Feline immunodeficiency virus envelope glycoprotein mediates apoptosis in activated PBMC by a mechanism dependent on gp41 function

Virology 330 (2004) 424 – 436 www.elsevier.com/locate/yviro

Feline immunodeficiency virus envelope glycoprotein mediates apoptosis in activated PBMC by a mechanism dependent on gp41 function Himanshu Garg1, Anjali Joshi, Wayne A. Tompkins* Immunology Program, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606, USA Received 27 August 2004; returned to author for revision 17 September 2004; accepted 5 October 2004

Abstract Feline Immunodeficiency Virus (FIV) is a lentivirus that causes immunodeficiency in cats, which parallels HIV-1-induced immunodeficiency in humans. It has been established that HIV envelope (Env) glycoprotein mediates T cell loss via a mechanism that requires CXCR4 binding. The Env glycoprotein of FIV, similar to HIV, requires CXCR4 binding for viral entry, as well as inducing membrane fusion leading to syncytia formation. However, the role of FIV Env in T cell loss and the molecular mechanisms governing this process have not been elucidated. We studied the role of Env glycoprotein in FIV-mediated T cell apoptosis in an in vitro model. Our studies demonstrate that membrane-expressed FIV Env induces apoptosis in activated feline peripheral blood mononuclear cells (PBMC) by a mechanism that requires CXCR4 binding, as the process was inhibited by CXCR4 antagonist AMD3100 in a dose-dependent manner. Interestingly, studies regarding the role of CD134, the recently identified primary receptor of FIV, suggest that binding to CD134 may not be important for induction of apoptosis in PBMC. However, inhibiting Env-mediated fusion post CXCR4 binding by FIV gp41-specific fusion inhibitor also inhibited apoptosis. Under similar conditions, a fusion-defective gp41 mutant was unable to induce apoptosis in activated PBMC. Our findings are the first report suggesting the potential of FIV Env to mediate apoptosis in bystander cells by a process that is dependent on gp41 function. D 2004 Elsevier Inc. All rights reserved. Keywords: FIV; Envelope; Apoptosis; Fusion; Syncytia; CD134; CXCR4

Introduction Feline Immunodeficiency Virus (FIV) infection, similar to HIV-1, is characterized by a progressive and irreversible depletion of CD4+ T lymphocytes leading to immunodeficiency in cats. Various hypotheses have been proposed for HIV-mediated T cell loss, including direct cytopathic effect due to infection (Lenardo et al., 2002), syncytia formation (Ferri et al., 2000), and autoimmune killing of T cells (Golding et al., 1989). Several inves-

* Corresponding author. Immunology Program, College of Veterinary Medicine, North Carolina State University, 4700 Hillsborough Street, Raleigh, NC 27606. Fax: +1 919 515 4237. E-mail address: [email protected] (W.A. Tompkins). 1 Current address: National Cancer Institute, Frederick, MD 21702, United States. 0042-6822/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2004.10.007

tigators have suggested that the loss of T cells in HIV infection is partly due to apoptosis of uninfected bystander cells (Alimonti et al., 2003; Famularo et al., 1997; Gougeon and Montagnier, 1993; Meyaar et al., 1992; Muro-Cacho et al., 1995). In this regard, Finkel et al. (1995) showed that apoptosis of lymph node T cells in HIV-infected individuals primarily occurs in uninfected bystander cells that lie in close proximity to productively infected cells. The mechanism(s) behind FIV-mediated T cell depletion are not clear, although observations mostly parallel findings in HIV-1. Several viral proteins including HIV Env (LaurentCrawford et al., 1995), Vpr (Desai et al., 2002), and Tat (Purvis et al., 1995) have been shown to induce apoptosis. However, determinants of HIV pathogenicity most often map to the Env glycoprotein (Etemad-Moghadam et al., 2001), suggesting that it may be a key regulator of CD4+ T

H. Garg et al. / Virology 330 (2004) 424–436

cell death. HIV Env is synthesized as a gp160 precursor that is later proteolytically cleaved into the gp120 surface unit and the gp41 transmembrane protein (Wyatt and Sodroski, 1998). While gp120 binds to CD4 and a chemokine coreceptor (CXCR4/CCR5), the gp41 mediates fusion of viral and cellular membranes leading to viral entry. Interestingly, FIV shares CXCR4 as a co-receptor (Willett et al., 1997) for Env binding, while the primary receptor has recently been identified as CD134 (Shimojima et al., 2004). Membraneexpressed Env glycoprotein also mediates fusion between HIV/FIV-infected cells and uninfected bystander cells expressing the necessary receptor/co-receptor, resulting in syncytia formation. HIV Env-mediated syncytia, frequently seen in in vitro infected T cell culture systems, undergo apoptosis and have been suggested as a mechanism of T cell loss by HIV (Ferri et al., 2000; Scheller and Jassoy, 2001). Membrane expressed Env glycoprotein has been shown to interact with bystander cells and cause apoptosis (LaurentCrawford et al., 1995), cytolysis (Nardelli et al., 1995), or reduced proliferation and G2 phase arrest followed by apoptosis (Kolesnitchenko et al., 1995; Schwartz et al., 1994). Studies with FIV have shown that T cells from infected cats exhibit reduced survival in vitro and die of apoptosis in a manner similar to that seen in HIV-1 infection (Guiot et al., 1997; Johnson et al., 1996; Momoi et al., 1996; Tompkins et al., 2003). The earliest distinctive pathological feature of FIV infections is a decline in the CD4+ T cell population (Hoffmann-Fezer et al., 1992) and an inversion of CD4/CD8 ratio (Novotney et al., 1990). The mechanism(s) of immune dysfunction in FIV infection are not clear, although Env glycoprotein of FIV shares remarkably similarities to HIV Env in terms of co-receptor usage and fusion mechanism. In this context, the Env glycoprotein of FIV may be an important target for further investigation. Although the role of FIV Env in T cell apoptosis has not been studied, it has been implicated in FIV-induced neuropathogenesis (Johnston et al., 2002). Further, FIV Env is a potent inducer of syncytia formation and similar to the X4 tropic strains of HIV, utilizes CXCR4 as co-receptor for mediating fusion and viral entry (Garg et al., 2004; Willett et al., 1997). Previously (Garg et al., 2004), we have shown that coculture of FIV Env expressing (CrFKenv/rev) cells with feline CD4+ T cell line (FCD4E) resulted in syncytia formation, which could be blocked by CXCR4 antagonist AMD3100 and gp41-specific fusion inhibitory peptide T1971. In this report, we show that membrane-expressed FIV Env is also capable of inducing apoptosis in activated PBMC and that this phenomenon correlates with Env-mediated fusion. This is the first report indicating the potential of FIV envelope glycoprotein in mediating T cell loss via a process that correlates withgp41 function rather than gp120 binding to receptor/co-receptor.

425

Results FIV Env mediates cell death in activated PBMC via apoptosis A number of studies have suggested that the progressive loss of CD4+ T cells in HIV and FIV infections is the result of bystander cell death (Alimonti et al., 2003; Corbeil and Richman, 1995; Laurent-Crawford et al., 1995; Tompkins et al., 2003). Although numerous studies have been conducted to elucidate the mechanism(s) of bystander cell death in HIV infection, there have been no studies pertaining to FIV infection in this regard. To address the hypothesis that FIV Env is capable of inducing cell death in bystander T cells, a coculture system was developed comprising of Env expressing CrFKenv/rev cells as effector cells and Concanvalin A (ConA)-activated PBMC as target cells. The model consisted of adherent effector cells and suspension target cells, which allows easy separation of the two populations for analysis. Previously, we have characterized the CrFKenv/rev cell line expressing Env glycoprotein from the NCSU1 isolate of FIV and studied the mechanisms of FIV Env-mediated syncytia formation in IL-2-dependent FCD4E cells (Garg et al., 2004). Activated PBMC were cocultured with Env expressing CrFKenv/rev cells and the viability of target cells was measured 24 and 48 h later by MTT assay. As shown in Fig. 1, coculture of activated PBMC with CrFKenv/rev cells resulted in loss of viability at 24 h as compared to cells cocultured with control CrFK cells (P b 0.01), and this difference was more marked at 48 h post coculture (P b 0.001). To further characterize the loss of viability in target cells, flow cytometry was used to assess cell morphology based on forward (FSC) versus side scatter (SSC). Twenty-four hours post coculture, the non-adherent cells were collected

Fig. 1. Coculture of activated PBMC with Env-expressing cells results in loss of viability. PBMC activated with 2 Ag/ml ConA were cocultured with either CrFK cells (Control) or Env-expressing CrFKenv/rev cells (Env) for 24 or 48 h. The non-adherent cells were collected after careful washing and viability determined by MTT assay. Data represent mean F standard deviation (SD) of quadruplicate observations. Statistical significance was assessed by Student’s t test (*P b 0.01, **P b 0.001).

426

H. Garg et al. / Virology 330 (2004) 424–436

and analyzed by flow cytometry. The cells were gated into R1 (high FCS, low SSC) and R2 (low FSC, high SSC) gates representing live and dead cells, respectively. As depicted in Fig. 2A, there was a significant decrease in percent cells in R1 (live) gate when activated PBMC were cocultured with CrFKenv/rev cells compared to control CrFK cells. Pooled data from three independent experiments shows an approximate 40% decrease in viability in PBMC cocultured with Env-expressing cells compared to control. To determine if FIV Env-mediated loss in cellular viability was a result of apoptosis, the FSC/SSC gate was applied to morphologically live cells (R1 gate) as depicted in Fig. 2A and percent apoptosis determined by Annexin V staining. Cells positive for Annexin V but negative for propidium iodide (PI) were considered apoptotic. As shown in Fig. 2B, a marked increase in apoptotic cells was seen when target cells were cocultured with CrFKenv/ rev cells (12.6%) compared to control CrFK cells (6.2%). Data from three separate experiments in Fig. 2B show a twofold increase in percent apoptotic cells when cocultured with Env-expressing cells compared to control cells (P b 0.001). Mitogenic stimulation is required for induction of apoptosis by FIV Env and correlates with up-regulation of surface CXCR4 but not CD134 To determine whether T cell activation was necessary for induction of apoptosis, experiments were conducted

with either resting cells or cells stimulated with ConA for 4 days. As shown in Fig. 3A, resting unstimulated PBMC did not show any significant cell death or apoptosis when cocultured with Env-expressing cells compared to activated cells (P b 0.01), suggesting that cellular activation is required for this process. The potential of FIV Env to mediate cell death only in activated but not resting PBMC suggests that there may be differences between the receptor (CD134) and co-receptor (CXCR4) expression levels between the two cell populations. Hence, we determined the expression of these receptors on activated versus resting cells as a determinant of differences seen in the previous experiments. Flow cytometric analysis of surface expression of CXCR4 and CD134 showed that ConA mediated activation of feline PBMC results in upregulation of CXCR4 but not CD134 (Fig. 3B). This suggests that Env-mediated apoptosis in ConA-activated PBMC correlates with up-regulation of surface CXCR4 expression and leads to the possibility that the majority of dying cells are more likely to be CXCR4 positive than CD134 positive. CD134 can act as a receptor for fusion mediated by FIV-NCSU1 Env glycoprotein In the previous experiment, apoptosis in ConA-stimulated cultures did not correlate with surface expression of CD134. Hence, we attempted to investigate whether the NCSU1 isolate of FIV similar to other primary isolates

Fig. 2. Coculture of activated PBMC with Env-expressing cells results in morphological changes and apoptosis. (A) ConA-activated PBMC were cocultured with either CrFK cells (Control) or Env-expressing CrFKenv/rev cells (Env) for 24 h. The non-adherent cells were collected after careful washing and analyzed by flow cytometry for morphological changes based on forward scatter (FSC) versus side scatter (SSC). Graphical representation of data from three independent experiments is shown in the right panel. Data represent mean of triplicate observations from three independent experiments. (B) Twenty-four hours post coculture, the non-adherent cells were stained with Annexin V-FITC and PI and analyzed by flow cytometry for apoptosis. The gate was applied to the live population (R1 gate) as in part A. Graphical representation of data from three independent experiments is shown in the right panel. Data are mean F SD of triplicate observations. Statistical significance was assessed by Student’s t test (*P b 0.01).

H. Garg et al. / Virology 330 (2004) 424–436

427

Fig. 3. Mitogenic stimulation of PBMC is required for induction of apoptosis via FIV Env. (A) Feline PBMC were either cultured in media alone (Resting) or in media supplemented with 2 Ag/ml ConA (Activated) for 4 days followed by coculture with either Env-expressing or control CrFK cells for 24 h. Apoptosis was subsequently determined by Annexin V and PI staining. (B) Flow cytometric analysis of surface expression of CXCR4 and CD134 in PBMC either with (Activated) or without (Resting) stimulation with ConA for 4 days. Thin line: secondary antibody alone, bold line: specific staining for CXCR4 or CD134. Numbers represent the percentage of cells expressing CXCR4 or CD134.

utilizes CD134 as the primary receptor. For this purpose, a syncytia forming assay was conducted using CD134 transfected HeLa cells and CrFKenv/rev cells to determine whether FIV-NCSU1 binds/utilizes CD134. Syncytia forming assay in Fig. 4 shows that expression of feline CD134 in HeLa cells renders them permissive to FIV-NCSU1 Envmediated fusion, which otherwise does not mediate fusion in parental HeLa cells (Fig. 4) as shown previously (Garg et al., 2004). For control purposes, fusion with CrFKpet cells is shown, which fuse with HeLa cells irrespective of CD134 expression (Fig. 4). Interestingly, though CXCR4 antagonist AMD3100 blocked fusion under different conditions, anti-human CD134 antibody previously shown to cross-react with feline CD134 (Shimojima et al., 2004) failed to block the fusion process (Table 1).

CXCR4 binding and not CD134 is required for FIV Env-mediated apoptosis Studies with HIV-1 have implicated CD4 and CXCR4 binding/signaling as one of the mechanisms for HIV Envmediated apoptosis in bystander T cells (Arthos et al., 2002; Roggero et al., 2001). Hence, we wished to study whether FIV Env binding to its surface receptor (CD134) and coreceptor (CXCR4) was required for induction of apoptosis. Our data in Fig. 4 indicates that FIV-NCSU1 similar to other primary isolates of FIV binds CD134, although a blocking antibody or reagent specific to feline CD134 is not available. To study the role of CD134 binding in FIV Env-mediated apoptosis, we utilized the cross-reactive monoclonal antibody to deplete activated PBMC of CD134-positive cells to

428

H. Garg et al. / Virology 330 (2004) 424–436

Fig. 4. FIV-NCSU1 can utilize feline CD134 to mediate fusion. Untransfected HeLa cells or those transfected with feline CD134 expressing plasmid (HeLa CD134) were cocultured either with cells expressing FIV-NCSU1 Env (NCSU1) or with CrFKpet cells chronically infected with FIV-Petaluma isolate (Petaluma). The cells were fixed and stained with Giemsa stain. Representative photomicrographs are shown. Magnification 200.

high purity (Fig. 5A) using Fluorescence Activated Cell Sorting (FACS) and then determine percent apoptosis either in total PBMC or CD134-depleted cells. As shown in Fig. 5B, there was no difference in apoptosis induced by FIV-NCSU1 Env in either total or CD134-depleted PBMC, suggesting that CD134 binding may not be an absolute requirement for this process. Further, infection studies with CD134-depleted PBMC (data not shown) indicated that they are equally infectable with FIV-NCSU1, suggesting that although NCSU1 can utilize CD134 it may not be restricted to using this receptor in primary cells. Our next objective was to determine whether CXCR4 binding was required for FIV Env-mediated apoptosis. Previously, we have shown that CXCR4 antagonist AMD3100 can block fusion mediated by FIV-NCSU1 Env in FCD4E cells in a dose dependent manner (Garg et al., 2004). Based on these findings, we expected to block FIV Env-mediated apoptosis in activated feline PBMC using AMD3100. As shown in Fig. 5C, AMD3100 blocked EnvTable 1 AMD3100 but not anti-CD134 monoclonal antibody blocks syncytia formation under different coculture conditions Medium CRFKrev/env + HeLa CrFKrev/env + HeLa CD134 CrFKpet + HeLa CrFKpet + HeLa CD134

+ + +

AMD

Anti-CD134 + + +

Cultures were set up with different cell types as described in Fig. 4. In some wells, AMD3100 or anti-CD134 antibody was added to block fusion. Twenty-four hours post-coculture, cells were fixed, stained with Giemsa stain, and subsequently photomicrographed. Presence of syncytia under different culture conditions was recorded as (+) and absence of syncytia as ( ).

mediated apoptosis in ConA-stimulated feline PBMC in a dose-dependent manner. The above results confirm that apoptosis seen in our model was due to membraneexpressed FIV Env and that CXCR4 binding was required for this process. FIV Env-mediated apoptosis is dependent on gp41 function FIV Env-mediated fusion can be inhibited by gp41specific peptide T1971 that corresponds to the C terminal heptad repeat region of gp41. T1971, an FIV-specific peptide, binds gp41 at a post-CXCR4 binding stage inhibiting the six-helix hairpin structure formation thereby blocking fusion (Garg et al., 2004; Medinas et al., 2002). Interestingly, in our coculture system, T1971 was able to block FIV Env-mediated apoptosis in a dose-dependent manner (Fig. 6A). This suggests that CXCR4 binding alone is not sufficient for Env-induced apoptosis and that steps subsequent to CXCR4 binding involving gp41 function are important in mediating the apoptotic process. To further establish the above findings, we utilized a gp41 mutant defective in mediating fusion in our study. A conserved tryptophan-rich region in HIV gp41 comprising the membrane proximal ectodomain, also referred to as the pre-transmembrane (pre-TM) region (Suarez et al., 2000), has been identified as being critical for fusion (Salzwedel et al., 1999). Previously, we have identified a similar conserved region in FIV gp41 (Garg et al., 2004) and established a fusion defective mutant, W(1–3)A, by replacing the tryptophans (W) at position 766, 769, and 772 with alanines (A) as represented in Fig. 6B. We assayed the apoptosis inducing potential of W(1–3)A mutant and compared it to the wild-type (WT) clone. As shown in

H. Garg et al. / Virology 330 (2004) 424–436

429

Fig. 5. FIV Env-mediated apoptosis correlates with CXCR4 binding but not with CD134 binding. (A) ConA-activated PBMC were depleted of CD134+ cells using the MoFlo fluorescence activated cells sorter. The purity of CD134-depleted cells is shown along with total PBMC. Numbers represent the percentage of cells within that quadrant. (B) Either total PBMC or PBMC depleted of CD134+ cells were cocultured with Env-expressing or control cells. Apoptosis was determined 24 h later by Annexin V and PI staining. (C) Total PBMC activated with 2 Ag/ml ConA for 4 days were cocultured with either CrFK cells or Envexpressing CrFKenv/rev cells for 24 h. CXCR4 antagonist AMD3100 was added at the indicated concentrations at the time of coculture. Apoptosis was determined by Annexin V staining followed by flow cytometry 24 h post coculture. Data are mean F SE from triplicate observations from two independent experiments.

Fig. 6C, the fusion defective mutant W(1–3)A was unable to induce apoptosis in activated PBMC compared to the WT Env ( P b 0.01) under similar expression levels (Fig. 6C inset). These findings further suggest that a fusion competent FIV Env glycoprotein is required for induction of apoptosis. Syncytia formation per se is not sufficient for induction of apoptosis Our previous experiments suggest a correlation between fusion and apoptosis even though syncytia formation

is not evident in cocultures using PBMC. Previously, we have utilized the feline CD4+ T cell line, FCD4E cells that form extensive syncytia when cocultured with CrFKenv/ rev cells (Garg et al., 2004). To determine whether syncytia formation was sufficient to induce apoptosis in our system, we cocultured FCD4E cells with Envexpressing cells and compared them to PBMC for apoptosis induction. As shown in Fig. 7A, FCD4E cells did not undergo significant apoptosis when cocultured with Env-expressing cells compared to activated PBMC. However, under similar conditions, FCD4E cells formed extensive syncytia while in cocultures using activated

430

H. Garg et al. / Virology 330 (2004) 424–436

FIV Env-mediated apoptosis correlates with Env-mediated fusion/gp41 function Our coculture experiments with PBMC did not show significant syncytia formation (Fig. 7B) even though we believe that small syncytia comprising of a few cells maybe formed. Fluorescent dye transfer assay may identify small syncytia or cells that have undergone cytoplasmic content mixing due to fusion pore formation. Hence, we utilized a dye transfer assay to determine the population of cells undergoing apoptosis, and whether there was a correlation between FIV Env-mediated apoptosis and fusion. For this purpose, FIV Env-expressing cells were labeled with CMTMR (Cell Tracker Orange) dye prior to coculture with CMFDA (Cell Tracker Green) labeled PBMC for studying fusion. For apoptotic assays, only the Env-expressing cells were labeled with CMTMR prior to coculture with unlabeled PBMC and followed by staining with Annexin V-FITC post coculture.

Fig. 6. FIV Env-mediated apoptosis is dependent on gp41 function. (A) ConA-activated PBMC were cocultured either with CrFK cells or Envexpressing CrFKenv/rev cells for 24 h. Gp41 C-terminal-specific inhibitory peptide T1971 was added at the indicated concentrations at the time of coculture. Apoptosis was determined by Annexin V staining 24 h post coculture followed by flow cytometry. Data are mean F SE of triplicate observations from two independent experiments. (B) Schematic diagram of FIV gp41 and fusion-defective mutant W(1–3)A. The position of heptad repeats (HR1 and HR2), transmembrane region (TM), and pre-transmembrane region (pre-TM) is shown. (C) CrFK cells were transiently transfected with either wild type (pWT) or mutant Env (pW(1–3)A), or vector control (pControl) plasmids. Inset shows protein expression by Western blot. Forty-eight hours post-transfection, cells were cocultured with ConA-activated PBMC. Twenty-four hours post-coculture, suspension cells were collected after careful washing, stained with Annexin V and PI and analyzed by flow cytometry for apoptosis. Data represent mean F SD of triplicate observations. Statistical significance was assessed by Student’s t test (*P b 0.01).

PBMC syncytia formation was not evident (Fig 7B). This suggests that syncytia formation per se may not be sufficient for FIV Env-mediated apoptosis, although the fusogenic potential or membrane disruption capacity of gp41 is necessary.

Fig. 7. Fusion per se is not sufficient for induction of apoptosis by FIV Env. (A) Either ConA activated PBMC or highly fusogenic FCD4E cells were cocultured with Env-expressing or control cells for 24 h. Apoptosis was determined by Annexin V staining followed by flow cytometry. Data are mean F SD of triplicate observations. (B) Photomicrographs of the different cocultures as described in A 24 h post-coculture and prior to analysis of apoptosis. Magnification 200.

H. Garg et al. / Virology 330 (2004) 424–436

431

Fig. 8. FIV Env-mediated apoptosis correlates with Env-mediated fusion. CrFKenv/rev or control CrFK cells were pre-labeled with CMTMR (red) and cocultured with CMFDA (green) labeled PBMC (A,B) or unlabeled PBMC (C) for 24 h. (A) CMFDA-labeled PBMC were cocultured with CMTMR-labeled CrFK control cells or Env-expressing CrFKenv/rev cells for 24 h. Following coculture, the non-adherent cells were collected, cytospun onto slides, and observed by fluorescent microscopy. Images acquired for red and green fluorescence were overlaid using Adobe Photoshop software. Arrows indicate multicellular syncytia appearing as yellow in the overlay. Arrowheads indicate single cells that are double positive. (B) Fusion was estimated 24 h postcoculture by flow cytometry as CMTMR- and CMFDA-positive cells. (C) Apoptosis was detected by Annexin V staining followed by flow cytometry 24 h post-coculture. For B and C, numbers represent the percentage of cells within each quadrant or within each specified gate. (D) Morphology of different cell populations (gates R2 through R5) in B and C based on forward scatter (FSC) and side scatter (SSC). (E) Graphical representation of data for (B) and (C). Data are mean F SD of triplicate observations.

432

H. Garg et al. / Virology 330 (2004) 424–436

Twenty-four hours post coculture, fusion was quantified by flow cytometry as CMTMR+ CMFDA+ double-positive cells and apoptosis was determined by staining with FITC labeled Annexin V (CMTMR+/ Annexin+/ ). As seen in Fig. 8A, fused cells (appearing as yellow in overlay) were distinctly seen by fluorescent microscopy when activated PBMC were cocultured with Env-expressing CrFKenv/rev cells compared to control cells. Fusion here did not necessarily represent syncytia (arrows) as single cells that were double positive were also seen (arrowheads) most likely due to formation of fusion pore without progressing to complete fusion. A quantification of fusion/dye transfer by flow cytometry revealed that approximately 10% of the cells were double positive in Env cocultures (11.88%) when compared to background control cocultures (1.62%). We then determined percent apoptosis in cocultured target cells by Annexin V staining under identical coculture conditions (except that the target PBMC were unlabeled). Interestingly, a similar increase of 10% (5.94–17.01%) was seen in CMTMR+Annexin V+ cells in Env cocultures compared to control cocultures. Further, there was no significant increase in apoptosis of CMTMR cells (9.17–10.54%, Fig. 8C). Since the previously unlabelled target cells (activated PBMC) had taken up the CMTMR dye, this population most likely represents cells that have formed fusion pores with Env-expressing cells. To further validate the above findings, we determined whether the fused and apoptotic populations were morphologically similar based on the FSC/SSC characteristics. As seen in Fig. 8D, the fused CMTMR+CMFDA+ (R2 gate) cells were almost identical to the apoptotic CMTMR+Annexin V+ (R4 gate) cells based on forward and side scatter characteristics. Moreover, they were morphologically distinct from CMTMR negative (R3 gate) and non-apoptotic (R5 gate) cells. These results suggest that apoptosis in feline PBMC was primarily occurring post fusion pore formation as a result of coculture with Env-expressing cells, indicating a strong correlation between FIV Env-mediated fusion and apoptosis.

Discussion Feline immunodeficiency virus infection is a relevant and well-characterized animal model for human AIDS (English et al., 1993; Mizuno et al., 2001). Based on the similarities between FIV and HIV pathogenesis, it is speculated that both viruses share common mechanism(s) for inducing immunodeficiency. Even though it has been known for sometime that FIV causes T cell depletion in a manner similar to HIV-1 (Guiot et al., 1997; Johnson et al., 1996; Tompkins et al., 2003), the molecular mechanisms involved in FIV-induced T cell loss have not been established. The Env glycoprotein of HIV has been implicated in apoptotic killing of bystander T cells leading to immunodeficiency (Etemad-Moghadam et al., 2001). Conflicting reports

suggest the role of gp120 binding to CD4 and chemokine receptors versus fusion mediated by Env glycoprotein in initiating apoptotic signaling. The objective of the present study was to determine (1) whether FIV Env glycoprotein was capable of inducing apoptosis in bystander cells and (2) if this process was dependent on receptor (CD134), coreceptor (CXCR4) binding, or Env-mediated fusion. The fact that FIV Env utilizes CD134 and not CD4 as its primary receptor (Hosie et al., 1993; Shimojima et al., 2004) but can still initiate apoptosis in bystander cells indicates that CD4 signaling may not be necessary for lentiviral Env-mediated apoptosis. We show here that membraneexpressed FIV Env can induce apoptosis in activated PBMC but not resting cells, and this correlates with increased surface expression of the CXCR4 but not CD134. While CXCR4 is a well-known activation marker (Bleul et al., 1997), our observations with regard to lack of CD134 upregulation following ConA stimulation are consistent with findings in other species where ConA-mediated activation failed to up-regulate surface CD134 expression (Barten et al., 2001). Interestingly, similar to HIV-1, FIV Env binding to CXCR4 is required for apoptosis, suggesting that it is a common requirement for both HIV and FIV Env-induced T cell death. However, inhibition of FIV Env-mediated apoptosis at steps post CXCR4 binding using either gp41specific inhibitory peptides or a mutation in gp41 pretransmembrane region suggests that CXCR4 itself may not be sufficient for activating the apoptotic signaling pathway. Although our studies regarding the primary receptor CD134 were not conclusive in the absence of suitable blocking antibodies, inhibition of the fusion process at steps post receptor and co-receptor binding abrogates apoptosis, suggesting that the binding of CD134 and/or CXCR4 is not sufficient for induction of FIV-NCSU1 Env-mediated apoptosis. The observation that fusion-inhibiting agents/ mutations inhibit T cell apoptosis indicates a strong correlation between the two processes. Our findings are contrary to the hypothesis that CXCR4 binding and signaling is sufficient for lentivirus Env-mediated apoptosis (Arthos et al., 2002; Penn et al., 1999), but are in agreement with recent reports showing that HIV Env-mediated apoptosis can be blocked by T20, an HIV gp41-specific fusion inhibitor (Barretina et al., 2003; Blanco et al., 2003; Scheller and Jassoy, 2001). The correlation between HIV Env-mediated fusion and apoptosis remains controversial. While some reports show the potential of soluble Env glycoprotein to induce apoptosis in T cells (Corbeil and Richman, 1995), others have demonstrated the necessity of membrane-bound Env glycoprotein for induction of apoptosis (Laurent-Crawford et al., 1995). Though it has been established that HIV Env can signal via chemokine receptors (CXCR4/CCR5) (Davis et al., 1997), and some authors suggest that HIV Env mediates apoptotic signaling via engaging CD4 and/or CXCR4 (Arthos et al., 2002; Vlahakis et al., 2001), others have shown that signaling via either of these receptors is

H. Garg et al. / Virology 330 (2004) 424–436

not required for HIV Env-mediated apoptosis (BiardPiechaczyk et al., 2000; Blanco et al., 1999). While pathogenicity of HIV is often related to the fusogenic potential of the Env glycoprotein (Alimonti et al., 2003; Gougeon and Montagnier, 1993), the requirement of fusion between infected and uninfected bystander cells for apoptosis remains controversial. HIV infections are characterized highly cytopathic syncytia-inducing (SI) and relatively less pathogenic nonsyncytia-inducing (NSI) strains based on the in vitro phenotype of the viruses. The likelihood of finding SI phenotype in a patient is associated with poor immunological parameters, rapid CD4+ T cell loss and disease progression (Koot et al., 1993; Richman and Bozzette, 1994). The mechanism by which SI strains tend to be more pathogenic than NSI strains is not established. SI strains utilize CXCR4 as co-receptor, which is expressed practically on all CD4+ T cells while NSI strains generally utilize CCR5, which is expressed only on a limited number of CD4+ T cells especially in the gut-associated lymphoid tissue (Anton et al., 2000). Based on co-receptor expression by T cells, X4 viruses tend to be more infectious, which may be a cause of their increased pathogenicity (Fouchier et al., 1996). On the other hand, cells infected with SI strains would be more effective at causing apoptosis of CXCR4 expressing bystander T cells via binding by Env glycoprotein and mediating fusion. Even though NSI strains are considered less cytopathic, they have been shown to be capable of CD4+ T cell killing by a mechanism that is not directly related to the rate of viral replication (Yu et al., 1994). In a recent report, LaBonte et al. (2003) have shown that CCR5 utilizing viruses can cause cytolysis of CCR5+ target cells by a process that is again dependent on membrane fusion. Our findings utilizing the highly fusogenic FCD4E cells suggests that although syncytia formation itself may not be the determining factor in inducing apoptosis, the fusogenic potential and membrane disruption properties of FIV Env are required for this process. Further analysis of this phenomenon is necessary utilizing membrane specific lipophilic dyes and nondiffusible cytoplasmic labeling. We believe that the discrepancy between FCD4E cells and PBMC might be due to differences in membrane structure and organization as well as signaling factors. This correlates with the observation that although pathogenicity of HIV correlates with the fusogenic potential of Env glycoprotein, formation of syncytia in vivo are rare. Based on the above evidence, it is clear that interaction of lentiviral Env glycoprotein with target cell membranes may result in membrane destabilization subsequently leading to apoptosis of target cells. Our findings are in agreement with others (Roumier et al., 2003), further strengthening the hypothesis that gp41-mediated membrane perturbation may be important for FIV as well as HIV-induced apoptosis. This is the first report showing the potential of membraneexpressed FIV Env to induce apoptosis in bystander cells.

433

Also, by blocking Env-mediated apoptosis at the level of gp41, we show a strong correlation between fusion and apoptosis mediated by FIV Env. Our findings not only strengthen the hypothesis that lentiviral Env glycoproteins may be involved in T cell apoptosis but also indicate strong similarities between FIV and HIV pathogenesis.

Materials and methods Cells and reagents Crandell Feline Kidney (CrFK), CrFK cells expressing FIV-NCSU1 Env glycoprotein (CrFKenv/rev) (Garg et al., 2004), and CrFKpet (CrFK cells chronically infected with the Petaluma isolate of FIV) were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) (Mediatech) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml), and streptomycin (100 Ag/ml). Medium for CrFKenv/rev cells was supplemented with G-418 (GibcoBRL, Gaithesberg, MD) at 700 Ag/ml. HeLa cells were maintained in DMEM medium supplemented with 10% FBS, penicillin (100 U/ml) and streptomycin (100 Ag/ml). FCD4E, a CD4+ T cell line highly permissive for FIVNCSU1 infection (English et al., 1993) was cultured in RPMI medium supplemented as described below. Peripheral blood lymphocytes were obtained from FIV negative cats by Percoll density gradient centrifugation as described previously (English et al., 1993). The donor cats were housed at the Laboratory Animal Resource facility of the College of Veterinary Medicine, North Carolina State University, as per the federal guidelines and institutional policies. Purified PBMC were cultured in RPMI 1640 medium (Mediatech) supplemented with 10% FBS, streptomycin (100 Ag/ml), penicillin (100 IU/ml), l-glutamine (2 mM), HEPES (15 mM), sodium pyruvate (2 mM) and hmercaptoethanol (2.5  10 5 M). PBMC were activated with 2 Ag/ml ConA (Sigma) and cultured for 4 days prior to use in the coculture experiments. Plasmid expressing feline CD134 was a kind gift from Dr. Brian Willett (University of Glasgow). CXCR4 antagonist AMD3100 was a kind gift from Dr. Edward Hoover (Colorado State University). FIV gp41-specific fusion inhibitor T1971, shown to inhibit FIV Env-mediated fusion, was a kind gift from Robyn Medinas (Trimeris Inc. Durham, NC). Flow cytometry and cell sorting Cell surface CXCR4 expression was measured using the cross-reacting anti-CXCR4 monoclonal antibody (44717, BD Biosciences) followed by staining with Phycoerythrin (PE) conjugated goat anti-mouse IgG (Jackson ImmunoResearch Laboratories, PA). PE-conjugated anti CD134 antibody BerACT35 (Ancell Corporation, Bayport, MN) previously shown to cross-react with the feline CD134 homolog (Shimojima et al., 2004) was used to detect surface

434

H. Garg et al. / Virology 330 (2004) 424–436

CD134 expression. Cells were acquired using the FACSCalibur flow cytometer (BD Biosciences) and analyzed using the CellQuest software. For blocking experiments, the nonconjugated anti-CD134 antibody was used. PBMC were sorted into purified CD134+ Gaithesberg and CD134 populations using a high-speed MoFlo fluorescence activated cell sorter (DakoCytomation) at the flow cytometry facility at North Carolina State University. Coculture experiment CrFK or CrFKenv/rev cells were seeded in 24-well plates at 2  105 cells/well and allowed to adhere overnight. After 24 h, medium was aspirated from the wells and 4  105 PBMC activated with 2 Ag/ml ConA for 4 days were added to the wells. Inhibitors AMD3100 or T1971 were added at the time of coculture. The cells were cocultured for 24 or 48 h. Subsequently, the suspension cells were collected by careful washing and used for viability or apoptotic assays. Transfection of cells HeLa cells were transfected using Effectene transfection reagent (Qiagen, CA) as per the manufacturer’s instructions. Twenty-four hours post transfection, the cells were seeded in 96-well plates and used for syncytia forming assay. Syncytia-forming assay HeLa cells either untransfected or transfected with feline CD134 were seeded in 96-well plates at 104 cells per well and allowed to adhere overnight. Subsequently, CrFK cells expressing FIV-NCSU1 Env (CrFKenv/rev) or chronically infected with FIV-Petaluma (CrFKpet) were added at 1000 cells per well. In some wells, either AMD3100 or antiCD134 antibody was added for blocking. The cells were cocultured for 24 h after which the plates were fixed and stained with Giemsa stain as described previously (Garg et al., 2004). Measurement of viability Viability of PBMC was measured by the MTT [3-(4,5dimethylthiozol-2-yl)-2,5-diphenyl tetrazolium bromide] (Sigma) assay as described previously (Mosmann, 1983) with slight modifications. Briefly, ConA-activated PBMC were cocultured with either CrFK or CrFKenv/rev cells for 24 or 48 h following which the non-adherent cells were collected and re-plated in 96-well plates at 100 Al/well. Thirty microlitres of MTT reagent (5 mg/ml in PBS) was added to each well and incubated for 4 h. Subsequently, the plates were centrifuged, supernatant discarded and 100 Al of 0.1 N HCl in isopropanol added to the wells. The plates were incubated for an additional 1 h and then 100 Al of distilled water added per well. Finally, the plates were read at 530/650 nm in an ELISA plate reader (Tecan).

Detection of apoptosis Morphological change in suspension cells was detected 24 h post coculture by measuring the forward scatter (FSC) versus side scatter (SSC) by flow cytometry on a FACSCalibur flow cytometer (BD Biosciences). At least 15 000 events were collected and analyzed using the CellQuest software. Apoptosis was detected by staining suspension cells with FITC-conjugated Annexin V, a 35–36 kDa protein that binds to phosphatidylserine exposed on the surface of apoptotic cells. Twenty-four hours post coculture, the suspension cells were stained with Annexin V and PI (Roche Diagnostics, IN) as per the manufacturer’s instructions and analyzed by flow cytometry. Cells positive for Annexin V but negative for PI were considered apoptotic, while PI-positive cells were excluded as dead cells. Inhibitors AMD3100 or T1971 were added at the time of coculture for some experiments. Detection of fusion and apoptosis using fluorescent labeling of cells To determine FIV Env-mediated fusion in activated PBMC, a two-color dye redistribution assay was developed. Briefly, CrFK or CrFKenv/rev cells were labeled with cytoplasmic dye 5- and 6-{[(4-choloromethyl) benzoyl] amino} tetramethyl rhodamine (CMTMR/Cell Tracker Orange, Molecular Probes) (5AM in PBS for 30 min), washed and seeded in 24-well plates at 2  105 cells per well. Next day, activated PBMC labeled with cytoplasmic dye 5-chloromethylfluorescein diacetate (CMFDA/Cell Tracker Green, Molecular Probes) (0.5 AM in PBS for 30 min) were added to the wells at 4  105 cells per well. The cells were cocultured for 24 h, following which suspension cells were analyzed either by fluorescent microscopy or by flow cytometry. A similar assay with slight modification was used to detect apoptosis in activated PBMC after coculture with Env-expressing cells. CrFKenv/rev or nontransfected CrFK cells labeled with CMTMR dye were cocultured with unlabeled ConA-activated PBMC. Apoptosis was detected the subsequent day by staining suspensions cells with Annexin V. Cells positive for Annexin V but negative for CMTMR were single unfused PBMC that were apoptotic. Cells positive for Annexin V as well as CMTMR were Env-expressing cells undergoing apoptosis, most likely post fusion.

Acknowledgments The authors would like to thank Dr. Edward Hoover (Colorado State University) for AMD3100, Robyn Medinas (Trimeris Inc. Durham, NC) for T1971, and Dr. Brian Willett (University of Glasgow) for the feline CD134 expression plasmid. The authors are also thankful to Dr. Frederick Fuller (N.C. State University) for critical review

H. Garg et al. / Virology 330 (2004) 424–436

of the manuscript. This work was supported in part by National Institute of Health grant AI43858.

References Alimonti, J.B., Ball, T.B., Fowke, K.R., 2003. Mechanisms of CD4+ T lymphocyte cell death in human immunodeficiency virus infection and AIDS. J. Gen. Virol. 84, 1649 – 1661. Anton, P.A., Elliott, J., Poles, M.A., McGowan, I.M., Matud, J., Hultin, L.E., Grovit-Ferbas, K., Mackay, C.R., Chen, I.S.Y., Giorgi, J.V., 2000. Enhanced levels of functional HIV-1 co-receptors on human mucosal T cells demonstrated using intestinal biopsy tissue. AIDS 14, 1761 – 1765. Arthos, J., Cicala, C., Selig, S.M., White, A.A., Ravindranath, H.M., Van Ryk, D., Steenbeke, T.D., Machado, E., Khazanie, P., Hanback, M.S., Hanback, D.B., Rabin, R.L., Fauci, A.S., 2002. The role of the CD4 receptor versus HIV co-receptors in envelope-mediated apoptosis in peripheral blood mononuclear cells. Virology 292, 98 – 106. Barretina, J., Blanco, J., Armand-Ugon, M., Gutierrez, A., Clotet, B., Este, J.A., 2003. Anti-HIV-1 activity of enfuvirtide (T-20) by inhibition of bystander cell death. Antiviral Ther. 8, 155 – 161. Barten, M.J, Gummert, J.F, van Gelder, T., Shorthouse, R., Morris, R.E., 2001. Flow cytometric quantitation of calcium-dependent and -independent mitogen-stimulation of T cell functions in whole blood: inhibition by immunosuppressive drugs in vitro. J. Immunol. Methods 253, 95 – 112. Biard-Piechaczyk, M., Robert-Hebmann, V., Richard, V., Roland, J., Hipskind, R.A., Devaux, C., 2000. Caspase-dependent apoptosis of cells expressing the chemokine receptor CXCR4 is induced by cell membrane-associated human immunodeficiency virus type 1 Envelope glycoprotein (gp120). Virology 268, 329 – 344. Blanco, J., Jacotot, E., Cabrera, C., Cardona, A., Clotet, B., De Clercq, E., Este, J.A., 1999. The implication of the chemokine receptor CXCR4 in HIV-1 Envelope protein-induced apoptosis is independent of the G protein-mediated signaling. AIDS 13, 909 – 917. Blanco, J., Barretina, J., Ferri, K.F., Jacotot, E., Gutierrez, A., ArmandUgon, M., Cabrera, C., Kroemer, G., Clotet, B., Este, J.A., 2003. Cellsurface-expressed HIV-1 envelope induces the death of CD4 T cells during GP41-mediated hemifusion-like events. Virology 305, 318 – 329. Bleul, C., Wu, C.L., Hoxie, J.A., Springer, T.A., Mackay, C.R., 1997. The HIV co-receptors CXCR4 and CCR5 are differentially expressed and regulated on human T lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 94, 1925 – 1930. Corbeil, J., Richman, D.D., 1995. Productive infection and subsequent interaction of CD4-gp120 at the cellular membrane is required for HIVinduced apoptosis of CD4+ T cells. J. Gen. Virol. 76, 681 – 690. Davis, C.B., Dikic, I., Unutmaz, D., Hill, C.M., Arthos, J., Siani, M.A., Thompson, D.A., Schlessinger, J., Littman, D.R., 1997. Signal transduction due to HIV-1 Envelope interactions with chemokine receptors CXCR4 or CCR5. J. Exp. Med. 86, 1793 – 1798. Desai, B.M., Zhang, D., Dayes, N., Green, D.R., Weiner Muthumani, K., Hwang, D.S., 2002. HIV-1 Vpr induces apoptosis through caspase 9 in T cells and peripheral blood mononuclear cells. J. Biol. Chem. 277, 37820 – 37831. English, R.V., Johnson, C.M., Gebhard, D.H., Tompkins, M.B., 1993. In vivo lymphocyte tropism of feline immunodeficiency virus. J. Virol. 67, 5175 – 5186. Etemad-Moghadam, B., Rhone, D., Steenbeke, T., Sun, Y., Manola, J., Gelman, R., Fanton, J.W., Racz, P., Tenner-Racz, K., Axthelm, M.K., Letvin, N.L., Sodroski, J., 2001. Membrane-fusing capacity of the human immunodeficiency virus envelope proteins determines the efficiency of CD+ T-cell depletion in macaques infected by a simianhuman immunodeficiency virus. J. Virol. 75, 5646 – 5655. Famularo, G., De Simone, C., Marcellini, S., 1997. Apoptosis: mechanisms and relation to AIDS. Med. Hypotheses 48, 423 – 429.

435

Ferri, K.F., Jacotot, E., Geuskens, M., Kroemer, G., 2000. Apoptosis and karyogamy in syncytia induced by the HIV-1-Envelope glycoprotein complex. Cell Death Differ. 7, 1137 – 1139. Finkel, T.H., Tudor-Williams, G., Banda, N.K., Cotton, M.F., Curiel, T., Monks, C., Baba, T.W., Ruprecht, R.M., Kupfer, A., 1995. Apoptosis occurs predominantly in bystander cells and not in productively infected cells of HIV- and SIV-infected lymph nodes. Nat. Med. 1, 129 – 134. Fouchier, R.A., Meyaard, L., Brouwer, M., Hovenkamp, E., Schuitemaker, H., 1996. Broader tropism and higher cytopathicity for CD4+ T cells of a syncytium-inducing compared to a non-syncytium-inducing HIV-1 isolate as a mechanism for accelerated CD4+ T cell decline in vivo. Virology 219, 87 – 95. Garg, H., Fuller, F., Tompkins, W., 2004. Mechanism of feline immunodeficiency virus envelope glycoprotein mediated fusion. Virology 321, 274 – 286. Golding, H., Shearer, G.M., Hillman, K., Lucas, P., Manischewitz, J., Zajac, R.A., Clerici, M., Gress, R.E., Boswell, R.N., Golding, B., 1989. Common epitope in human immunodeficiency virus (HIV) I-GP41 and HLA class II elicits immunosuppressive autoantibodies capable of contributing to immune dysfunction in HIV I-infected individuals. J. Clin. Invest. 83, 1430 – 1435. Gougeon, M.L., Montagnier, L., 1993. Apoptosis in AIDS. Science 260, 1269 – 1270. Guiot, A.L., Rigal, D., Chappuis, G., 1997. Spontaneous programmed cell death (PCD) process of lymphocytes of FIV-infected cats: pharmacological modulation in vitro. Int. J. Immunopharmacol. 19, 167 – 179. Hoffmann-Fezer, G., Thum, J., Ackley, C., Herbold, M., Mysliwietz, J., Thefeld, S., Hartmann, K., Kraft, W., 1992. Decline in CD4+ cell numbers in cats with naturally acquired feline immunodeficiency virus infection. J. Virol. 66, 1484 – 1488. Hosie, M.J., Willett, B.J., Dunsford, T.H., Jarrett, O., Neil, J.C., 1993. A monoclonal antibody which blocks infection with feline immunodeficiency virus identifies a possible non-CD4 receptor. J. Virol. 67, 1667 – 1671. Johnson, C.M., Benson, N.A., Papadi, G.P., 1996. Apoptosis and CD4+ lymphocyte depletion following feline immunodeficiency virus infection of a T-lymphocyte cell line. Vet. Pathol. 33, 195 – 203. Johnston, J.B., Silva, C., Power, C., 2002. Envelope gene-mediated neurovirulence in feline immunodeficiency virus infection: induction of matrix metalloproteinases and neuronal injury. Virology 76, 2622 – 2633. Kolesnitchenko, V., Wahl, L.M., Tian, H., 1995. Human immunodeficiency virus 1 Envelope-initiated G2-phase programmed cell death. Proc. Natl. Acad. Sci. U.S.A. 92, 11889 – 11893. Koot, M., Keet, I.P., Vos, A.H., de Goede, R.E., Roos, M.T., Coutinho, R.A., Miedema, F., Schellekens, P.T., Tersmette, M., 1993. Prognostic value of HIV-1 syncytium-inducing phenotype for rate of CD4+ cell depletion and progression to AIDS. Ann. Intern. Med. 118, 681 – 688. LaBonte, J.A., Madani, N., Sodroski, J., 2003. Cytolysis by CCR5-using human immunodeficiency virus type 1 envelope glycoproteins is dependent on membrane fusion and can be inhibited by high levels of CD4 expression. J. Virol. 77, 6645 – 6659. Laurent-Crawford, A.G., Coccia, E., Krust, B., Hovanessian, A.G., 1995. Membrane-expressed HIV Envelope glycoprotein heterodimer is a powerful inducer of cell death in uninfected CD4+ target cells. Res. Virol. 146, 5 – 17. Lenardo, M.J., Angleman, S.B., Bounkeua, V., Bolton, D.L., 2002. Cytopathic killing of peripheral blood CD4(+) T lymphocytes by human immunodeficiency virus type 1 appears necrotic rather than apoptotic and does not require Env. J. Virol. 76, 5082 – 5093. Medinas, R.J., Lambert, D.M., Tompkins, W.A., 2002. C-Terminal gp40 peptide analogs inhibit feline immunodeficiency virus: cell fusion and virus spread. J. Virol. 76, 9079 – 9086. Meyaar, L., Otto, S.A., Jonker, R.R., Mijnster, M.J., Keet, R.P., Miedema, F., 1992. Programmed death of T cells in HIV-1 infection. Science 257, 217 – 219.

436

H. Garg et al. / Virology 330 (2004) 424–436

Mizuno, T., Goto, Y., Baba, K., Masuda, K., Ohno, K., Tsujimoto, H., 2001. TNF-alpha-induced cell death in feline immunodeficiency virusinfected cells is mediated by the caspase cascade. Virology 287, 446 – 455. Momoi, Y., Mizuno, T., Nishimura, Y., Endo, Y., Ohno, K., Watari, T., Goitsuka, R., Tsujimoto, H., Hasegawa, A., 1996. Detection of apoptosis induced in peripheral blood lymphocytes from cats infected with feline immunodeficiency virus. Arch. Virol. 141, 1651 – 1659. Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55 – 63. Muro-Cacho, C.A., Pantaleo, G., Fauci, A.S., 1995. Analysis of apoptosis in lymph nodes of HIV-infected persons. Intensity of apoptosis correlates with the general state of activation of the lymphoid tissue and not with stage of disease or viral burden. J. Immunol. 154, 5555 – 5566. Nardelli, B., Gonzalez, C.J., Schechter, M., Valentine, F.T., 1995. CD4+ blood lymphocytes are rapidly killed in vitro by contact with autologous human immunodeficiency virus-infected cells. Proc. Natl. Acad. Sci. U.S.A. 92, 7312 – 7316. Novotney, C., English, R.V., Housman, J., Davidson, M.G., Nasisse, M.P., Jeng, C.R., Davis, W.C., Tompkins, M.B., 1990. Lymphocyte population changes in cats naturally infected with feline immunodeficiency virus. AIDS 4, 1213 – 1218. Penn, M.L., Grivel, J.C., Schramm, B., Goldsmith, M.A., Margolis, L., 1999. CXCR4 utilization is sufficient to trigger CD4+ T cell depletion in HIV-1-infected human lymphoid tissue. Proc. Natl. Acad. Sci. U.S.A. 96, 663 – 668. Purvis, S.F., Jacobberger, J.W., Sramkoski, R.M., Patki, A.H., Lederman, M.M., 1995. HIV type 1 Tat protein induces apoptosis and death in Jurkat cells. AIDS Res. Hum. Retroviruses 11, 443 – 450. Richman, D.D., Bozzette, S.A., 1994. The impact of the syncytiuminducing phenotype of human immunodeficiency virus on disease progression. J. Infect. Dis. 169, 968 – 974. Roggero, R., Robert-Hebmann, V., Harrington, S., Roland, J., Vergne, L., Jaleco, S., Devaux, C., Biard-Piechaczyk, M., 2001. Binding of human immunodeficiency virus type 1 gp120 to CXCR4 induces mitochondrial transmembrane depolarization and cytochrome c-mediated apoptosis independently of Fas signaling. J. Virol. 75, 7637 – 7650.

Roumier, T., Castedo, M., Perfettini, J.L., Andreau, K., Metivier, D., Zamzami, N., Kroemer, G., 2003. Mitochondrion-dependent caspase activation by the HIV-1 envelope. Biochem. Pharmacol. 66, 1321 – 1329. Salzwedel, K., West, J.T., Hunter, E., 1999. A conserved tryptophan-rich motif in the membrane-proximal region of the human immunodeficiency virus type 1 gp41 ectodomain is important for Env-mediated fusion and virus infectivity. J. Virol. 73, 2469 – 2480. Scheller, C., Jassoy, C., 2001. Syncytium formation amplifies apoptotic signals: a new view on apoptosis in HIV infection in vitro. Virology 282, 48 – 55. Schwartz, O., Alizon, M., Heard, J.M., Danos, O., 1994. Impairment of T cell receptor-dependent stimulation in CD4+ lymphocytes after contact with membrane-bound HIV-1 Envelope glycoprotein. Virology 198, 360 – 365. Shimojima, M., Miyazawa, T., Ikeda, Y., McMonagle, E.L., Haining, H., Akashi, H., Takeuchi, Y., Hosie, M.J., Willett, B.J., 2004. Use of CD134 as a primary receptor by the feline immunodeficiency virus. Science 303, 1192 – 1195. Suarez, T., Nir, S., Goni, F.M., Saez-Cirion, A., Nieva, J.L., 2000. The pretransmembrane region of the human immunodeficiency virus type-1 glycoprotein: a novel fusogenic sequence. FEBS Lett. 477, 145 – 149. Tompkins, M.B., Bull, M.E., Dow, J.L., Ball, J.M., Collisson, E.W., Winslow, B.J., Phadke, A.P., Vahlenkamp, T.W., Tompkins, W.A., 2003. Feline immunodeficiency virus infection is characterized by B7 + CTLA4 + T cell apoptosis. J. Infect. Dis. 185, 1077 – 1093. Vlahakis, S.R., Algeciras-Schimnich, A., Bou, G., Heppelmann, C.J., Villasis-Keever, A., Collman, R.C., Paya, C.V., 2001. Chemokinereceptor activation by Env determines the mechanism of death in HIVinfected and uninfected T lymphocytes. J. Clin. Invest. 107, 207 – 215. Willett, B.J., Picard, L., Hosie, M.J., Turner, J.D., Adema, K., Clapham, P.R., 1997. Shared usage of the chemokine receptor CXCR4 by the feline and human immunodeficiency viruses. J. Virol. 71, 6407 – 6415. Wyatt, R., Sodroski, J., 1998. The HIV-1 Envelope glycoproteins: fusogens, antigens, and immunogens. Science 280, 1884 – 1888. Yu, X., McLane, M.F., Ratner, L., O’Brien, W., Collman, R., Essex, M., Lee, T.H., 1994. Killing of primary CD4+ T cells by non-syncytiuminducing macrophage-tropic human immunodeficiency virus type 1. Proc. Natl. Acad. Sci. U.S.A. 91, 10237 – 10241.