technology. c. Representative images of Bright field, Dapi (blue) IRF3 (green), DAPI/IRF3 composite images, for cells with either a diffuse intracellular (non translocated, low similiarity value) or a nuclear localization (translocated, high similarity value) are shown. d. 293T-4X4 cells with nuclear-localized IRF3 have higher similarity values, because the aspect of IRF3 and DAPI staining is similar. The percentage of cells with similarity values above an arbitrary value of 1.5 (where most, if not all of the cells fall in the translocated category) is indicated on each histogram. Data are representative of 4 independent experiments.