Vpr transcriptionally suppresses NF-κB through a pathway that does not require a functional glucocorticoid receptor. Vpr suppresses tumour necrosis factor-α (TNF-α)-induced NF-κB-mediated transcription in HeLa (a), CV-1 (b), Jurkat (c), U937 (d), PBL (e) and macrophage (f) cells. A total of 1 × 10⁶ cells were co-transfected with the indicated expression vectors and luciferase activity was measured. Total proteins were extracted from the HeLa and CV-1 cells, as indicated above, and the presence of Vpr was analysed by western blotting (WB) using specific Vpr antibodies. (g) Vpr suppresses NF-κB-dependent transcription in the absence of new protein synthesis. HeLa cells were treated with or without cycloheximide (CHX; 10 μg ml⁻¹) for 30 min, and transfected with pNF-κB–Luc plasmids (2 μg). After 6 h of transfection, the medium was replaced with or without a CHX-containing medium, and cells were incubated for 24 h before RNA extraction. TNF-α (5 ng ml⁻¹) was added to cells and incubated for 6 h before RNA extraction. Total RNA (20 μg) was resolved on 1.2% formaldehyde gel and analysed by northern blotting using a radiolabelled luciferase DNA as a probe. The loading markers were 28S rRNA level. (h) Nuclear levels of RelA (p65) in both functional GR (HeLa) or non-functional GR (CV-1) cells. Cells were treated with or without rVpr (10 pg) and stimulated with TNF-α (5 ng ml⁻¹). Cells were collected post-treatment, as indicated, and were immunoblotted with a polyclonal antibody against NF-κB (p65). As a control, Vpr effects on GR were also analysed with antibody against GR and anti-PCNA as an internal control for nuclear loading. (i) IKKβ kinase activity in cells treated with rVpr and TNF-α. Serum-starved HeLa cells (1.5 × 10⁶) were treated with TNF-α (5 ng ml⁻¹) at indicated time points with or without rVpr (10 pg) and were then subjected to kinase activity, as described in Methods. Top, kinase assay (KA) revealed by autoradiography; bottom, the immunoprecipitates were also subjected to western blotting (WB) to determine the amount of precipitated proteins. (j) I-κBα phosphorylation and its turnover in rVpr-treated cells was measured from lysates simulated with the conditions described above. (k) HeLa cells were treated with TNF-α at indicated time points in the presence or absence of rVpr (10 pg). Cells were collected at indicated time points and nuclear proteins (10 μg) were incubated in 96-well plates lined with oligonucleotides specific for NF-κB (p65) for the ELISA-based transcription-factor assay. Values and bars represent mean (n = 3) and SD. Scale bars in b, c and i represent 10 μm.

Vpr interacts with PARP-1 and inhibits its nuclear localization through a GR interaction-dependent pathway. (a) Vpr prevents the nuclear migration of poly(ADP-ribose) polymerase-1 (PARP-1). HeLa or Jurkat T cells were stimulated with tumour necrosis factor-α (TNF-α) with or without rVpr (10 pg and 1 pg). PARP-1 localization was detected by protein blotting with an anti-PARP-1 polyclonal antibody. Immunoblotting was also performed using an anti-PCNA antibody as an internal control for nuclear loading. (b–c) Cellular localization of PARP-1. HeLa cells were treated with or without rVpr (10 pg) in the presence or absence of anti-Vpr and were stimulated with or without of TNF-α for 3 h. Cells were fixed and stained with an anti-PARP-1 antibody, followed by FITC-conjugated goat anti-rabbit secondary antibody, as described in Methods (b). Localization of eIF4E was also performed to show the specificity of Vpr towards PARP-1. The third panel is eIF4E staining in leptomycin B (LMB; 50 nM)-treated cells to show trafficking (c). Scale bars, 10 μM. (d–e) Biochemical fractionation of nuclear (Nu) versus cytoplasmic (Cy) contents were performed from TNF-α or TNF-α + Vpr-treated cells. Protein samples were separated by SDS–PAGE and analysed by western blot with a polyclonal antibody against PARP-1 or eIF4E antibody for cytosolic versus nuclear expression. (f–h) Vpr and PARP-1 form a complex in multiple cells. CV-1, peripheral blood leukocyte (PBL) and macrophage cells were transfected with a vector control (pDNA) or with pVpr (4 μg) and analysed for co-immunoprecipitation (IP) with respective antibodies. WB, western blot. (i) Dexamethasone is insufficient to regulate PARP-1, whereas the ability of Vpr to regulate PARP-1 localization is sensitive to mifpristone. HeLa cells were treated with Mock, rVpr (10 pg), dexamethasone (Dex; 1 μM), mifpristone (Mif; 1 μM), rVpr (10 pg) + mifpristone (1 μM) or Vpr (10 pg) + anti-Vpr (1:200). Twenty-four hours post-treatment, cells were stimulated with or without 5 ng ml⁻¹ of TNF-α for 3 h, and immunofluorescence was performed using an antibody against PARP-1 followed by FITC-conjugated goat anti-rabbit secondary antibody. Scale bars, 10 μM. (j–k) The interaction of Vpr with PARP-1 is sensitive to mifpristone. HeLa or PBL cells were transfected with vector or pVpr (4 μg) in the presence or absence of mifpristone (1 μM). Protein lysates were immunoprecipitated with an anti-Vpr polyclonal antibody and analysed via protein blotting with anti-PARP-1, glucocorticoid receptor (GR) and Vpr antibodies. Western blots were also performed for PARP-1 expression for the input cell lysate that was extracted from the vector or Vpr-transfected samples.

The GR interaction with Vpr is both necessary and sufficient to recruit PARP-1. (a–c) Expression plasmids (Vpr, glucocorticoid receptor (GR) and poly(ADP-ribose) polymerase-1 (PARP-1)) (2 μg) were used for coupled in vitro transcription/translation reactions and were carried out as described in Methods. (a) Immunoprecipitation (IP) of the in vitro-translated proteins were performed with anti-Vpr, anti-GR and anti-PARP-1 antibodies. Equal amounts of in vitro-translated proteins were mixed in the following groups: (b) Vpr + PARP-1, GR + PARP-1, Vpr + GR or (c) Vpr + GR + PARP-1, and interactions were performed as described in Methods. (d–e) A glutathione S-transferase (GST) pull-down assay with recombinant forms of each protein was performed as described in Methods. (d) Represents a Coomassie blue staining of the gels, whereas (e) represents the western blot (WB) analysis of the pull-down assay with the indicated antibodies. Input represents 25% of total protein used for pull-down; flow-through represents 10% of unbound material. K, relative molecular mass in thousands. (f) Small interfering RNA (siRNA)-mediated knockdown of GR. Three different GR target siRNA expression vectors were constructed as described in Methods. (g) GR is required for Vpr and PARP-1 interaction in cells. GR-21 stable clone cells were established as described in Methods and transfected with pVpr (4 μg) and co-immunoprecipitation assays were performed with the indicated antibodies. (h) Vpr, PARP-1 and GR reside as a single complex in vivo. Equal amounts of total protein in whole-cell extracts that were prepared from HeLa cells were transfected with mock or pVpr–His + pGR + pPARP-1, respectively, and were bound to Ni-NTA columns, eluted and immunoprecipitated for GR and PARP-1 with Vpr–His-tagged eluted protein (bottom). Vpr–His-tagged eluted proteins were immunoprecipitated with anti-PARP or anti-GR antibodies and blotted back for the other.

Recruitment of PARP-1 to the Vpr–GR complex is necessary for Vpr-mediated transcriptional suppression of NF-κB. HeLa (a), CV-1 (b), Jurkat (c) and U937 (d) cells were transfected with the nuclear factor of κB (NF-κB)-dependent reporter plasmid with RelA plasmids and with pCMVLacZ (1 μg) with pVpr (4 μg) in the presence or absence of mifpristone (Mif; 1 μM), and luciferase assays were conducted as described in Methods. (e) Glucocorticoid receptor (GR) is necessary for the transcriptional suppression induced by Vpr. HeLa cells were transfected with the NF-κB-dependent reporter plasmid (2 μg), with 5 μg of short hairpin RNA (shRNA) pGR-91 or shRNA pGR-21 or shRNA control plasmid with or without Vpr (4 μg). Luciferase activity was measured as described in Methods. Asterisks (**) over a bar indicates statistically significant results (p < 0. 005) whereas the hash symobl (#) indicates results that are not statistically significant (p < 0.01). (f) Vpr H71Y interacts with GR but fails to recruit poly(ADP-ribose) polymerase-1 (PARP-1). Expression plasmids (Vpr and different Vpr mutants, GR and PARP-1) (2 μg) were used for coupled in vitro transcription/translation reactions and interaction action assays were carried out as described in Methods. IP, immunoprecipitation; WT, wild type. (g) The H71Y Vpr mutant fails to prevent PARP-1 nuclear localization. Total nuclear (Nu) and cytosol (Cy) protein extracts of HeLa cells were prepared 48 h post-transfection from wild-type Vpr (Vpr-Wt) or mutant Vpr (Vpr-H71Y)-transfected cells and were western blotted for PARP-1 antibody and anti-PCNA as an internal control for nuclear loading. (h) Analysis of Vpr and differing mutants for NF-κB-mediated transcription was measured as described in Methods. HeLa cells were transfected with pLuc–NF-κB (5 μg), pCMVLacZ (1 μg), with or without Vpr-Wt and Vpr mutants (2 μg), and luciferase activity in whole-cell lysates was assayed as described in Methods.

The PARP-1–Vpr–GR interaction is relevant in vivo. (a) Groups of 10 female, Balb/C mice for each group aged 6–12 weeks were sensitized by intravenous (i.v) injection with 20 mg D-galactosamine for 30 min prior to challenge with 10 μg of LPS or SEB toxin. Mice were challenged with rVpr (100 μg) in the presence or absence of 10 μM mifpristone (Mif), 1 h after SEB introduction via an intravenous method. We examined the survival of the SEB-challenged animals. (b) Immunoblot analysis of Vpr in the spleen post-rVpr injection. Total cellular proteins were prepared from the splenocytes challenged with SEB or SEB + Vpr. Protein lysates were immunoprecipitated with anti-Vpr antibody, followed by western blot (WB) analysis with an anti-Vpr antibody blotted back for Vpr. Lysates from the splenocytes were also loaded and blotted for actin as an internal control for immunoprecipitation. (c) Vpr-mediated suppression of cytokine secretion in mice challenged with SEB, SEB + rVpr or SEB + rVpr + mifpristone. Serum from ten mice per group were harvested prior to SEB treatment and serum from mice in the tested groups was also harvested. Serum tumour necrosis factor-α (TNF-α), interleukin (IL)-12 and IL-1β levels were determined by standard cytokine ELISA, as described in Methods. Results are expressed as pg ml⁻¹ ± SD of individual determinants. (d) Vpr prevents poly(ADP-ribose) polymerase-1 (PARP-1) nuclear localization in vivo. Total nuclear (Nu) and cytosolic (Cy) protein extracts were prepared from the challenged splenocytes and analysed by western blot with a polyclonal antibody against PARP-1 for cytosolic versus nuclear location. Dex, dexamethasone.