Vpu-deficient HIV strains stimulate innate immune signaling responses in target cells

Abstract

Acute virus infection induces a cell-intrinsic innate immune response comprising our first line of immunity to limit virus replication and spread, but viruses have developed strategies to overcome these defenses. HIV-1 is a major public health problem; however, the virus-host interactions that regulate innate immune defenses against HIV-1 are not fully defined. We have recently identified the viral protein Vpu to be a key determinant responsible for HIV-1 targeting and degradation of interferon regulatory factor 3 (IRF3), a central transcription factor driving host cell innate immunity. IRF3 plays a major role in pathogen recognition receptor (PRR) signaling of innate immunity to drive the expression of type I interferon (IFN) and interferon-stimulated genes (ISGs), including a variety of HIV restriction factors, that serve to limit viral replication directly and/or program adaptive immunity. Here we interrogate the cellular responses to target cell infection with Vpu-deficient HIV-1 strains. Remarkably, in the absence of Vpu, HIV-1 triggers a potent intracellular innate immune response that suppresses infection. Thus, HIV-1 can be recognized by PRRs within the host cell to trigger an innate immune response, and this response is unmasked only in the absence of Vpu. Vpu modulation of IRF3 therefore prevents virus induction of specific innate defense programs that could otherwise limit infection. These observations show that HIV-1 can indeed be recognized as a pathogen in infected cells and provide a novel and effective platform for defining the native innate immune programs of target cells of HIV-1 infection.

Fig 1

Vpu-deficient HIV-1 strains stimulate innate immune signaling. (A) PMA-differentiated THP-1 cells were infected for 24 h with a multiplicity of infection of 0.5 or 1 of wild-type JR-CSF or YU2, mock treated, or untreated (Ø) and then analyzed by immunoblotting for ISG56, p24, and actin as a loading control. (B and C) Jurkat (R5 positive) and PBMC-derived macrophages (Mϕ) tested as described for panel A with virus at a multiplicity of infection of 1. (D) ISG56, MX1, HIV-1 p24, and actin protein levels in vaginal mucosal T cells infected ex vivo with SeV, HIV-1_JR-CSF, or HIV-1_YU2. Virus was added to cultures and allowed to infect for 24 h, after which extracts were prepared and subjected to immunoblot analysis. (E) Differentiated THP-1 cells were challenged with increasing multiplicities of infection of JR-CSF or Vpu-deficient JR-CSF A/C, with SeV challenge used as a control. Protein lysates were blotted as described for panels A to C and quantified; fold induction above that for mock-treated cells is shown. Representative immunoblots of at least 3 experiments are shown.

Fig 2

Vpu-deficient HIV-1 strains stimulate an IRF3-dependent ISG response. (A) PMA-differentiated THP-1 cells were either infected with JR-CSF, YU2, or SeV, mock treated, or untreated (Ø) for 8 h and harvested for immunoblot analysis. Cell lysates were probed for total IRF3 (T-IRF3) and beta-actin as a loading control (multiplicity of infection = 1). (B) PMA-differentiated THP-1 cells were either mock treated or infected with YU2 or SeV for 8 h. Cells were fractionated into cytoplasmic (C) or nuclear (N) compartments and analyzed by immunoblotting for phosphorylated IRF3 (P-IRF3). Lamin B and tubulin mark loading of the nuclear and cytoplasmic fractions, respectively. (C) PMA-differentiated THP-1-knockdown cell lines for IRF3, IPS-1, or MyD88 or a nontargeting control were infected with SeV or HIV-1_YU2 or mock treated. Cells were treated and analyzed as described for Fig. 1E. (D) Comparison of non-targeting-vector (NTV) and knockdown (KD) cell protein levels for IRF3, IPS-1, and MyD88. Cells were harvested and immunoblotted for the targeted protein. Quantification is shown. Error bars are SDs.

Fig 3

Partial rescue of Vpu-deficient HIV-1 by knockdown of IRF3. Wild-type or IRF3-knockdown (KD) PMA-differentiated THP-1 cells were infected with JR-CSF or JR-CSF A/C at a multiplicity of infection of 1 for 4 h and then washed to remove free virus. Samples were harvested for immunoblotting of p24 and actin levels as a loading control at 8 and 24 h of infection. (A) Normalized 8-h p24 levels were set to 100 for each virus and compared to the 24-h values. (B) Percent increase of viral p24 values at 24 h postinfection, comparing virus growth in IRF3-knockdown cells and wt. **, P < 0.005. Mean values with SDs are plotted.

Fig 4

Vpu-deficient HIV-1 stimulates the specific expression of a wide range of innate antiviral immune modulators. JR-CSF, Vpu-deficient JR-CSF A/C, and YU2 virus strains were used to infect PMA-differentiated THP-1 cells for 8 or 16 h. Synthetic HCV PAMP RNA was transfected as a positive control for IRF3-dependent innate immune stimulation. RNA was isolated and subjected to microarray analyses targeted at an experimentally derived set of ISGs and innate immune-associated genes. (A) Heat map cluster of selected significantly regulated innate immune and ISGs in log₂ scale. (B) Fold induction of known anti-HIV-1 ISGs and restriction factors during HIV-1 infection with or without Vpu at 8 h postinfection. Mean of four biological replicates. *, P < 0.0005; **, P < 0.000001 (FDR-adjusted P values). Results for all JR-CSF samples were not significant, with P values of >0.05. (C) Fold induction of known NF-κB-responsive genes during HIV-1 infection with wt or Vpu-deficient strains at 16 h postinfection. Mean of four biological replicates. Results for all samples were significant, with FDR-adjusted P values of <0.000001, with the exception of JR-CSF samples TNFRSF14 (*, P < 0.0025) and TNFSF14 (**, not significant).