Vpx rescues HIV-1 transduction of dendritic cells from the antiviral state established by type 1 interferon

Abstract

Background: Vpx is a virion-associated protein encoded by SIVSM, a lentivirus endemic to the West African sooty mangabey (Cercocebus atys). HIV-2 and SIVMAC, zoonoses resulting from SIVSM transmission to humans or Asian rhesus macaques (Macaca mulatta), also encode Vpx. In myeloid cells, Vpx promotes reverse transcription and transduction by these viruses. This activity correlates with Vpx binding to DCAF1 (VPRBP) and association with the DDB1/RBX1/CUL4A E3 ubiquitin ligase complex. When delivered experimentally to myeloid cells using VSV G-pseudotyped virus-like particles (VLPs), Vpx promotes reverse transcription of retroviruses that do not normally encode Vpx.

Results: Here we show that Vpx has the extraordinary ability to completely rescue HIV-1 transduction of human monocyte-derived dendritic cells (MDDCs) from the potent antiviral state established by prior treatment with exogenous type 1 interferon (IFN). The magnitude of rescue was up to 1,000-fold, depending on the blood donor, and was also observed after induction of endogenous IFN and IFN-stimulated genes (ISGs) by LPS, poly(I:C), or poly(dA:dT). The effect was relatively specific in that Vpx-associated suppression of soluble IFN-β production, of mRNA levels for ISGs, or of cell surface markers for MDDC differentiation, was not detected. Vpx did not rescue HIV-2 or SIVMAC transduction from the antiviral state, even in the presence of SIVMAC or HIV-2 VLPs bearing additional Vpx, or in the presence of HIV-1 VLPs bearing all accessory genes. In contrast to the effect of Vpx on transduction of untreated MDDCs, HIV-1 rescue from the antiviral state was not dependent upon Vpx interaction with DCAF1 or on the presence of DCAF1 within the MDDC target cells. Additionally, although Vpx increased the level of HIV-1 reverse transcripts in MDDCs to the same extent whether or not MDDCs were treated with IFN or LPS, Vpx rescued a block specific to the antiviral state that occurred after HIV-1 cDNA penetrated the nucleus.

Conclusion: Vpx provides a tool for the characterization of a potent, new HIV-1 restriction activity, which acts in the nucleus of type 1 IFN-treated dendritic cells.

Figure 1

SIV_MAC virus-like particles (VLPs) rescue HIV-1 transduction of human monocyte-derived dendritic cells (MDDCs) from pretreatment with type I ifN. MDDCs were incubated for 24 h with 10 ng⁄mL IFN-α (A) or 10 ng⁄mL IFN-β (B). The cells were then treated for 3 h with media or VSV-G-pseudotyped SIV_MAC-251 VLPs, followed by challenge with a VSV-G-pseudotyped HIV-1_NL4-3 GFP reporter virus. The percent GFP-positive cells was determined by flow cytometry 72 h after transduction. Error bars represent ± standard deviation (SD) (n = 3). In each case, one representative example of at least three independent experiments is shown.

Figure 2

SIV_MAC VLPs rescue HIV-1 infectivity from pretreatment of MDDC with pattern recognition receptor (PRR) agonists. (A) MDDCs were incubated for 24 h with recombinant type I interferon (10 ng⁄mL IFN-α, 10 ng⁄mL IFN-β), or PRR agonists as indicated: 100 ng⁄mL LPS, 25 μg⁄mL poly(I:C) with no lipid carrier, or 2 μg⁄mL poly(dA-dT). Then, cells were treated for 3 h with media or VSV-G-pseudotyped SIV_MAC-251 VLPs, followed by challenge with a VSV G-pseudotyped HIV-1_NL4-3 GFP reporter virus (A) or with a CCR5-tropic, HIV-1_NL4-3 GFP reporter virus (B). The percent GFP-positive cells was determined by flow cytometry 72 h after addition of the reporter virus. Error bars represent ± SD (n = 3). In each case, one representative example of at least three independent experiments is shown.

Figure 3

Among VLP constituents, Vpx is necessary and sufficient to rescue HIV-1 from type I IFN. (A) MDDCs were treated with LPS for 24 hrs, then treated for 3 hrs with media or the indicated VSV-G-pseudotyped HIV-2_ROD or SIV_MAC-251 VLPs, and finally challenged with a VSV-G-pseudotyped HIV-1_NL4-3 GFP reporter virus. Infectivity was measured by flow cytometry. (B) As indicated, 293T cells were co-transfected with a codon optimized SIV_MAC251 vpx expression plasmid and HIV-1 GFP reporter vectors bearing either wild-type Gag or Gag with an engineered Vpx binding motif (DPAVDLL). Proteins from the cell lysate and from virion preparations were separated by SDS-PAGE and then immunoblotted with anti-Vpx or anti-p24 antibodies. (C) MDDCs treated with IFN-β for 24 h and were then challenged with VSV-G-pseudotyped HIV-1 GFP reporter vectors with wild-type HIV-1 Gag or HIV-1 Gag bearing the engineered Vpx binding motif (DPAVDLL). Both HIV-1 reporter vectors were produced in the presence of empty pcDNA3.1 plasmid or pcDNA3.1 containing a codon-optimized SIV_MAC-251 vpx cDNA. Data are representative of one of at least three independent experiments. Error bars represent ± SD (n = 3).

Figure 4

Vpx rescues HIV-1, but not SIV_MAC or HIV-2, from the type I IFN response in MDDC. (A) MDDCs were treated with the indicated compounds for 24 h, and then challenged with VSV-G-pseudotyped, vpx⁺ or Δvpx SIV_MAC GFP reporter virus. (B, C) MDDCs were treated with the indicated compounds for 24 h, then treated with either VSV-G-pseudotyped vpx⁺ or Δvpx SIV_MAC-251 VLPs, and then challenged with either VSV-G-pseudotyped SIV_MAC-239 (B) or HIV-1_NL4-3 (C) GFP reporter viruses. (D, E) MDDCs were treated with LPS, then treated with either media or VSV-G pseudotyped HIV-1_NL4-3 or SIV_MAC-239 VLPs (containing all accessory genes) for 3 h, and then challenged with either VSV-G-pseudotyped SIV_MAC-239 (D) or HIV-1_NL4-3 (E) GFP reporter viruses. Data are representative of one of at least three independent experiments. Error bars represent ± SD (n = 3).

Figure 5

SIV_MAC Vpx association with DCAF1 (VPRBP) is dispensable for Vpx-mediated rescue of HIV-1 from the antiviral state. (A) 293T cells were transfected with FLAG-tagged DCAF1 and either wild type SIV_MAC-251 Vpx or SIV_MAC-251 Vpx containing the indicated alanine-substitution mutations that disrupt associated with DCAF1. Immune complexes were isolated from clarified, 0.5% CHAPSO detergent lysates using anti-FLAG antibody conjugated to Protein G magnetic beads. Panels show immunoblots (IB) of the immunoprecipiated (IP) proteins (top panels) and immunoblots of the inputs (bottom panels). (B) Immunoblots of wild-type Vpx and the indicated mutants incorporated into SIV_MAC-251 VLPs (top panels) and expression in the 293T producer cells (bottom panels). (C) MDDCs were treated with LPS, then treated with SIV_MAC-251 VLPs containing wild-type Vpx or the indicated mutants, and challenged with an HIV-1_NL4-3 GFP reporter virus. Data represent one of at least three independent experiments. Error bars represent ± SD (n = 3).

Figure 6

DCAF1 (VPRBP) knockdown does not prevent Vpx rescue of HIV-1 from the antiviral state in MDDCs. MDDCs were transduced with lentiviral knockdown vectors targeting either DCAF1, or a control RNA, in the presence of SIV VLPs. DCAF1 KD and control KD cells were then treated with IFN-β for 24 hrs, and lysates were probed in immunoblots with antibodies against the indicated proteins (A), or cells were challenged with a VSV-G-pseudotyped HIV-1_NL4-3 GFP reporter virus (B). (C) DCAF1 KD and control KD MDDCs were treated with LPS for 24 h, and challenged with either VSV-G-pseudotyped HIV-1_NL4-3 or SIV_MAC-239 GFP reporter viruses. Data represent one of at least three independent experiments. Error bars represent ± SD (n = 3).

Figure 7

SIV_MAC Vpx rescues HIV-1 from the antiviral state in MDDC prior to establishment of the provirus. MDDCs were treated with IFN-α for 24 h, and then treated with SIV_MAC VLPs or media for 3 h, and finally challenged with a VSV-G-pseudotyped HIV-1_NL4-3 GFP reporter virus. Total DNA was extracted from 5 × 10⁶ MDDCs and qPCR was performed for HIV-1 full-length linear reverse transcription products (A), 2-LTR circles (B), and provirus (C). (D) Data from A, B, C, represented as (fold-rescue of HIV-1 by Vpx from IFN-α treatment) divided by (fold-rescue of HIV-1 by Vpx in the absence of exogenous IFN). Data represent one of at least three independent experiments. Error bars represent ± SEM (n = 4).