Published online before print December 5, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Center for Cancer Research Nanobiology Program, National Cancer Institute, National Institutes of Health, Frederick, Maryland
1Correspondence: Center for Cancer Research, NCI-Frederick, P.O. Box B, Building 469, Room 152, Miller Drive, Frederick, MD 21702-1201. E-mail: blumen{at}helix.nih.gov
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: hemifusion • Env glycoprotein • CXCR4
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Various viral proteins including Env, Vpr, and Tat [2 3 4 5 ] have been implicated in HIV-mediated apoptosis. The Env glycoprotein is composed of a gp120 surface unit, which interacts with CD4 (receptor) and CXC chemokine receptor 4 (CXCR4)/CC chemokine receptor 5 (CCR5; coreceptor) on target cells and a gp41 transmembrane protein, which is a class I fusion protein mediating fusion of membranes via a pH-independent mechanism [6 ]. As Env is expressed on the surface of infected cells and can interact with CD4 on uninfected cells, the preferential depletion of CD4+ T cells in HIV infection suggests that Env may be the most likely candidate for induction of apoptosis in bystander cells. In fact, HIV Env, expressed on the cell surface, has been shown to interact with CD4 and CXCR4 to trigger apoptosis in uninfected cells [5 , 7 ]. The function of Env glycoprotein is not restricted to receptor/coreceptor binding but, in fact, culminates in fusion of effector and target membranes. In this context, pathogenecity of HIV has been associated with the fusogenic potential of the Env glycoprotein [8 9 10 ]. In a bystander scenario, whether Env signals apoptosis via binding to CD4 and CXCR4 or via the fusion process remains controversial. In this regard, a recent report shows that Env-mediated apoptosis in bystander cells correlates with a gp41-dependent, hemifusion-like event and can be inhibited by gp41-specific fusion inhibitory peptides [11 ]. Furthermore, the requirement of membrane fusion for induction of apoptosis has also been reported for CCR5 tropic viruses [12 , 13 ].
Although the interaction of HIV gp120 with receptor and coreceptor is required for the induction of apoptosis, signaling via either of these receptors seems dispensable for this process [14 , 15 ]. Hence, the signaling mechanism via which HIV Env mediates apoptosis in bystander cells is not clear, although several reports indicate that caspase activation [16 , 17 ], mitochondrial dysfunction [18 , 19 ], and reactive oxygen species (ROS) may be involved [20 ]. Nevertheless, the sequence of these events in HIV Env-mediated apoptosis remains unclear. In this study, we have tried to determine whether Env-mediated apoptosis in bystander cells is a result of a gp41-dependent hemifusion-like event and the signaling pathway activated.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Isolation of CD4+ lymphocytes from peripheral blood mononuclear cells (PBMC)
Whole blood was collected from healthy donors via the anonymous blood donor program at the National Cancer Institute (NCI)-Frederick (MD). PBMC were isolated using Ficoll gradient (Amersham, Little Chalfont, UK). The purified PBMC were washed with phosphate-buffered saline and passed through a CD4 T cell enrichment column (Cedarlane Labs, Ontario, Canada). The purified CD4+ T cells were activated with 2 µg/ml phytohemagglutinin (PHA) and interleukin (IL)-2, 10 U/ml, and cultured for 3 days prior to use. The purity of CD4+ T cells (>95%) was determined via staining with phycoerythrin-conjugated anti-CD4 antibody (BD PharMingen, San Diego, CA) followed by flow cytometry prior to use.
Reagents
CXCR4 antagonist AMD3100; peptides C34, T20, and N36; and protease inhibitors nelfinavir, saquinavir, and indinavir were obtained from the AIDS Reference and Reagent program. Caspase inhibitors E64D, pepstatin, SB203580, and cyclic pifithrin were from Calbiochem (San Diego, CA). Anti-Fas antibody clone CH-11 was from Upstate Inc. (Lake Placid, NY). All fluorescent dyes were from Molecular Probes (Eugene, OR). N36mut(e,g) peptide was a kind gift from Dr. Marius Clore (NIH, Bethesda, MD), 2F5 antibody was a kind gift from Dr. Hermann Katinger (University of Natural Resources, Vienna, Austria), and NC-1 antibody was a gift from Dr. Shibo Jiang (New York Blood Center, New York, NY).
Induction of apoptosis
CHO WT, GPI-anchored Env (CHO PI), or vector-transfected (CHO EE) were seeded in 24-well plates at 105 cells per well and allowed to adhere overnight. Subsequent day media were removed, and 0.5–1.0 x 106 SupT1 cells were added per well for a final effector-to-target ratio of 1:5–1:10. Different inhibitors were preincubated with SupT1 cells or added at the time of coculture. The cells were cocultured for 24 h, following which the nonadherent target cells were collected and used for various analyses.
Detection of apoptosis
Apoptosis was determined in target cells 24 h post-coculture, as phosphatidyl serine exposure via Annexin V staining (Molecular Probes) or as DNA fragmentation using the deoxyuridine triphosphate nick-end labeling (TUNEL) assay (Roche, Nutley, NJ), as per the manufacturers’ instructions, followed by flow cytometry (FACSCalibur, BD Biosciences, San Jose, CA). At least 10,000 events were collected and analyzed using Cellquest software.
Detection of mitochondrial depolarization and ROS
Mitochondrial depolarization was determined by staining cells with 10 nM DiOC6 dye for 15–20 min at 37°C, followed by flow cytometry. ROS production in cells was determined via dihydroethidium (DHE) staining, 2.5 µM at 37°C for 30 min, followed by flow cytometry.
Caspase activation
Caspase-3 activation was determined by staining cells with fluorescein isothiocyante (FITC)-labeled anti-active caspase-3 monoclonal antibody (mAb; BD PharMingen), as per the manufacturer’s instructions, followed by flow cytometry. Conversely, caspase-3/7 or caspase-8 activity was determined by substrate-based luminescent Caspase-Glo assay (Promega, Madison, WI), as per the manufacturer’s instructions.
Dye transfer assay
CHO cells were labeled with nondiffusible cytoplasmic dye 4-chloromethyl benzoyl amino tetramethyl rhodamine (CMTMR; 10 µM) or diffusible cytoplasmic dye Calcein Red (10 µM) or lipophilic membrane dye 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine (DiI; Vybrant cell labeling kit, Molecular Probes). The cells were then cocultured with unlabeled SupT1 cells for 24 h. Subsequently, the nonadherent target cells were collected and stained with Annexin V for detection of apoptosis and analyzed by flow cytometry. A transfer of dye (cytoplasmic or membrane) was acquired in the red channel, and the green channel was used for apoptosis detection via Annexin V-FITC. This assay was used to determine whether the apoptotic target cells seen in our system had taken up cytoplasmic or membrane dyes from the effector cells.
Fusion from without
High-titer HIV-1 MN virions were a kind gift from Jeff Lifson and Julian Bess [National Institutes of Health (NIH) AIDS vaccine program, NCI-Frederick]. SupT1 cells were incubated with 200 ng/ml p24 equivalent of virus in the presence of 1 µM zidovudine (AZT) to prevent reverse transcription and any productive infection. Inhibitors C34 or AMD3100 were added prior to addition of virus. Fusion from without was observed, and photomicrographs were collected at 2 h. Caspase-3 activity was determined at 24 h post-virus addition.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (22K): [in a new window] |
Figure 1. Fusion-competent Env glycoprotein mediates apoptosis in bystander cells. SupT1 cells were cocultured with CHO WT, CHO PI, or CHO EE cells at a ratio of 1:10. Twenty-four hours post-coculture, the nonadherent target cells were collected and analyzed for apoptosis via Annexin V (a) or TUNEL (b) staining.
|
 View larger version (12K): [in a new window] |
Figure 2. Env-mediated apoptosis can be inhibited by gp41-specific fusion inhibitors. (a) CHO WT cells were cocultured with SupT1 cells in the presence or absence of CXCR4 antagonist AMD3100 (2 µM) or gp41 fusion inhibitor C34 (2 µM), and apoptosis was determined 24 h later via TUNEL staining. Gp41-specific fusion-inhibitory peptides C34, T20, N36, and N36mut(e,g) (4 µM; b) or HIV gp41-specific neutralizing 2F5 or non-neutralizing NC-1 antibodies (20 µg/ml; c) were added to inhibit Env-mediated apoptosis, determined by Annexin V staining. Data are mean ± SD triplicate observations. All experiments were repeated twice with similar results. Statistical significance was assessed by Student’s t-test (*, P<0.001).
|
HIV Env-mediated apoptosis correlates with a hemifusion-like event
Hemifusion is a phenomenon mediated via a variety of viral fusion proteins [32
33
34
], whereby the outer leaflet of effector and target membranes fuse without fusion of the inner leaflets or the formation of the fusion pore. Under hemifusion conditions, mixing of lipid components between the membranes occurs, which can be demonstrated via transfer of fluorescently labeled lipid dyes, and full fusion results in syncytia formation, which can be studied by intracytoplasmic labeling of effector and target cells. To determine whether the apoptosis seen under our coculture conditions was a result of full fusion of effector and target cells (as suggested by Ferri et al. [35
]) or mixing of lipid contents between cells (as suggested by Blanco et al. [11
]), a dye transfer assay was used, whereby CHO cells labeled with nondiffusable CMTMR dye, diffusible calcein red dye, or the lipophilic DiI dye were cocultured with unlabeled SupT1 cells. Mixing of cellular or lipid contents could be detected via transfer of the dye to the unlabeled target cells. Further, to detect apoptosis, Annexin V was used as the marker. As seen in Figure 3
, apoptosis occurred primarily in cells that had taken up the lipid dye DiI from the effector cells, and there was little transfer of the diffusible (calcein) or nondiffusible (CMTMR) cytoplasmic dyes. Apoptosis and DiI transfer was blocked by C34, indicating that it was an Env-mediated phenomenon and not a passive transfer of dye. This suggests that HIV Env-mediated apoptosis correlates with exchange of lipid components between effector and target cells.
 View larger version (24K): [in a new window] |
Figure 3. HIV gp41-mediated apoptosis occurs predominantly in cells after a hemifusion-like event. CHO WT cells were labeled with DiI, calcein red, or CMTMR and cocultured with unlabeled SupT1 cells. Twenty-four hours post-coculture, the nonadherent cells were stained with Annexin V-FITC and analyzed for dye transfer (FL-2) or apoptosis (FL-1) via flow cytometry. C34 (2 µM) was added to inhibit fusion (right panels).
|
 View larger version (26K): [in a new window] |
Figure 4. HIV gp41-mediated apoptosis involves mitochondrial depolarization and ROS production. SupT1 cells cocultured with CHO WT cells in the presence or absence of C34 or AMD were analyzed for mitochondrial depolarization via DiOC6 staining (a) or ROS production via DHE staining (b) followed by flow cytometry.
|
Caspase-3 but not caspase-8 is involved in HIV Env-mediated apoptosis
Caspases are critical mediators of apoptosis and have been shown to be involved in HIV Env-mediated apoptosis [14
, 17
, 37
]. To determine whether caspase-3 was activated under our apoptosis-inducing conditions, a direct, intracytoplasmic staining of cells with a mAb against active caspase-3 was used. As seen in Figure 5a
, Env-mediated apoptosis involved activation of caspase-3, which was again inhibited by C34, suggesting the requirement of gp41. Caspase-3 activation has often been shown to be downstream of caspase-8 activation in receptor (Fas, tumor necrosis factor receptor)-mediated apoptosis. To determine whether caspase-3 activation was downstream of caspase-8, a substrate-based assay was used to determine the activity of both of these enzymes under identical culture conditions. As seen in Figure 5b
, there was no significant activation of caspase-8 under conditions where caspase-3 activity was fourfold higher in Env-exposed cells. As a control, we induced apoptosis in CEM cells or SupT cells using anti-Fas antibody CH-11. As shown in Figure 5c
, although anti-Fas induced caspase-3 and caspase-8 activation in CEM cells, it did not induce apoptosis in SupT cells, consisitent with previous reports [38
]. This suggests that caspase-3 activation and apoptosis, under our culture conditions, were most likely independent of caspase-8 activity.
 View larger version (19K): [in a new window] |
Figure 5. HIV gp41-mediated apoptosis involves caspase-3 but not caspase-8 activation. (a) SupT1 cells were cocultured with CHO WT in the presence or absence of C34 or AMD for 24 h, following which caspase-3 activation was determined via anti-caspase-3 antibody (Ab) staining. (b) CHO WT SupT1 cocultures were assayed for caspase-3 and -8 activity using a substrate-based Caspase-Glo assay. AMD and C34 were used as inhibitors of Env-mediated fusion. (c) CEM or SupT1 cells were incubated with 2 µg/ml anti-Fas antibody CH-11 for 24 h, following which caspase-3 and -8 activity was determined. Data are mean ± SD of triplicate observations. All experiments were repeated twice with similar results.
|
 View larger version (21K): [in a new window] |
Figure 6. Env-mediated apoptosis is dependent on caspase-mediated activation of mitochondrial depolarization. (a) SupT1 cells were cultured with CHO WT cells in the presence or absence of C34 (2 µM) or Z-VAD-FMK (40 µM). Mitochondrial depolarization was determined via DiOC6 staining. (b) Pan caspase inhibitor Z-VAD-FMK (40 µM) or caspase-3 inhibitor Ac-Asp-Glu-Val-Asp (DEVD)-CHO (40 µM) was used to inhibit apoptosis. PS exposure via Annexin V staining, mitochondrial depolarization by DiOC6 staining, and ROS by DHE (HE) staining was determined 24 h post-coculture. (c) Caspase-3 (DEVD), caspase-9 [Ac-Leu-Glu-His-Asp (LEHD)-CHO], caspase-8 [Ac-Ile-Glu-Thr-Asp (IETD)-CHO], or pan caspase inhibitor (Z-VAD) was added to inhibit Env-mediated apoptosis. Mitochondrial depolarization (mito depol.) was determined via DiOC6 staining and normalized to percent inhibition of C34 control. Data are mean ± SD of triplicate observations. All experiments were repeated twice with similar results.
|
 View larger version (10K): [in a new window] |
Figure 7. HIV gp41-mediated apoptosis is independent of p38 MAPK and p53. Coculture experiments were conducted in the presence or absence of p53 inhibitor cyclic pifithrin (10 µM) or p38 MAPK inhibitor SB203580 (1 µM). Apoptosis was determined 24 h later via Annexin V staining. Data are average ± SD of triplicate observations.
|
 View larger version (18K): [in a new window] |
Figure 8. HIV-specific protease inhibitor nelfinavir inhibits apoptosis via inhibition of mitochondrial depolarizatiuon. (a) SupT1 cells were cocultured with CHO WT cells in the presence of varying concentrations of nelfinavir. Apoptosis was determined 24 h later via Annexin V staining. (b) SupT1 cells were pretreated for 30 min with HIV protease inhibitors nelfinavir (Nel; 10 µM), inidnavir (Ind; 10 µM), saquinavir (Saq; 10 µM), or the cysteine protease inhibtor E64D (50 µM) or the aspartate protease inhibitor pepstatin (50 µM) prior to coculture with CHO WT cells. Apoptosis was determined via Annexin V staining. (c) Nelfinavir (10 µM)-mediated inhibition of apoptosis was determined by Annexin V, DiOC6, or DHE (ROS) staining. Data are average ± SD.
|
 View larger version (14K): [in a new window] |
Figure 9. CD4+ PBMC are susceptible to gp41-dependent apoptosis. Peripheral blood-derived CD4+ T cells were activated with PHA, 2 µg/ml, + IL-2, 10 U/ml, for 4 days and subsequently cocultured with CHO WT cells. Apoptosis was determined 24 h post-coculture via TUNEL (a), Annexin V (b), or caspase-3 or -8 activities via a substrate-based assay (c).
|
 View larger version (40K): [in a new window] |
Figure 10. Virion-induced fusion from without activates caspase-3, which is inhibited by C34. SupT1 cells were incubated with 200 ng/ml p24 equivalents of HIV virions in the presence of 1 µM AZT to prevent productive infection and in the presence or absence of C34 or AMD. (a) Photomicrographs collected at 2 h post-infection show fusion from without in the absence of inhibitors AMD or C34. (b) Caspase-3 and -8 activities were determined 24 h later using Caspase-Glo assay. Data are average ± SD.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our results confirm that under in vitro culture conditions, whereby HIV receptor/coreceptor- expressing target cells are exposed to Env-expressing effector cells, apoptosis of target cells is initiated. This process can be inhibited by the gp41-specific fusion-inhibitory peptides, indicating that it is gp41-dependent. We also show here that GPI-anchored Env lacking fusion capabilities [24 ] but having an intact gp120 [21 ] does not induce apoptosis. This argues against the role of 120 binding to CD4 and CXCR4 as the apoptosis-inducing signal, which is consistent with findings by others [14 , 15 ]. The inhibition of Env-mediated apoptosis by the anti-gp41-neutralizing antibody 2F5 further strengthens the hypothesis that this phenomenon is dependent on gp41. Furthermore, we show here for the first time that virion-induced fusion from without induces apoptosis in CD4+ T cells. Using a dye redistribution assay, we confirm findings by Blanco et al. [11 ], who have shown that apoptosis via membrane-expressed HIV Env occurs predominantly in cells after a hemifusion-like event.
We have extended these findings by looking at various intracellular events leading to HIV Env-induced apoptosis. We observe that the apoptosis in our model was mediated via a mitochondrial pathway, as has been seen by others in Env cocultures [49 ] and virus infection [18 ]. We also find that the cells undergoing apoptosis show signs of high ROS, a marker of oxidative stress, and a phenomenon reported in HIV-infected cultures [16 ] or in lymphocytes from infected patients ex vivo [20 ]. Also, an analysis of the caspase activation suggests a role of caspase-3 but not caspase-8 in Env-mediated apoptosis. Interestingly that inhibition of mitochondrial depolarization by the pan-caspase inhibitor Z-VAD-FMK suggests that caspase activation was upstream of mitochondrial depolarization. The extrinsic pathway for apoptosis initiated via Fas ligation involves caspase-8 activation upstream of mitochondrial depolarization. This pathway is unlikely a result of the absence of caspase-8 activation observed in PBMC and lack of Fas-induced apoptosis in SupT1 cells [38 ]. This is consistent with previous reports showing that Env-mediated apoptosis is Fas-independent [17 , 49 ]. Interestingly, caspase-3 inhibitor Ac-DEVD-CHO also inhibited mitochondrial depolarization and apoptosis similar to pan caspase inhibitor Z-VAD, and other inhibitors were not as effective. However, partial inhibition was seen with the caspase-9 inhibitor, suggesting that HIV Env-mediated apoptosis may involve early caspase-3 activation followed by amplification of signal via a caspase-3-mediated mitochondrial amplification loop initiated by caspase-9. There is mounting evidence in favor of a caspase-3-mediated mitochondrial amplification loop in a variety of stimuli including genotoxic drugs [40 ] and granzyme B [50 ]-mediated apoptosis. In fact, caspase-3 has been shown to cleave the proapoptotic protein bid directly, which in turn mediates mitochondrial depolarization [39 ]. Furthermore, a recent report shows caspase-3 to be a component of lipid rafts and is activated early after Fas ligation in type II cells, which poorly recruit caspase-8 to the Fas-associated death domain [51 ]. This suggests that caspase-3 may act as an initiator and effector caspase and may be a key regulator of HIV Env-mediated apoptosis.
We also show that the apoptosis seen in our system was distinct from syncytial apoptosis, as p53 or p38 MAPK inhibitors, shown to inhibit syncytial apoptosis [42 ], failed to inhibit apoptosis in our system. Furthermore, we observe that mitochondrial depolarization was caspase-dependent, contrary to what has been reported in syncytial apoptosis [37 ]. This suggests that although gp41-mediated fusion and hemifusion can result in apoptosis, the mechanism of apoptosis under these conditions is most likely distinct. We believe that successful fusion events resulting in syncytia formation might result in less membrane perturbation or a suboptimal signal, and thus, the apoptosis seen in syncytia is delayed and a result of intrinsic signaling via p53. Consistent with this idea, in a model of contagious apoptosis, suboptimal apoptotic signal in effector/target cells can be transmitted to full apoptosis in syncytia post-fusion via a p53-independent mechanism and might be similar to apoptosis seen in our system [43 ].
We demonstrate here for the first time the ability of HIV-specific protease inhibitor nelfinavir to inhibit gp41-mediated apoptosis. Although other HIV protease inhibitors failed to inhibit apoptosis, it is conceivable that nelfinavir is distinct from other inhibitors, as it has been shown recently to inhibit mitochondrial depolarization by acting on the adenosine nucleotide transporter subunit of the mitochondrial permeability transition pore complex [46 ]. This holds clinical relevance, as the use of nelfinavir has been shown to reduce apoptosis in infected individuals beyond virus suppression [52 ]. We believe that our findings have twofold implications. First, that the inhibition of apoptosis via nelfinavir seen in our system is also seen in vivo suggests that the mechanism of HIV-induced apoptosis in vivo may be via gp41-mediated membrane perturbation. Second, it also suggests that nelfinavir may be a drug of choice for inhibiting HIV replication and Env-mediated apoptosis.
Although we find that pathogenesis of HIV is associated with the fusogenic property of the Env glycoprotein, how gp41-mediated hemifusion events lead to apoptotic signaling is a matter of further investigation. Based on our preliminary findings, we hypothesize that gp41-mediated membrane perturbation may activate membrane-associated caspase-3, followed by activation of proapoptotic members of the bcl-2 family such as bid and/or bax. Although Fas or caspase-8 is not involved, gp41-mediated apoptosis has similarities to extrinsic pathway of apoptosis initiated via type II Fas signaling as a result of several facts. First, gp41-mediated hemifusion most likely generates signals at the membrane, making it similar to extrinsic pathway. Second, caspase-3 activation was upstream of mitochondrial depolarization, as has been shown recently in type II Fas signaling [51 ]. Finally, nelfinavir, which inhibits type II but not type I Fas-mediated apoptosis [45 ], also inhibits gp41-mediated apoptosis.
Overall, we demonstrate here an involvement of gp41 in induction of CD4+ T cell apoptosis by a process that is dependent on caspase-3 activation upstream of nelfinavir-inhibitable mitochondrial depolarization. The requirement of an amplification pathway for apoptosis via HIV gp41 suggests that the signal generated via the interaction of gp41 with cellular membranes may be weak and hence may induce apoptosis in vitro but may only prime cells for apoptosis in vivo. Further analysis of this pathway would be of interest to correlate in vitro with in vivo findings. Although we believe that the loss of T cells in HIV infection is multifaceted involving several phenomenon, we do wish to emphasize that the potential of Env glycoprotein to induce apoptosis is largely dependent on gp41 function and not on gp120 binding to receptor/coreceptor on bystander cells.
Received August 2, 2005; revised October 7, 2005; accepted October 17, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Garg, A. Joshi, E. O. Freed, and R. Blumenthal Site-specific Mutations in HIV-1 gp41 Reveal a Correlation between HIV-1-mediated Bystander Apoptosis and Fusion/Hemifusion J. Biol. Chem., June 8, 2007; 282(23): 16899 - 16906. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Pauls, J. Senserrich, B. Clotet, and J. A. Este Inhibition of HIV-1 replication by RNA interference of p53 expression J. Leukoc. Biol., September 1, 2006; 80(3): 659 - 667. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
