Transcription factor profiling of syncytia elicited by HIV-Env. (A) Systematic analysis of transcription factors activated in syncytia. HeLa CD4 cells were transfected with luciferase constructs placed under the control of the indicated transcription factor. 24 h after transfection, cells were cultured alone (100% control value) or in the presence of HeLa Env cells, and the expression of luciferase was monitored 24 h later. Asterisks indicate a significant (P < 0.01; Student's t test) increase in the luciferase activity (mean ± SD; n = 3). (B) Mutual relationship among p53 and NF-κB. Cells were cotransfected with luciferase expression plasmids controlled by p53, NF-κB, or AP1 plus empty vector (pcDNA3.1) or the IκB super-repressor (IKSR) or a DN p53 (p53H273), and the increase in luciferase activities in syncytia (vs. single cells) was quantified. Results are representative of three independent experiments. Asterisks indicate significant induction of luciferase constructs.

Rules governing the activation of p53 in Env-elicited syncytia. (A) Genetic inhibition of p53-dependent transcription. HeLa CD4 cells were cotransfected with a p53-inducible GFP construct, together with vector only (pcDNA3.1), a dominant-negative (DN) Cdk1 mutant, DN p53 (p53H273), or IKSR. After 24 h, the cells were either left alone (single cells) or cocultured with HeLa Env cells for 36 h, followed by cytofluorometric determination of GFP. (B) Pharmacological inhibition of p53-dependent transcription. HeLa CD4 cells were transfected with a p53-inducible GFP construct. After 24 h, cells were cultured in the presence of the indicated agents, in the presence or absence of HeLa Env cells for 24 h, followed by determination of the frequency of GFP-expression single cells or syncytia. (C) Inhibition of karyogamy by IKSR. HeLa CD4 cells were transfected with the indicated constructs as in A, and the frequency of cells exhibiting nuclear fusion was scored upon coculture with HeLa Env cells. (D) Inhibition of karyogamy by NF-κB inhibitors. Syncytia were treated with various drugs as in B, and the frequency of karyogamic cells was scored. (E) Effect of IKSR and DN p53 on cyclin B1, karyogamy, phosphorylation of IκB or p53, and apoptotic parameters. Syncytia were generated as in A, after transfection with pcDNA3.1 (Control), IKSR, or DN p53, followed by immunofluorescence detection of cyclin B1. The frequency of syncytia exhibiting an increase in cyclin B immunostaining was determined in three independent experiments after 24 h of coculture. Moreover, the frequency of syncytia exhibiting positive cytoplasmic or nuclear IκBS32/36P staining, p53S15P or p53S46P, mitochondrial cytochrome c release, or nuclear apoptosis was determined. Asterisks indicate significant inhibitory effects (P < 0.01; paired Student's t test; mean ± SD; n = 3).

Identification of p53 target genes by macroarray. Labeled cDNA from single cells or syncytia was hybridized to filters containing p53 target genes. The signal/background ratio was determined for all spots (A, with β-actin as an internal control) and the change in mRNA expression was determined as described in Materials and Methods (B).

Functional impact of Puma on the Bax/Bak-mediated mitochondrial membrane permeabilization. (A) Effect of the Puma antisense oligonucleotide on the conformational change of Bax and Bak. Untreated control syncytia or syncytia generated in the presence of a Puma-specific antisense thiorate oligonucleotide were stained with antibodies specific for the NH₂ terminus of Bax (red) or Bak (green) and counterstained with Hoechst 33324. Note that these antibodies only recognize the active, apoptotic conformation of Bax and Bak. (B) Quantitation of the effect of the Puma antisense oligonucleotide, as compared with a scrambled control oligonucleotide, determined 36 h after syncytium formation. The insert confirms effective down-regulation of Puma as determined by immunoblot. (C) Effect of IKSR on the activation of Bax. Cells were either transfected with vector only (pcDNA3.1) or with IKSR, 24 h before the generation of syncytia, and the indicated parameters were assessed in triplicate. (D) Effect of SN50 and cyclic pifithrin-α as compared with its control (SN50M). The indicated agents were added to HeLa Env and HeLa CD4 at the beginning of coculture, and the frequency of cells exhibiting activation of Bax or Bak, cytochrome c release and nuclear apoptosis was assessed. (E) Effect of Bax and Bak small interfering RNA (siRNA) on the apoptosis of Env-elicited syncytia. Cells were transfected with small interfering RNA oligonucleotides 24 h before fusion, and the indicated parameters were assessed. The inset demonstrates the impact of siRNA on Bax and Bak expression, as determined by immunoblot. This experiment was repeated three times with similar results.

Induction of Puma in HIV-1–infected patients. (A) Immunocytochemical localization of Puma in lymph node biopsies of HIV-1–infected donors as compared with a healthy control. Similar results were obtained with three pairs of uninfected and HIV-1–infected donors (>10⁵ copies/ml). (B) Frequency of PUMA⁺ cells among defined subsets of mononuclear cells. Patients (n = 4) and controls (n = 4) with the same clinical characteristics as in A were subjected to two-color immunofluorescence stainings, allowing for the determination of the percentage (mean ± SD) of cells expressing Puma among CD4⁺ or CD8⁺ lymphocytes, as well as among CD14⁺ monocytes. (C) Correlation between PUMA expression and viral titers. The frequency of PUMA⁺ cells was determined among total PBMCs of 10 untreated HIV-1⁺ patients. Each dot represents one patient. Calculations were performed following the Pearson method. (D) Expression of Puma in purified circulating CD4⁺ lymphocytes from uninfected control donors, naive HIV-1–infected donors, and HIV-1 carriers undergoing successful HAART. (E) Longitudinal study of individual patients. Whole PBMCs were drawn from freshly diagnosed (untreated) patients with viral titers >10⁵ copies/ml, as well as several months after HAART. In these patients, HAART led to a reduction of retroviral titers to levels <5,000 within 4 wk. Immunoblot detection of Puma and β-tubulin (loading control) is shown.

Phosphorylation of IκBα and p53 in Env-elicited syncytia. (A–C) Representative immunofluorescence microphotographs of prekaryogamic and karyogamic syncytia (36 h) with antibodies specific for IκBαS32/36P (A), p53S15P (A and B), p53S46P (C), or cytochrome c (B and C). The frequency of the indicated phosphorylation events was determined both among single cells (SCs) and karyogamic (KG) or prekaryogamic syncytia. (D) Immunoblot confirmation of the phosphorylation of IκBαS32/36P, p53S15P, and p53S46P in syncytia. The immunodetection of GAPDH was performed to control equal loading. (E) Kinetic analysis of phosphorylation events. Syncytia (obtained by coculture of HeLa Env and HeLa CD4 cells during the indicated interval) were subjected to immunostainings as aforementioned, and the percentage of cells exhibiting the indicated parameters was quantified among the total population of syncytia.

Phosphorylation of IκBα and p53 in lymph nodes from HIV-1 carriers. Lymph nodes from age- and sex-matched controls or untreated HIV-1 carriers were stained for the immunocytochemical detection (red) of IκBα (S32/36)P (A) and p53S46P (B). Nuclei are counterstained in blue. Similar results indicating a scattered p53S46 staining and staining of mitotic cells with IκBα(S32/36)P were found in three independent asymptomatic, untreated HIV-1–infected donors with a viremia of >10⁵ copies/ml.

Up-regulation of Puma at the mRNA and protein levels. (A) Quantitative RT-PCR was used to determine the relative abundance of the indicated transcripts. (B) Protein blot detection of Puma expression. Syncytia of different ages (18 or 36 h of coculture), generated after transfection with vector only or a p53 DN construct, were subjected the immunodetection of Puma (∼21 kD), Bax, and GAPDH as a loading control.

Induction of Puma by Env and HIV-1 in vitro. (A) Induction of Puma in U937 cells cocultured with HeLa Env cells. U937 cells were mixed at different ratios with HeLa Env cells just before the preparation of cell lysates for immunoblot detection of Puma (Control) or coculture for 36 h in the absence or presence of the p53 inhibitor cyclic pifithrin-α (5 μM). The percentage of dead U937 cells was determined by trypan blue exclusion. (B) Induction of Puma by recombinant gp120 protein in U937 cells. U937 were exposed to the indicated gp120 protein (at 500 ng/ml) and cell death, Puma, and GAPDH expression were determined after 4 d. (C) Cooperation between gp120 and p53 to induce Puma in U937 cells. Cells were mock transfected or transfected with wild-type p53. 1 d later, the indicated gp120 proteins were added, and the expression of 53 and Puma were determined 48 h later. (D) Induction of Puma by recombinant gp120 protein in Jurkat cells as in B. (E) Effect of cyclic pifithrin-α. Jurkat cells were treated for 4 d with 500 ng/ml gp120 protein and/or 10 μM cyclic pifithrin-α, followed by determination of Puma expression. (F) Induction of Puma by infection of primary lymphoblasts in vitro. CD4⁺ lymphoblasts from a healthy donor were infected with HIV-1_LAI/IIIb for the indicated period, and proteins (40 μg/lane) were subjected to immunoblot determination of p53 phosphorylation and Puma expression. (G) Puma induction in CD4⁺ lymphoblasts from a healthy donor infected with a clinical HIV-1 isolate. 5 d after infection, cells were subjected to immunohistochemical detection of Puma. Uninfected cells served as a negative control. Results typical for five independent experiments are shown.