Bacterial DNA Protects Monocytic Cells against HIV-Vpr–Induced Mitochondrial Membrane Depolarization

Cells of the monocytic lineage, including macrophages, serve as a major source of HIV reservoir by supporting accumulation of latent yet potentially replication competent virus (1–3). These cells are refractory to antiviral drug penetration and allow low levels of viral replication during and after highly active antiretroviral therapy (1, 2, 4–6). Human macrophages, unlike primary monocytes, are productively infected in vitro and in vivo with HIV and can survive HIV-induced cytopathic effects (3, 5–8). Several factors have been implicated in induction of resistance to HIV-mediated apoptosis in monocytic cells, including downregulation of cell-surface death receptors and differential regulation of apoptosis-related genes (9–11). However, the exact mechanism enabling macrophages to withstand HIV cytopathic effects is poorly understood.

Because primary monocytes are refractory to productive HIV infection in vitro (8), we and others have employed C-terminal viral protein R (Vpr) peptide 52–96 aa (referred to hereafter as Vpr) as an apoptosis-inducing agent to understand mechanisms of resistance to HIV-induced apoptosis (9, 10, 12, 13). In the absence of therapy, HIV, either through direct infection or through its components such as Vpr, contributes to mitochondrial dysfunction by altering mitochondrial membrane potential (14–16). Vpr specifically has been shown to interact with a component of mitochondrial permeability transition pore complex, adenine nucleotide translocator (ANT), which leads to the release of mitochondrial apoptogenic factors such as cytochrome c, apoptosis-inducing factor (AIF), and second mitochondria-derived activator of caspases (SMAC) (12, 17, 18). However, the role of Vpr-ANT interaction in apoptosis is controversial, as small interfering RNA (siRNA)–mediated ANT knockdown did not affect Vpr’s ability to permeabilize mitochondria (19). We have previously shown that primary monocytes undergo apoptosis when exposed to Vpr. However, monocytes stimulated with CpG oligodeoxynucleotide (ODN) or bacterial DNA and the differentiated macrophages exhibit resistance to the apoptotic effects of Vpr, an effect that was attributed to the antiapoptotic inhibitor of apoptosis (c-IAP)-2 gene (9, 10). However, the mechanism by which c-IAP-2 confers resistance to Vpr-induced apoptosis in CpG ODN/bacterial DNA-stimulated monocytic cells remains unknown.

c-IAP-1/2 play a key role in innate immune signaling and cell survival by maintaining NF-κB activation downstream of TLRs and nucleotide-binding and oligomerization domain–like receptors (20–23). c-IAP-2 associates with the TNF-α receptor complex-I via TNFR-associated factor (TRAF)-1/2 to mediate activation of the classical NF-κB pathway (21). Furthermore, it functions as an E3 ubiquitin ligase that maintains constitutive ubiquitination of the receptor-interacting protein (RIP)-1 (24). This suppresses formation of the death-inducing complex-II, the activation platform for caspase-8 that promotes cell death via the extrinsic pathway (23, 24). From a signaling perspective, TLR4/TLR9 activation leads to the formation of a signaling complex comprised of MyD88, TRAF-6, c-IAP-1, and c-IAP-2. In this complex, TRAF-6 undergoes autoubiquitination and recruitment of TGF-β–activated kinase-2, TGF-β–activated kinase-3, TGF-β–activated kinase-1, and inhibitor of NF-κB kinase subunit α– inhibitor of NF-κB kinase subunit β–Nemo complexes and eventually leading to transcription of inflammation-related genes (21).

IAPs, in addition to being a modulator of immune signaling, regulate caspases (21, 25–27). c-IAP-1/2, unlike X-linked IAP, physically interact with caspases but cannot inhibit caspase activity by binding alone. Indeed, c-IAP-1/2 were reported to polyubiquitinate caspase-3 and caspase-7 using E3-ligase activity (28–30), leading to their degradation and nondegradative inactivation downstream of mitochondria. In the intrinsic pathway of apoptosis, c-IAPs bind and inactivate SMAC released following mitochondrial membrane depolarization (31). However, the involvement of c-IAP-2 or TRAF-1/2 in maintaining mitochondrial integrity and controlling release of cytochrome c or other caspase-independent, proapoptotic proteins such as AIF remains unknown. In this study, we show for the first time, to our knowledge, that Vpr induces mitochondrial depolarization through TRAF-1/2 degradation and by activation of proapoptotic caspase-8, Bid, and Bax. Furthermore, CpG ODN/bacterial DNA–induced c-IAP-2 prevented Vpr-induced mitochondrial depolarization and release of cytochrome c and AIF through prevention of TRAF-1/2 degradation and sequential inhibition of caspase-8, bid, and bax upstream of mitochondria.

Blood drawn from healthy volunteers was used to isolate PBMCs. PBMCs isolated by density gradient centrifugation over Ficoll–Hypaque (Pharmacia Biotech, Piscataway, NJ) were subjected to Automacs negative selection (Miltenyi Biotec, Auburn, CA) as per the manufacturer’s instructions. THP-1 cells, a promonocytic cell line derived from a patient with acute monocytic leukemia (32), were obtained from the American Type Culture Collection (Manassas, VA). Cells were cultured in IMDM-10 (Sigma-Aldrich, St. Louis, MO) supplemented with 10% FBS (Invitrogen, Grand Island, NY), 100 U/ml penicillin, 100 μg/ml gentamicin, 10 mmol HEPES, and 2 mmol glutamine (all from Sigma-Aldrich). Second mitochondrial activator of caspases (SMAC) mimetic AEG-730 was a gift from Dr. Korneluk (Apoptosis Research Centre, Ottawa, ON, Canada). CpG-B ODN 2006 (Hycult Biotech, Plymouth Meeting, PA), Lyovec Escherichia coli DNA and GpC control ODN (Invitrogen), and TNF-α (Invitrogen) were purchased. The concentrations of CpG ODN (5 μmol) and bacterial DNA (25 μg/ml) used in this study were based on the dose-response kinetics for protection against apoptosis induced by Vpr (9). The following signaling inhibitors were used: chloroquine, EGTA, Bcl2 inhibitor HA14-1, and cycloheximide (Sigma-Aldrich); W-7 hydrochloride, KN-93, and SKF-96365 hydrochloride (Calbiochem, San Diego, CA) (33, 34); and caspase-8–specific inhibitor z-Leu-Glu-His-Asp-fluoromethyl ketone (R&D Systems, Minneapolis, MN). All other chemicals used for electrophoresis and immunoblot analysis were obtained from (Sigma-Aldrich).

For generation of monocyte-derived macrophages (MDMs), briefly, PBMCs were resuspended in serum-free medium (5 × 10⁶/ml) and cultured in 12-well polystyrene plates (BD Biosciences, Mississauga, ON, Canada) for 3 h to adhere to the plate. The nonadherent cells were washed off, and adherent cells were cultured for another 6 d in IMDM-10 supplemented with 10 ng/ml M-CSF (9, 10) (R&D Systems, Minneapolis, MN). THP-1–derived macrophages (THP-1 Macs) were generated as described before (10) by differentiating THP-1 cells (5 × 10⁵/ml) with 20 ng/ml PMA (Sigma-Aldrich) for 2 d.

The Vpr peptides were synthesized by automated solid-phase synthesis and purified by reverse-phase HPLC (>95%) (Invitrogen). The amino acid sequence of Vpr peptide is ⁵²DTWAGVEAI IRILQQLLFI HFRIGCRHSR IGVTRQRRAR NGASRS⁹⁶. The mutant Vpr peptide with three arginine to alanine mutations at sites R73, R77, and R80 is indicated in boldface letters in the above sequence (Genemed Synthesis, San Antonio, TX). Cells were cultured for 12 h in serum-free media before treatment with Vpr peptides, as described previously (10, 35).

Loss of Rhodamine 123 (10) and MitoTracker Green FM (36) (both from Invitrogen Molecular Probes) fluorescence was used as an indicator of mitochondrial membrane depolarization. Briefly, THP-1 cells and MDMs were loaded with 100 ng/ml Rhodamine 123, and primary monocytes were loaded with 200 nmol/ml MitoTracker in serum-free medium and incubated for 30 min at 37°C after treatment with Vpr. Subsequently, the cells were washed with PBS and analyzed by flow cytometry for green fluorescence.

Apoptotic cells exhibiting sub-G₀ DNA content were identified and analyzed by flow cytometry using propidium iodide (PI) staining of permeabilized cells, as described previously (10). Briefly, cells (1.0 × 10⁶/ml) were washed twice with PBS containing 1% FBS, fixed with methanol for 15 min at 4°C, and treated with 1 μg/ml RNAse A (Roche Applied Science, Laval, Quebec, Canada) followed by staining with 50 μg/ml PI (Sigma-Aldrich) at 4°C for 1 h. The DNA content was then analyzed by flow cytometry (BD FACSCanto equipped with BD FACSDiva software v5.0.3; BD Biosciences). Apoptosis was also measured by staining cells (1.0 × 10⁶/ml) with FITC-labeled Annexin-V (Molecular Probes, Eugene, OR) for 15 min at room temperature in the dark followed by flow cytometry and data analysis using Win-MDI version 2.8 software (J. Trotter, Scripps Institute, San Diego, CA).

Briefly, total proteins from cell lysates were subjected to SDS-PAGE followed by transfer onto polyvinylidene difluoride membranes (Bio-Rad, Hercules, CA). The membranes were probed with Abs specific for c-IAP-1, c-IAP-2, Ca²⁺/calmodulin-dependent protein kinase II (CaMK-II), Bid, full-length caspase-8, Bax, TRAF-1, TRAF-2, Mcl-1, BcL-X_L, and GAPDH (all from Cell Signaling Technology, Danvers, MA), cleaved caspase-8 (Santa Cruz Biotechnology, Santa Cruz, CA) followed by donkey anti-rabbit secondary polyclonal Abs conjugated to HRP (Amersham Bioscience, Montreal, Quebec). All immunoblots were visualized by ECL (Amersham Bioscience), as described previously (10).

The siRNAs against c-IAP-2 and CaMK-IIγ (Santa Cruz Biotechnology) were used as described previously (9). Briefly, the control nonsilencing siRNA (Qiagen, Mississauga, ON, Canada) and c-IAP-2 siRNA were incubated with 3 μl Fugene6 (Roche Applied Science) at a 1:3 ratio (μg/μl) in 100 μl serum-free medium for 30 min at room temperature before adding to THP-1 cells (0.25 × 10⁶/0.5 ml). After 5 h of transfection, cells were transferred into complete medium and stimulated with CpG ODN (5 μmol) for 48 h. Similarly, CaMK-IIγ siRNA was transfected using Fugene6 (Roche Applied Science) transfection reagent at a 1:3 ratio (μg/μl) as described above. After 5 h of transfection, cells were transferred into complete medium and incubated for 24 h followed by stimulation with CpG ODN (5 μmol ) for 12 h. Mcl-1, BclX_L, TRAF-1, and TRAF-2 siRNAs were transfected using Mirus TransIT-TKO transfection reagent (Mirus, Madison, WI) at a 20:2 ratio (nmol/μl) in 50 μl serum-free medium for 30 min at room temperature before adding to (0.25 × 10⁶) THP-1 cells suspended in 250 μl complete medium in 24-well plates. After 5 h of incubation, an additional 200 μl complete medium was added to the wells followed by a 48-h incubation, following which the cells were stimulated with CpG ODN (5 μmol) for 12 h. The caspase-8–, Bid-, and Bax-specific siRNAs (Santa Cruz Biotechnology) were transfected using the Mirus TransIT-TKO transfection reagent (Mirus) as described above for 24–48 h using the following siRNA to transfection reagent ratio: caspase-8 (10 nmol/1 μl), Bid (40 nmol/4 μl), and Bax (10 nmol/1–4 μl). Following transfections, the cells were treated with 1.5 μmol Vpr for 5 h (mitochondrial permeabilization measurement) or 24 h (apoptosis measurement).

Caspase-8 activation was detected using an FITC-IETD-FMK caspase-8 activation kit (Calbiochem, San Diego, CA) according to the manufacturer’s protocol. This kit provides a cell-permeable FITC-conjugated caspase-8 inhibitor that binds with cleaved caspase-8. The measure of FITC fluorescence reflects the degree of caspase-8 activation (37). Briefly, monocytes and THP-1 cells, treated with 1.5 μmol Vpr for 2 h and 5 h, respectively, suspended in 300 μl culture medium and 1 μl FITC-IETD-FMK (caspase-8 detection kit; Calbiochem) were added. After 1 h of incubation at 37°C, cells were washed and analyzed for caspase-8 activation by flow cytometry.

Briefly, THP-1 cells and monocytes were washed with PBS and plated on poly-l-lysine–coated (Sigma-Aldrich) coverslips in 12-well plates for 20 min at 37°C. MDMs and THP-1 macrophages were generated as mentioned above and cultured directly on glass coverslips in six-well plates. The adhered cells were fixed and quenched as described previously (9). This was followed by permeabilization in 0.1% Triton in PBS (10 min) and overnight incubation at 4°C with either: 1) Alexa Fluor 488–conjugated cytochrome c and FITC-conjugated AIF-specific Abs (Santa Cruz Biotechnology); or 2) Tom20, Bax 6A7, and cleaved caspase-8–specific primary Abs (all from Cell Signaling Technology) overnight at 4°C. Cells treated with non–Alexa Fluor or FITC-conjugated primary Abs alone were washed with blocking solution three times and incubated with the secondary Ab conjugated with Alexa Fluor 488 (following Bax 6A7 and cleaved caspase-8) or Alexa Fluor 688 (following Tom 20) (Invitrogen) for 1 h at room temperature. MDMs and THP-1 macrophages were treated with 200 nmol/ml MitoTracker green (Invitrogen Molecular Probes) for 30 min following staining with the secondary Abs. Thereafter, the coverslips were washed with blocking solution three times and mounted using Prolong Gold Antifade reagent with DAPI nuclear stain (Invitrogen). Confocal fluorescent images were obtained using a Zeiss LSM510 confocal scan head mounted on a Zeiss Axiovert 200M on an inverted-base microscope with a 63× objective (Carl Zeiss). Images were analyzed by Zeiss software and ImageJ (National Institutes of Health freeware).

Cells were centrifuged and precipitated in glutaraldehyde followed by standard double fixation using glutaraldehyde and osmium teraoxide. The samples were processed using Leica EMTP, cut on Leica Ultra UC6 ultra microtome (Leica Microsystems), and counterstained with lead citrate and uranyl acetate. The images were visualized and captured using a Jeol 1230 transmission electron microscope (Jeol) equipped with AMT digital software.

When comparisons between measurements from two different samples were required, the Mann–Whitney U test was used. Comparisons of paired data from multiple treatments were performed using one-way repeated-measures ANOVA. If a significant difference was found, Tukey test with Geisser-Greenhouse correction was then used to determine which treatments yielded significant differences. Throughout, two-sided p values <0.05 were considered statistically significant. Data sets in which n is >20 were analyzed using two-tailed paired Student t test. Data sets were graphed and analyzed using GraphPad Prism software version 6 (GraphPad). Apoptosis results were normalized against the percentage Annexin-V–positive cells in unstimulated samples. The protective effect of TLR ligands was calculated as percentage apoptosis relative to the Vpr-induced apoptosis.

Blood was obtained from healthy volunteers after approval of the protocol by the ethics review committee of the Ottawa Hospital, Ottawa, Ontario, Canada. A written informed consent was obtained from the study participants.

We have previously shown that pretreatment with CpG ODN/bacterial DNA reversed Vpr-mediated apoptosis in human monocytic cells (Supplemental Fig. 1A) (9). Because Vpr interacts with the proteins on the mitochondrial membrane, causing loss of mitochondrial membrane potential (12, 17, 19), we reasoned that protection induced by bacterial DNA/CpG ODN may be mediated through prevention of Vpr-induced mitochondrial depolarization. The results show that Vpr caused damage to mitochondrial structure in THP-1 cells as evidenced by loss of mitochondrian with dense cristae as observed under the electron microscopy (Fig. 1A). The mitochondrial damage was accompanied by the loss of mitochondrial membrane potential in both THP-1 cells (Fig. 1B, 1C) and monocytes (Fig. 1D). The control Vpr peptide with Arg to Ala mutations at 73, 77, and 80 aa essential for Vpr-mediated apoptosis (38) did not induce mitochondrial depolarization in THP-1 cells (Fig. 1B). Pretreatment with E. coli DNA or CpG ODN prevented Vpr-mediated damage to mitochondrial structure (Fig. 1A) as well as loss of mitochondrial potential in both THP-1 cells (Fig. 1B, 1C) and monocytes (Fig. 1D). Interestingly, pretreatment with non-TLR9–stimulating GpC control also exhibited significant protection against Vpr-mediated mitochondrial depolarization (Fig. 1C). In addition, treatment with TLR9-inhibiting drug chloroquine (39) failed to reverse CpG ODN-mediated protection against Vpr-induced mitochondrial depolarization in both THP-1 cells and monocytes (Fig. 1C, 1D). THP-1 cells treated with CpG ODN alone did not show any change in mitochondrial morphology (Fig. 1A) suggesting that CpG ODN- and bacterial DNA-induced protection occurs in a TLR9-independent manner.

We have shown that CpG ODN and E. coli DNA stimulation enhanced expression of c-IAP-2 (9). Because Vpr caused mitochondrial depolarization and prior treatment with CpG ODN and E. coli DNA induced protection against it, we hypothesized that CpG ODN and E. coli DNA-induced cIAP-2 may prevent Vpr-mediated mitochondrial depolarization. Therefore, c-IAP-2 was knocked down by siRNA (Fig. 2A) or SMAC mimetic (SMC) followed by CpG ODN stimulation and treatment with Vpr. Interestingly, c-IAP-2 siRNA transfected cells stimulated with CpG ODN exhibited significantly higher mitochondrial depolarization in response to Vpr treatment as compared with cells transfected with control nonsilencing siRNA (Fig. 2A). Similarly, SMC treatment degraded c-IAP-2 in CpG ODN-stimulated cells (Fig. 2B) and significantly inhibited CpG ODN- and E. coli DNA-induced protection from Vpr-induced mitochondrial depolarization in both THP-1 cells (Fig. 2C) and monocytes (Fig. 2D). There was no significant difference in mitochondrial depolarization between cells transfected with c-IAP-2 siRNA or control siRNA alone and Vpr-mediated mitochondrial depolarization between cells transfected with c-IAP-2 siRNA or control siRNA (Supplemental Fig. 1B).

Although monocytes and THP-1 cells are susceptible to Vpr-induced apoptosis, these cells acquire antiapoptotic properties following differentiation into MDMs due to upregulation of IAPs (10). Therefore, we used differentiated MDMs and THP-1 Macs as c-IAP-2 overexpression models to determine whether knocking down c-IAP-2 would render these cells sensitive to Vpr-mediated mitochondrial depolarization. Treatment of MDMs with SMC degraded c-IAP-2 strongly and downregulated c-IAP-1 less strongly (Fig. 3A). Treatment of MDMs with Vpr alone did not cause apoptosis (Fig. 3B) or mitochondrial depolarization (Fig. 3C). However, treatment of MDMs with SMC prior to Vpr exposure rendered these cells responsive to Vpr-mediated apoptosis (Fig. 3B) and caused loss of mitochondrial membrane potential (Fig. 3C).

Vpr-induced mitochondrial depolarization may cause release of apoptogenic factors such as cytochrome c and AIF (12, 17, 18). Because c-IAP-2 protected mitochondrial integrity, we reasoned that c-IAP-2 may also prevent the release of cytochrome c and AIF following Vpr treatment. Treatment of MDMs with Vpr alone did not cause release of cytochrome c (Fig. 3D) or AIF (Supplemental Fig. 2A). However, treatment of MDMs with SMC prior to Vpr exposure resulted in cytochrome c and AIF release from the mitochondria as compared with cells treated with Vpr alone (Fig. 3D, Supplemental Fig. 2A). Similarly, THP-1-Macs transfected with c-IAP-2–siRNA (Fig. 3E, right panel) exhibited increased release of AIF as compared with control siRNA-transfected cells in response to Vpr (Fig. 3E, left panel). SMC alone did not cause mitochondrial depolarization in monocytes (Fig. 2D) and MDMs (Fig. 3C).

We previously demonstrated that CpG ODN-induced activation of CaM/CaMK-II leads to c-IAP-2 induction and subsequent protection against Vpr-mediated apoptosis (9). We reasoned that inhibition of c-IAP-2 expression by blocking CaM/CaMK-II activation should prevent bacterial DNA/CpG ODN-mediated protection against Vpr-induced mitochondrial depolarization. Treatment of THP-1 cells with Vpr caused the release of cytochrome c and AIF from the mitochondria and stimulation with CpG ODN or E. coli DNA prevented the release of both cytochrome c and AIF induced by Vpr (Fig. 4A). Interestingly, prior treatment of THP-1 cells with pharmacological inhibitors of influx of calcium ions through receptor gated calcium channels on the cell surface (SKF-96365), inhibitor of CaM (W-7), and CaMK-II activation (KN-93), and calcium chelator (EGTA) reversed the CpG ODN-mediated protection against Vpr-induced cytochrome c and AIF release (Fig. 4A). Treatment of THP-1 cells with the highest concentrations of the inhibitors alone did not cause release of AIF (Supplemental Fig. 2B). Transfecting THP-1 cells with CaMK-II–siRNA prior to CpG ODN stimulation resulted in the loss of c-IAP-2 induction in response to CpG ODN (Fig. 4B). Moreover, CaMK-II siRNA-transfected cells displayed a significant decrease in protection afforded by CpG ODN against Vpr-mediated loss of mitochondrial membrane potential as compared to cells transfected with control nonsilencing siRNA (Fig. 4C). Moreover, CaMK-II–silencing siRNA alone did not affect Vpr-induced mitochondrial depolarization (Supplemental Fig. 3A).

TRAF-1 and TRAF-2 serve as adaptors for physically interacting with and recruiting c-IAP-2 to the membrane-bound signaling complex (40). Therefore, we determined the functional relevance of TRAF-1 and TRAF-2 for c-IAP-2–mediated protection against apoptosis and mitochondrial membrane depolarization caused by Vpr. First, we demonstrated that Vpr induced downregulation of both TRAF-1 and TRAF-2 in THP-1 cells. Stimulation of cells with CpG ODN or E. coli DNA prevented Vpr-induced downregulation of TRAF-1 and TRAF-2 (Fig. 5A). However, CpG ODN or E. coli DNA treatment failed to rescue TRAF-1/2 from Vpr-mediated downregulation if c-IAP-2 was degraded using SMC prior to stimulation with bacterial DNA (Fig. 5A). Treatment of cells with SMC alone did not alter TRAF-1 and TRAF-2 expression (data not shown). Furthermore, siRNA-mediated knockdown of TRAF-1 and TRAF-2 (Fig. 5B) induced significant loss of CpG ODN-mediated protection against Vpr-triggered apoptosis (Fig. 5C) and mitochondrial depolarization (Fig. 5D). These results suggest that CpG ODN/bacterial DNA-induced c-IAP-2 prevents Vpr-mediated apoptosis and mitochondrial depolarization by inhibiting Vpr-induced TRAF-1/2 degradation.

To understand the mechanism by which the c-IAP-2–TRAF-1–TRAF-2 complex regulates Vpr-induced mitochondrial depolarization, we first determined the mechanism of Vpr-mediated apoptosis in monocytic cells. Vpr has been shown to induce apoptosis through the caspase-8–mediated apoptotic pathway in human neuronal cells, renal tubular epithelial cells, and Jurkat cells (41–43). To determine the role of caspase-8 activation in Vpr-mediated mitochondrial depolarization, monocytes were treated with specific inhibitor of caspase-8 activation Z-IETD-FMK (44) followed by Vpr treatment. Z-IETD-FMK significantly reduced Vpr-mediated apoptosis (Fig. 6A) and mitochondrial depolarization (Fig. 6B). Cycloheximide and TNF-α have been widely shown to be inducers of caspase-8–mediated apoptosis (40, 45). Hence, the biological activity of Z-IETD-FMK was confirmed by its ability to inhibit cycloheximide and TNF-α–mediated apoptosis in THP-1 cells (Supplemental Fig. 3B). The role of caspase-8 in Vpr-induced apoptosis and mitochondrial depolarization was confirmed by knocking down caspase-8 expression using siRNA. THP-1 cells transfected with caspase-8 siRNA exhibited significantly lower apoptosis (Fig. 6C) and mitochondrial depolarization (Fig. 6D) in response to Vpr treatment. As expected, caspase-8 siRNA inhibited caspase-8 expression (Fig. 6E).

To determine whether CpG ODN exerts its antiapoptotic effects via inhibition of caspase-8 activation caused by Vpr, we show that Vpr induced caspase-8 cleavage, which could be reversed by prior stimulation with CpG ODN in THP-1 cells (Fig. 7A, 7B). Moreover, CpG ODN-induced inhibition of caspase-8 cleavage was reversed if c-IAP-2 was degraded by SMC prior to stimulation with CpG ODN and treating cells with Vpr as shown by immune blotting with Abs against full-length caspase-8 (Fig. 7A) and visualizing cleaved caspase-8 with confocal microscopy using Abs against activated caspase-8 in both THP-1 cells and monocytes (Fig. 7B). The protection against Vpr-mediated caspase-8 activation by CpG ODN-induced c-IAP-2 was confirmed by employing FITC-conjugated caspase-8 inhibitor (FITC-IETD ZFMK) that specifically binds to cleaved caspase-8 and allows for analysis of caspase-8 activation by flow cytometry (Fig. 7C).

Because CpG ODN/bacterial DNA-induced c-IAP-2 prevents Vpr-mediated downregulation of TRAF-1 and TRAF-2 as well as caspase-8 activation, we determined if c-IAP-2–mediated protection of caspase-8 cleavage is regulated by TRAF-1/2. Our results show that Vpr treatment caused caspase-8 cleavage. Prior stimulation of cells with CpG ODN prevented Vpr-induced caspase-8 cleavage. siRNA-mediated knockdown of TRAF-1 and TRAF-2 significantly abrogated the protection afforded by CpG ODN against Vpr-mediated caspase-8 activation in THP-1 cells (Fig. 8).

Activation of Bid is known to mediate apoptosis downstream of caspase-8 (46). Upon activation by cleavage, truncated Bid may directly translocate to the mitochondrial surface and induce membrane permeabilization (46). The involvement of Bid was demonstrated by knocking down Bid in THP-1 cells (Fig. 9A). Cells transfected with Bid siRNA were found to be significantly less sensitive to apoptosis (Fig. 9B) and mitochondrial depolarization (Fig. 9C) in response to Vpr treatment as compared with control siRNA-transfected cells. As a control, Bid siRNA was shown to prevent TNF-α and cycloheximide-induced apoptosis (Supplemental Fig. 3C).

Bid may also activate another proapoptotic Bcl2 family member, Bax (46). Activated Bax has been shown to mediate apoptosis by embedding and oligomerizing within the mitochondrial membrane, causing loss of mitochondrial membrane potential and apoptosis (47). The knockdown of Bax was recently shown to abrogate Vpr-mediated cell death in HeLa cells and human epithelial cells (19, 43). Therefore, we determined whether Vpr-mediated caspase-8–dependent apoptosis involved Bax activation in human monocytic cells. THP-1 cells transfected with Bax siRNA (Fig. 9D) exhibited significantly lower apoptosis (Fig. 9E) and mitochondrial depolarization (Fig. 9F) in response to Vpr treatment as compared with cells transfected with nonsilencing siRNA. Bax activation entails undergoing a conformational change revealing amino terminal epitope recognized by the mAb 6A7 (48). Vpr treatment of THP-1 cells caused Bax activation as determined by staining with 6A7 Ab. Prior treatment with CpG ODN, E. coli DNA (data not shown), and GpC control ODN (data not shown) inhibited Vpr-mediated Bax activation in both THP-1 cells (Fig. 9G) and monocytes (Fig. 9H). These results suggest that CpG ODN stimulation prevents Vpr-induced Bid and Bax activation and subsequent mitochondrial depolarization and apoptosis, as shown in the model in Fig. 10.

We provide evidence that HIV-Vpr causes mitochondrial depolarization by downregulating upstream TRAF-1/TRAF-2 and activating caspase-8, Bid, and Bax, eventually leading to mitochondrial membrane permeabilization and release of cytochrome c and AIF in human monocytic cells. However, stimulation of cells with CpG ODN or bacterial DNA prior to Vpr treatment protected against mitochondrial depolarization via induction of antiapoptotic c-IAP-2 gene, which inhibited downregulation of TRAF-1 and TRAF-2, activation of caspase-8, Bid, and Bax, and subsequent release of cytochrome c and AIF from the mitochondria. These results suggest a novel and key role played by c-IAP-2 in protection against HIV-Vpr–mediated mitochondrial depolarization through the intrinsic pathway in human monocytic cells. With the exception of survivin using a transgenic mouse model (49), no other member of IAP family has been shown to exert antiapoptotic effects by directly protecting mitochondria from depolarization and release of apoptotic factors in either mouse or human systems.

Mitochondria perform several key cellular processes such as oxidative phosphorylation and generation of adenosine triphosphate, homeostasis of cellular reactive oxygen species, and regulation of the intrinsic apoptotic pathway through cytochrome c release (50). Recently, HIV infection in the absence of antiretroviral therapy was shown to contribute to mitochondrial dysfunction (51). HIV infection decreased mitochondrial DNA in adipose tissue possibly through immune activation (52, 53). Concomitantly, plasma mitochondrial DNA levels were significantly elevated in acute HIV infection and late HIV presenters (52, 54). Moreover, reactive oxygen species production, mitochondrial membrane potential, and other morphological parameters are significantly altered in HIV infection (51). HIV-Vpr physically interacts with mitochondrial ANT, which leads to the release of cytochrome c, AIF, and SMAC (12, 17). However, the mechanism involved in Vpr-mediated mitochondrial membrane permeabilization and release of apoptogenic factors are not well understood. Our results suggest that Vpr-induced mitochondrial depolarization is due to the events upstream of mitochondria through TRAF-1/2 degradation, caspase-8 cleavage, and Bid and Bax activation. Moreover, Vpr-mediated downregulation of TRAF-1/2 is essential for caspase-8 activation and mitochondrial depolarization. Cells treated with caspase-8 inhibitors or transfected with caspase-8 siRNA exhibited significantly reduced mitochondrial depolarization. Vpr takes advantage of the cell’s proteasomal degradation mechanisms by enlisting E3 ubiquitin ligase cullin-4A via DCAF-1 to target various cellular substrates for degradation to induce cell-cycle arrest or apoptosis (55, 56). Additionally, TRAF-2 is an E3 ligase and contains a RING domain facilitating protein–protein interaction (57–60). Hence, given the rapid downregulation of TRAF-1/2 by Vpr, it is possible that Vpr exploits the host ubiquitin proteasome system for caspase-8 cleavage and TRAF-1/2 degradation.

Vpr-induced release of cytochrome c into the cytosol facilitates aggregation of the apoptosome, consisting of cytochrome c, procaspase-9, and Apaf-1 and subsequent sequential activation of caspase-9 and caspase-3, eventually leading to apoptosis (61, 62). However, the role of caspase-8 in Vpr-induced apoptosis is controversial (13, 14, 41–43, 63, 64), and its role in Vpr-induced mitochondrial depolarization is not known. Our results show that Vpr-induced mitochondrial depolarization is mediated by caspase-8 activation in macrophages.

Because TRAF-1/2 bind c-IAP-2 (65, 66), and Vpr-mediated caspase-8 activation requires TRAF-1/2 downregulation, we hypothesized that c-IAP-2 inhibited Vpr-induced caspase-8 activation by preventing TRAF-1/2 degradation (Fig. 10). Indeed, knockdown of TRAF-1/2 by their specific siRNAs rendered cells sensitive to Vpr-mediated caspase-8 activation and mitochondrial depolarization. Interestingly, individual knockdowns of TRAF-1 and TRAF-2 did not completely abrogate c-IAP-2–mediated protection, suggesting redundancy in the functions of TRAF-1/2. c-IAP-1/2 have been shown to maintain caspase-8 in an inactive state by ubiquitinating RIP-1 in the ripoptosome (27). However, contrary to previously published mechanisms of c-IAP-2–mediated protection, RIP-1 inhibitor necrostatin (67) did not inhibit the protective effect of c-IAP-2 against Vpr-mediated apoptosis (data not shown). It is possible that TRAF-1/2 exist as a complex with procaspase-8 that serves to maintain caspase-8 in an inactivated form. Our results suggest that when cells are treated with Vpr, TRAF-1/2 are targeted for ubiquitination and proteasomal degradation, leaving caspase-8 free for cleavage/activation. Because c-IAP-2 binds TRAF-1/2 (65, 66), binding of c-IAP-2 to TRAF-1/2 may make TRAF-1/2 refractory to Vpr-mediated degradation, caspase-8 activation, and mitochondrial depolarization. These observations also suggest that signaling pathways involved in c-IAP-2 synthesis will mediate such processes. We have previously shown that CpG ODN/bacterial DNA-induced c-IAP-2 expression is regulated by calcium signaling via CaMK-II (9). In this study, we provide evidence that c-IAP-2–mediated protection against mitochondrial depolarization and release of apoptogenic factors is regulated via activation of CaM/CaMK-II–mediated calcium signaling.

Differentiated MDMs and THP-1 macrophages exhibited resistance to Vpr-induced apoptosis that was attributed to the overexpression of c-IAP-1/2 (10). In this study, differentiated macrophages were refractory to Vpr-induced mitochondrial depolarization. Inhibition of c-IAP-2 expression by SMAC mimetics or specific siRNAs rendered differentiated macrophages sensitive to Vpr-induced mitochondrial depolarization and release of cytochrome c and AIF. These results confirmed that c-IAP-2 protects against mitochondrial depolarization in differentiated macrophages similar to the CpG ODN/bacterial DNA–stimulated monocytes.

Antiapoptotic Bcl-2 family of proteins such as Bcl-2, Bcl-xL, and Mcl-1 exert their protective effects by inhibiting mitochondrial permeabilization via neutralization of proapoptotic Bid and Bax (10, 68). Considering that protection of mitochondrial viability is critical for inhibiting Vpr-mediated apoptosis and taking into account the significance of Bid and Bax in causing Vpr-induced mitochondrial depolarization, it is possible that Bcl-2, Bcl-xL, or Mcl-1 may also play a role in CpG ODN-endowed protection against mitochondrial depolarization caused by Vpr. Moreover, CpG via TLR9 stimulation enhanced human neutrophil survival by directly upregulating Mcl-1 expression (69). However, neither the knockdown nor the inhibition of Bcl-xL or Mcl-1 expression significantly affected the CpG ODN-mediated protection against Vpr-induced mitochondrial depolarization or apoptosis (Supplemental Fig. 4A–C). In addition, inhibition of biological activity of Bcl2 family members by its inhibitor HA14-1 did not affect protection against Vpr-induced apoptosis (Supplemental Fig. 4D). Overall, our results suggest that Bcl2 family members do not affect Vpr-induced apoptosis or mitochondrial depolarization. However, further studies are needed to address the canonical and noncanonical survival proteins required to precisely determine how monocytes control their sensitivity to HIV peptides.

We show that Vpr-induced mitochondrial damage, both structural and functional, can be restored by pretreating cells with TLR9 ligand, CpG ODN, or bacterial DNA. The role of TLR-9 in responsiveness to CpG in human monocytes has recently been questioned, as these cells express very little TLR9 (70, 71). In fact, induction of cytokines by human monocytes in response to CpG was attributed to contaminating plamacytoid dendritic cells (71). Moreover, experiments performed with non-TLR9–stimulating ODNs have challenged the overall TLR9 dependence for CpG-mediated signaling in these cells (72–74). We have shown that although human monocytes express TLR9, immunomodulatory effects of CpG ODN were found to be TLR9 independent (9). The nature of the DNA cytosolic recognition receptor in immune cells is not clear. Several proteins have been suggested to act as sensors of DNA including DNA-dependent activator of IFN regulatory factors, also referred to as DLM-1/ZBP1, and stimulator of IFN genes (75–78). However, none has met with a universal acceptance linking all of the recent observations.

Chronically HIV-infected individuals contain higher levels of endotoxins and bacterial products, including DNA in their plasma following translocation from the gut resulting in chronic immune activation (79). The disruption of gut barrier due to HIV infection leads to translocation of luminal microbes and microbial products like LPS and bacterial DNA in the circulation. Moreover, presence of these microbial products in circulation has been correlated with systemic immune activation and disease progression (79, 80). Our current data shown in this manuscript and previous results (9) indicate that prior exposure of primary human monocytes and THP-1 cells to microbial products renders these cells resistant to HIV-Vpr–induced cell death. Enhanced resistance to apoptosis has been associated with maintenance of HIV and its replication in monocytic cells and eventual development of reservoir formation (81)

In summary, this is the first report, to our knowledge, describing a role for bacterial DNA and CpG ODN-induced c-IAP-2 in protecting against mitochondrial membrane depolarization by inhibition of HIV-Vpr–mediated TRAF-1 and TRAF-2 degradation and activation of caspase-8, Bid, and Bax. We also show the involvement of TRAF-1/2 in Vpr-mediated cell death through the activation of caspase-8, Bid, and Bax. Considering the significance of c-IAP-2, calcium signaling, and TRAF-1/2 in establishing resistance to cell death, drugs aimed at destroying this resistance by targeting c-IAP-2 such as Smac mimetics, CaM/CaMK-II, or TRAF-1/2, may cause selective death of HIV-infected macrophages and eventually prevent HIV reservoir formation in these cells.

We thank Dr. M. Kozlowski for critically reading the manuscript and Dr. Nick Barrowman for help in statistical analysis.

The authors have no financial conflicts of interest.