Stimulation of P2X receptors enhances lipooligosaccharide-mediated ...

Stimulation of P2X receptors enhances lipooligosaccharidemediated apoptosis of endothelial cells Matt J. Sylte,* Chris J. Kuckleburg,* Thomas J. Inzana,† Paul J. Bertics,‡ and Charles J. Czuprynski*,1 *Department of Pathobiological Sciences, School of Veterinary Medicine, and ‡Department of Biomolecular Chemistry, School of Medicine, University of Wisconsin, Madison; and †Center for Molecular Medicine and Infectious Diseases, Virginia-Maryland Regional College of Veterinary Medicine, Virginia Polytechnic Institute and State University, Blacksburg

Abstract: Exposure of endothelial cells to lipid A-containing molecules, such as lipopolysaccharide (LPS) or lipooligosaccharide (LOS), causes the release of purinergic compounds [e.g., adenosine 5ⴕ-triphosphate (ATP)] and can lead to apoptosis. The P2X family of purinergic receptors (e.g., P2X7) has been reported to modulate LPS signaling events and to participate in apoptosis. We investigated the role that P2X receptors play in the apoptosis that follows exposure of bovine endothelial cells to Haemophilus somnus LOS. Addition of P2X inhibitors, such as periodate-oxidized ATP (oATP) or pyridoxal-phosphate-6-azophenyl-2ⴕ,4ⴕdisulfonic acid tetrasodium, significantly reduced LOS-induced apoptosis. Incubation of endothelial cells with apyrase, which degrades ATP, diminished LOS-induced apoptosis of endothelial cells. Concomitant addition of P2X agonists [e.g., 2ⴕ,3ⴕ(4-benzoyl)-benzoyl ATP or ATP] to LOS-treated endothelial cells significantly enhanced caspase-3 activation. The P2X antagonist oATP significantly blocked caspase-8 but not caspase-9 activation in LOS-treated endothelial cells. Together, these data indicate that stimulation of P2X receptors enhances LOS-induced apoptosis of endothelial cells, possibly as a result of endogenous release of ATP, which results in caspase-8 activation. J. Leukoc. Biol. 77: 958 –965; 2005. Key Words: purinergic receptor 䡠 ATP

INTRODUCTION During gram-negative sepsis, endothelial cells are exposed to the lipid A-containing glycolipids lipopolysaccharide (LPS) or lipooligosaccharide (LOS), which induce endothelial cell activation and apoptosis [1]. Endothelial cells carefully orchestrate the inflammatory response to LPS, which includes the local production of cytokines, reactive oxygen and nitrogen intermediates, arachidonic acid metabolites, and purinergic compounds [2, 3]. Local production and release of these molecules may affect endothelial cell homeostasis and survival. We have previously reported that the LOS of Haemophilus somnus, a 958

Journal of Leukocyte Biology Volume 77, June 2005

gram-negative pathogen of cattle, which frequently causes vasculitis and thrombosis, induces apoptosis of endothelial cells [4, 5]. Addition of LPS to endothelial cells is known to cause autocrine or paracrine release of purinergic compounds such as adenosine 5⬘-triphosphate (ATP) [6 – 8], which in turn, has been reported to cause apoptosis in macrophages and other cell types, in part as a result of activation of the purinergic receptor P2X7, formerly known as P2Z [9]. The contribution of endothelial-derived purinergic compounds to LOS-induced apoptosis is unknown. Purinoreceptors bind extracellular nucleotides and nucleosides and are divided into three subfamilies based on their substrate affinity: adenosine (P1), P2Y, and P2X receptors [10 –12]. The P2X purinoreceptor family binds extracellular adenosine di- and trinucleotides and is comprised of seven members (P2X1–7) [13]. Ligands that activate P2X receptors, such as ATP or 2⬘,3⬘-(4-benzoyl)-benzoyl ATP (BzATP), are considered more potent ligands of P2X7 than adenosine diphosphate (ADP) or adenosine [9, 14]. However, ATP or BzATP are not specific agonists of P2X7, as they can activate other purinergic receptors (e.g., P2X1–7 and P2Y11) [15, 16]. In contrast to other P2X receptors, P2X7 possesses several unique biological properties. For example, P2X7 has a longer Cterminal domain, which contains protein–protein interaction domains, and an LPS-binding motif [17] and requires a greater concentration of ATP for activation than other P2X receptors [14]. Expression of P2X7 has been detected in several cell types including macrophages, dendritic cells (DC), endothelial cells, and various cells in the central nervous system [18 –22]. Stimulation of P2X7 is also distinct from other P2X receptors in its involvement in apoptotic cell death. For example, Pizzo et al. [23] demonstrated that P2X7 was involved in ATP-induced apoptotic cell death of a lymphocyte cell line. Subsequent studies have demonstrated that P2X7 participates in ATPdependent apoptosis of several cell types, such as human macrophages and mesangial cells [24, 25] and murine DC and macrophages [19, 26]. The proapoptotic mechanisms respon-

1

Correspondence: Department of Pathobiological Sciences, School of Veterinary Medicine, 2015 Linden Drive, Madison, WI 63706. E-mail: [email protected] Received October 19, 2004; revised January 7, 2005; accepted January 25, 2005; doi: 10.1189/jlb.1004597.

0741-5400/05/0077-958 © Society for Leukocyte Biology

sible for P2X7-induced apoptosis have begun to be characterized. Ferrari et al. [21] noted that addition of ATP stimulated activation of proapoptotic caspases (e.g., caspase-1, caspase-3, and caspase-8) and apoptosis in a murine microglial cell line. Recently, Humphreys et al. [22] demonstrated that ATP-induced apoptosis in murine macrophages involved activation of members of the stress-activated protein kinase family. However, the effect of P2X7 stimulation in endothelial cell apoptosis is controversial [27, 28]. P2X7 has been shown to participate in LPS signaling events in murine macrophages. Hu et al. [29] found that treatment of macrophages with P2X7 agonists stimulated LPS-induced production of nitric oxide and activation of nuclear factor-␬B, which were blocked by selective inhibitors of P2X7. Denlinger et al. [17] and Sommer et al. [30] reported an LPS-binding motif in the C-terminal domain of P2X7, which is important for its LPS binding in vitro. Recently, several investigators have identified expression of P2X7 in endothelial cells in vivo [31, 32] and in vitro [20, 33]. Although LPS induces release of ATP from endothelial cells [6 – 8], it is unknown whether stimulation of P2X receptors is involved in lipid A-induced apoptosis of endothelial cells. In this present study, we tested the hypothesis that stimulation of P2X receptors enhances LOS-induced apoptosis of endothelial cells. We provide evidence that P2X receptors are involved in transducing the apoptotic signal from LOS in bovine endothelial cells. Blocking P2X receptors diminished LOS-induced caspase-8 activation and apoptosis. Conversely, stimulation of P2X receptors enhanced LOS-induced caspase activation and apoptosis. These data suggest that P2X receptors (e.g., P2X7 and others) participate in LOS-mediated apoptosis, perhaps by activation of caspase-8.

MATERIALS AND METHODS

LOS LOS was isolated from H. somnus strain 649 [35], as described previously [36]. Protein contamination was not detected in the LOS preparation, as analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and silver-staining. Stock solutions of LOS were prepared in endotoxin-free dH2O and stored at –20°C.

Hoechst 33342 staining Endothelial cells adherent to glass coverslips were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 30 min and then stained for 10 min with a solution of Hoechst 33342 (Molecular Probes, Eugene, OR) diluted to 10 ␮g/mL in PBS, as described previously [4, 37]. Medium and staurosporine (200 nM for 6 h)-treated endothelial cells served as negative and positive controls, respectively [4]. At least 100 endothelial cells in five random areas per coverslip were examined microscopically (400⫻ magnification) and scored for apoptosis, based on nucleic acid condensation.

Caspase-8 and caspase-9 assays Caspase-8 and caspase-9 activation was detected using colorimetric caspase assays (R&D Systems, Minneapolis, MN) following the manufacturer’s recommended protocol as described previously [38]. Briefly, 100 ␮g total protein was added to the appropriate caspase reaction buffer and substrate (200 ␮M) in a 96-well plate format. The plate was incubated at 37°C for 4 h, and A405 was analyzed using a microplate reader (EL-312, Bio-Tek Instruments, Winooski, VT). Fold-increase in caspase activation was determined by comparing the absorbance of the experimental samples with that of the media control, which was normalized to onefold activation.

Caspase-3 assay Endothelial cells (1⫻104) were cultured in a 48-well plate. After treatment, the plate was immediately stored at –70°C overnight. The plate was thawed by floating in a 37°C water bath for 5 min. Approximately 100 ␮l ApoONE™ homologous caspase-3 reagent (Promega), containing Asp-Glu-Val-Asp-rhodamine-110 substrate, was added to each well. After incubating for 4 h in the dark at 22°C, the plate was analyzed for rhodamine-110 fluorescence (excitation of 488 nm and emission at 525 nm) using a fluorescent plate reader (Millipore, Bedford, MA). The fold-change in caspase-3 activity was normalized by comparing the fluorescence of the experimental samples with that of the media control.

ERK1/2 immunoblots

Chemicals and media Staurosporine, paraformaldehyde, amphotericin B, gentamicin sulfate, dimethyl sulfoxide (DMSO), penicillin, streptomycin, ADP, ATP, BzATP, periodate-oxidized ATP (oATP), uridine triphosphate (UTP), pyridoxal-phosphate6-azophenyl-2⬘,4⬘-disulfonic acid tetrasodium (PPADS), phorbol 12-myristate13-acetate (PMA), and Dulbecco’s modified Eagle’s medium (DMEM; containing phenol red, 25 mM HEPES, 4.5 g/L dextrose, and 2 mM Lglutamine) were obtained from Sigma Chemical Co. (St. Louis, MO). The rabbit polyclonal anti-ACTIVETM mitogen-activated protein kinase [MAPK; dualphosphorylated extracellular signal-regulated kinase 1/2 (pERK1/2)] and total ERK1/2 antibodies were obtained from Promega (Madison, WI). Horseradish peroxidase (HRP)-conjugated goat anti-rabbit immunoglobulin G (IgG) was purchased from Santa Cruz Biotechnology (CA).

Isolation and culture of endothelial cells Primary bovine pulmonary artery endothelial cells were isolated by luminal scraping of dissected vessels from several adult Hereford cattle, as described previously [4, 34] and were cultured in DMEM tissue-culture medium supplemented with 20% fetal bovine serum (FBS), penicillin (100 U/mL), and streptomycin (100 ␮g/mL) and amphotericin B (2.5 ␮g/mL) at 37°C with 5% CO2. Isolated endothelial cells were characterized by their “cobblestone” morphology and expression of von Willebrand-related factor on their surface, as determined by fluorescence microscopy [34]. For experiments, endothelial cells were used from passage 4 to 12, and frozen stocks were stored in a liquid nitrogen tank with 20% FBS and 5% DMSO added as cryopreservants.

Endothelial cells (1⫻106) were seeded into 100 mm dishes and incubated overnight in DMEM supplemented with 20% FBS. Five hours prior to treatment, endothelial cells were cultured in DMEM supplemented with 1% FBS. The endothelial cells were then treated for 2 h, with or without oATP (300 ␮M) [39, 40] or PPADS (500 ␮M) [39, 41, 42] and were incubated for an additional 15 min with LOS (500 ng/mL), ATP (5 mM), BzATP (500 ␮M), or PMA (100 nM). Total cell lysates were prepared using the MPER™ cell lysis buffer (Pierce, Rockford, IL), using the manufacturer’s protocol. Protein concentrations of lysates were determined by the bicinchoninic acid method (Pierce). Total protein (40 ␮g) per sample was separated using SDS-PAGE (4 –20% gradient, Bio-Rad, Hercules, CA) and was transferred to nitrocellulose membrane. Expression of pERK1/2 was detected by immunoblotting with polyclonal antibodies specific for pERK1/2, using a 1:2000 dilution in Trisbuffered saline/Tween-20 and 5% dried milk, and was incubated overnight at 4°C. Immunoreactive proteins were visualized by adding HRP-conjugated goat-anti rabbit IgG (1:10,000 dilution) and SuperSignal™ Pico Western chemiluminescent substrate (Pierce) and then exposing the membrane to X-ray film (Eastman Kodak, Rochester, NY). The nitrocellulose membrane was stripped using a commercial Western blot stripping buffer (Restore™, Pierce), blocked with 5% dried milk for 1 h at 22°C and probed again with rabbit polyclonal anti-total ERK1/2 antibodies (1:5000) at 4°C overnight. Immunoreactive bands against total ERK1/2 were determined, as described above.

Treatment of endothelial cells with purinergic agonists or antagonists Endothelial cells (1⫻106 for caspase-8 and -9 assays, 1⫻104 for caspase-3 assay, and 5⫻104 for Hoechst 33342) were incubated with oATP (300 ␮M) or

Sylte et al. Purinergic stimulation enhances LOS-mediated apoptosis

959

PPADS (500 ␮M) for 2 h, washed three times with DMEM and 20% FBS, and then incubated with 500 ng/mL LOS for 6 h. In other experiments, endothelial cells were incubated concomitantly with BzATP (500 ␮M), ATP (0.5–5 mM), or ADP (100 –1000 ␮M) and LOS, with or without pretreatment of oATP or PPADS. Apoptosis was detected using caspase-3, -8, or -9 assays and Hoechst 33342 staining.

Degradation of extracellular ATP Endothelial cells (5⫻104) were incubated for 6 h in DMEM with 20% FBS with 5 units apyrase (grade VI) [43], with or without 500 ng/mL of LOS. Apoptosis was detected using Hoechst 33342 staining.

Detection of released ATP Endothelial cells (5⫻105) were incubated with or without LOS (500 ng/mL). At specific time-points (0 –30 min), conditioned medium was assayed for release of ATP by incubating with luciferin-luciferase reagent (Chrono-Lume, ChronoLog, Havertown, PA) in a Chrono-Log Model 560-Ca dual sample Lumi-ionized calcium aggregometer. The amount ATP released (nM) from LOS or untreated endothelial cells was quantified by comparison with an ATP standard curve using AGGRO/LINK software (Chrono-Log) and subtracting background ATP from control tissue-culture medium.

Statistical analysis Data were analyzed for statistical significance using a one-way repeated measure ANOVA, followed by the Tukey Kramer multiple comparison posttest, as performed by the Instat software package (GraphPad, San Diego, CA). Data are presented as the mean ⫾ SEM of at least three independent experiments. Statistical significance was set at P ⬍ 0.05.

RESULTS P2X antagonists block LOS-induced ERK1/2 activation in endothelial cells We first sought to determine whether bovine pulmonary artery endothelial cells could respond to P2X agonists or H. somnus LOS by detecting formation of pERK1/2, as described previously [29]. Addition of LOS or the P2X agonists ATP or BzATP induced the formation of pERK1/2 (Fig. 1A). As a control, addition of PMA induced pERK1/2 formation, independent of P2X receptors. Conversely, the selective P2X antagonists oATP or PPADS inhibited BzATP and LOS-induced formation of pERK1/2 (Fig. 1B) but failed to prevent pERK1/2 formation by UTP, a P2Y2 agonist (data not shown). These results suggest that P2X agonist or LOS-induced activation of ERK1/2 was dependent on P2X but not P2Y2 receptors in endothelial cells.

Inhibition of P2X diminishes LOS-induced apoptosis As LOS-induced pERK1/2 formation was inhibited by addition of P2X antagonists, and stimulation of P2X7 is involved in apoptosis [19], we examined whether P2X was involved in LOS-mediated apoptosis of endothelial cells. Inhibition of P2X receptors by oATP or PPADS significantly reduced (P⬍0.01) LOS-induced chromatin condensation, as compared with endothelial cells treated with LOS alone (Fig. 2A). Likewise, addition of oATP significantly reduced LOS-induced caspase-3 activation (P⬍0.01; Fig. 2B). However, inhibition of P2X had no effect on staurosporine-induced apoptosis (Fig. 2A). Overall, these data suggest that P2X receptors are involved in LOS-induced apoptosis of endothelial cells. 960


Fig. 1. P2X-mediated activation of ERK1/2 is inhibited by oATP and PPADS. (A) Endothelial cells (1⫻106) were incubated in DMEM with 0.5% FBS for 4 h and then treated with medium (Control), BzATP (500 ␮M), ATP (5 mM), PMA (100 nM), or 500 ng/mL LOS for 15 min (upper panel). (B) Endothelial cells were incubated with oATP (300 ␮M) or PPADS (500 ␮M) for 2 h before the addition of BzATP or LOS for 15 min (lower panel). Cell lysates were prepared and analyzed for pERK1/2 formation by immunoblotting, using an antibody specific for the threonine/tyrosine-pERK1/2. Levels of total ERK1/2 (the multiple bands represent nonphosphorylated, monophosphorylated, or dually phosphorylated forms) were determined by immunoblotting of the same samples with total ERK1/2 polyclonal antibodies. These data are from one representative experiment of three independent experiments that were performed.

Endothelial cell release of ATP contributes to LOS-induced apoptosis The previous experiments suggested a role for P2X receptors in LOS-mediated apoptosis of endothelial cells. This could reflect, in part, release of ATP from LOS-treated endothelial cells, which in turn, reacts with P2X or P2Y receptors. We, therefore, tested whether LOS induced the release of ATP from endothelial cells. Addition of LOS induced significant release of ATP within 15 min (P⬍0.05), which returned to baseline levels at 30 min (Fig. 3A). To test the possibility that endogenous ATP is involved in LOS-mediated apoptosis, apyrase, which selectively degrades ATP to adenosine 5⬘-phosphate and pyrophosphate, was used. Apyrase completely degraded 270 nM ATP within 1 s (data not shown), demonstrating its potent and rapid activity against extracellular ATP. Addition of 5 units apyrase significantly diminished LOS-induced chromatin condensation (P⬍0.01; Fig. 3B). These data suggest that LOS induces release of ATP, which in turn, could bind to several P2X receptors to enhance apoptosis of endothelial cells.

P2X stimulation enhances LOS-induced apoptosis We next examined whether addition of P2X agonists would amplify LOS-induced apoptosis. Concomitant addition of ATP (1–5 mM) or BzATP (500 ␮M) to LOS-treated endothelial cells significantly increased (P⬍ 0.001) caspase-3 activation, as compared with endothelial cells incubated with LOS alone (Fig. 4). Treatment of endothelial cells with ATP or BzATP http://www.jleukbio.org

Fig. 2. Inhibition of P2X receptors diminishes LOS-mediated apoptosis. Endothelial cells (5⫻104) were incubated with LOS (500 ng/mL) for 6 h. Some endothelial cells were also treated with oATP (300 ␮M) or PPADS (500 ␮M) for 2 h prior to adding LOS. Controls consisted of endothelial cells incubated in medium (control) or medium to which oATP, PPADS, or 200 nM staurosporine (Stauro) were added. The latter served as a positive control. Apoptosis was evaluated using Hoechst 33342 staining to detect chromatin condensation (A) or by a caspase-3 assay (B). These data illustrate the mean ⫾ SEM of three separate experiments. Pretreatment of endothelial cells with oATP or PPADS significantly reduced LOS-mediated chromatin condensation and oATP-reduced caspase-3 activation, as compared with endothelial cells incubated with LOS alone (*, P⬍ 0.01).

alone had no significant effect on caspase-3 activation. Conversely, pretreatment of endothelial cells with oATP significantly reduced LOS and BzATP-induced caspase-3 activation (P⬍0.001; Fig. 4). Addition of ADP (0 –1000 ␮M), a P2Y1 agonist, had no effect on caspase-3 activation (data not shown). These data support the working hypothesis that stimulation of P2X receptors enhances LOS-induced apoptosis.

P2X antagonists significantly reduce LOSinduced caspase-8 activation We have previously reported that caspase-3 activation and apoptosis in LOS-treated endothelial cells are primarily dependent on caspase-8 activation [5]. Therefore, we sought to determine if inhibition of P2X receptors would affect caspase-8 and caspase-9 activation. Addition of the P2X antagonist oATP

significantly reduced (P⬍0.01) LOS-induced caspase-8 but had no effect on caspase-9 activation (Fig. 5).

DISCUSSION Although members of the P2X purinergic receptor family are known to be expressed in endothelial cells, their role in endothelial cell biology and vascular physiology is not well characterized. Expression of P2X purinergic receptors by normal or damaged endothelial cells from various mammalian species at different anatomical locations has been reported. For example, healthy rat thymic endothelial cells expressed predominately P2X3 [32]. In contrast, healthy rabbit aortic endothelial cells expressed all seven members of the P2X family but expressed


961

Fig. 3. LOS-induced release of ATP is involved in endothelial cell apoptosis. (A) Endothelial cells (5⫻105) were incubated with or without LOS (500 ng/mL) for up to 30 min, and release of ATP was measured. Endothelial cells incubated in medium alone served as a control. The data illustrate the mean ⫾ SEM nM ATP released in three separate experiments. Addition of LOS induced significant release of ATP, as compared with control endothelial cells (*, P⬍0.01). (B) Endothelial cells (5⫻104) were incubated with 5 units Grade VI apyrase (AP) for 15 min before a 6-h incubation with LOS (500 ng/mL). Controls consisted of endothelial cells incubated in medium alone, medium with 5 units AP, or medium with 200 nM staurosporine (Stauro) as a positive control. Apoptosis was determined by Hoechst 33342 staining to detect chromatin condensation. The data illustrate the mean ⫾ SEM percent chromatin condensation of three separate experiments. Addition of AP significantly reduced LOS-mediated chromatin condensation, as compared with endothelial cells incubated with LOS alone (*, P⬍0.05).

significantly more P2X4 following experimental injury [31]. Endothelial cells from human arterial explants and human umbilical vein and aortic endothelial cells express P2X7 mRNA and protein [33]. Likewise, bovine aortic endothelial cells express P2X4, P2X5, and P2X7 mRNA and protein [32], and expression of P2X7 protein appears to be located exclusively on the cell surface [20]. In this study, we demonstrate that bovine pulmonary artery endothelial cells responded to the addition of P2X agonists by formation of pERK1/2. Addition of ATP, BzATP, or LOS induced formation of pERK1/2 in bovine pulmonary artery endothelial cells, which was inhibited by the P2X antagonists oATP or PPADS. These findings are similar to those of Hu et al. [29], who reported that oATP and PPADS blocked BzATP or LPS-induced formation of pERK1/2 in murine macrophages. As addition of LOS activated ERK1/2, we tested whether ERK1/2 activation affected LOS-mediated apoptosis of endothelial cells by addition of U0126 (10 ␮M), a selective inhibitor of MAPK kinase 1/2 [44]. Addition of U0126 significantly reduced LOS-mediated caspase-3 activation (82⫾8% less caspase-3 activity than cells treated with LOS alone; data not shown), which suggests that the downstream events of LOSmediated pERK1/2 participate in enhancing apoptosis. oATP 962


inhibits P2X7 function by irreversibly binding to its ATPbinding site [39]. Although oATP is selective for P2X7, it also has been shown to inhibit P2X1–2 [15] and affect P2Y1 function, ecto-ATPase activity, and cytokine production [45]. It is possible that oATP may diminish LOS-induced apoptosis by inhibiting cytokine release or blocking pattern recognition receptors, such as Toll-like receptors (TLRs) [46]. Likewise, PPADS may affect other P2X receptors, including P2X1–3, P2X5, and P2X7, as well as some P2Y receptors [14, 39, 42, 47, 48]. Our findings also indicate that addition of the P2X agonists ATP or BzATP enhanced LOS-induced caspase-3 activation. In interpreting these data, it should be noted that ATP and BzATP bind several purinergic receptors (e.g., P2X1–7 and P2Y11) [49 –51]. Endothelial cells are reported to express P2X1, P2X4, P2X5, P2X7, and P2Y11 [20, 31, 33, 52]. It is possible that one or more of these receptors are involved in enhancing LOS-induced apoptosis. It has been reported that addition of BzATP (which is more resistant to ecto-ATPases [53]) to endothelial cells results in activation of P2X4 and P2X7 but not P2X5 receptors [20]. Thus, it seems less likely that P2X5 is involved in enhancing LOS-induced apoptosis Although P2X4 was reported to be expressed at higher levels than http://www.jleukbio.org

Fig. 4. P2X agonists enhance LOS-mediated caspase-3 activation. Bovine endothelial cells (1⫻104) were incubated with LOS (500 ng/mL), with or without BzATP (500 ␮M) or ATP (0.5–5 mM) for 6 h, and a caspase-3 assay was performed. In some instances, endothelial cells were incubated with 300 ␮M oATP for 2 h prior to treatment with BzATP (500 ␮M) or BzATP and LOS. Endothelial cells incubated in medium alone (–LOS) or with LOS (⫹LOS) served as negative and positive controls, respectively. These data represent the mean ⫾ SEM fold-change in caspase-3 activation of three independent experiments. Treatment of endothelial cells with BzATP or ATP (1 and 5 mM) significantly enhanced LOS-mediated caspase-3 activation (*, P⬍0.01), and addition of oATP significantly (‡, P⬍0.001) reduced LOS and BzATPinduced caspase-3 activation.

P2X7 in bovine aortic endothelial cells [20], there are no reports linking P2X4 stimulation with apoptotic cell death, nor does P2X4 possess an LPS-binding motif. Recently, a role for other P2X or P2Y receptors in apoptotic cell death of prostate cancer cells was described [53, 54]. These results indicate that activation of ion channels and modulation of intracellular Ca2⫹ by P2X or P2Y receptors may trigger apoptotic events independent of P2X7. Endothelial cells also express several P2Y receptors (e.g., P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, and P2Y12) [31, 52, 55, 56]. However, in the present study, addition of ADP, a P2Y1 agonist, failed to enhance LOS-induced caspase-3 activation, suggesting that P2Y1 receptors are not involved in LOS-induced apoptosis of endothelial cells. Based on our results, we cannot exclude the possibility that P2Y11, which is activated by BzATP, enhances LOS-induced apoptosis [57]. As the effector function of P2X7 is linked to apoptosis, it is tempting to speculate that the addition of ATP or BzATP

enhances apoptosis as a result of stimulation of P2X7. However, the ligands and antagonists used in the present study activate or inhibit multiple P2X and P2Y receptors. Thus, we cannot exclude the possibility that other P2X or P2Y receptors enhance LOS-mediated apoptosis. We found that incubation of endothelial cells with purinergic agonists alone (e.g., BzATP or ATP) did not cause apoptosis, which is consistent with a previous report that BzATP was not directly cytotoxic to human endothelial cells [8]. These data are contrary to the findings of von Albertini et al. [28], who reported that addition of ATP or ADP induced apoptosis of endothelial cells. Likewise, Rounds et al. [27] reported that addition of ATP resulted in endothelial cell apoptosis as a result of formation of intracellular adenosine over a 24-h period. In the present study, BzATP, a slowly hydrolysable analog of ATP, which resists degradation by ecto-ATPases, was a more potent enhancer of LOS-induced apoptosis than was

Fig. 5. oATP reduces LOS-induced caspase-8 activation. Bovine endothelial cells (1⫻106) were incubated with LOS (500 ng/mL) for 6 h. Some cultures of endothelial cells were also incubated with oATP (300 ␮M) for 2 h prior to adding LOS. Endothelial cells incubated in medium alone (control) served as a negative control. Cell lysates were assayed for caspase-8 or caspae-9 activities, and these data represent the mean ⫾ SEM fold-change in caspase-8 and caspase-9 activation of three independent experiments. Treatment of endothelial cells with oATP prior to the addition of LOS significantly reduced caspase-8 activation (*, P⬍0.01), as compared with LOS-treated endothelial cells.


963

ATP [58]. The possible role of P2X7 or other P2X receptors in LOS-induced apoptosis was strengthened by observed release of ATP from LOS-treated endothelial cells. Although the amount of ATP detected was less than that used to augment apoptosis, it is possible that ecto-ATPases in the serum used in the tissue-culture medium rapidly degraded some of the ATP in the medium [59]. The effective local concentration of ATP to which the endothelial cells were exposed could be considerably greater than what we detected in the culture medium. Functional evidence for the role of endogenous ATP in endothelial cell apoptosis was provided by the observation that addition of apyrase, which efficiently degrades ATP, significantly diminished LOS-mediated apoptosis (Fig. 3). These results are consistent with previous reports that LPS induced a significant but brief release of ATP from human endothelial cells [6, 7]. Overall, our results suggest that the autocrine or paracrine release of ATP affects lipid A-induced apoptosis of endothelial cells. An alternative source of ATP in vivo could be provided by activated platelets, which we have recently reported release ATP when activated by exposure to H. somnus or its LOS in vitro [60]. Perhaps development of strategies to degrade extracellular ATP in acute gram-negative sepsis could provide some therapeutic benefit. We have previously demonstrated that H. somnus LOSmediated apoptosis of bovine endothelial cells was caspase-8dependent [5]. There are conflicting reports regarding initiator caspase activation by P2X agonists. Ferrari et al. [21] previously demonstrated that addition of ATP to myeloid cells induced caspase-8 activation and apoptosis. In contrast, addition of a caspase-9 inhibitor blocked BzATP-induced DNA fragmentation in primary, cervical, epithelial cells, although caspase-3, caspase-8, and caspase-9 activities were not reported in that study [59]. In the present study, addition of oATP inhibited LOS-induced activation of caspase-8 but not caspase-9. Perhaps oATP prevented interaction of caspase-8 with a putative death domain found in the C-terminal domain on P2X7 [17] or affected presentation of LOS to pattern recognition receptors such as TLR4 [46]. Although we did not determine whether ATP or BzATP alone induced caspase-8 or caspase-9 activity, this is unlikely, as their addition failed to induce significant levels of caspase-3 activation (Fig. 4). It has been reported previously that P2X7 stimulation leads to pore formation and calcium influx, which can, in turn, lead to apoptosis [61]. We were unable to detect any P2X7-induced pore activity, as measured by YO-PRO1 dye translocation in BzATP, ATP, or LOS-treated endothelial cells (data not shown). Ramirez and Kunze [20] also failed to detect BzATPinduced YO-PRO1 dye uptake or pore formation in bovine aortic endothelial cells. We infer that P2X agonists enhance LOS-mediated apoptosis by stimulation of P2X or P2Y receptors that activate initiator caspases, rather than as a consequence of pore formation. In summary, we provide evidence that P2X receptors participate in LOS-mediated apoptosis of bovine endothelial cells and that endogenously released ATP is involved in this process. Treatment of endothelial cells with P2X agonists (e.g., BzATP or ATP) enhanced LOS-mediated apoptosis. Conversely, addition of P2X.antagonists (e.g., oATP or PPADS) reduced LOS-mediated apoptosis. We also provide evidence 964


that addition of P2X agonists, perhaps through stimulation of P2X7, activates caspase-8, which, in turn, activates caspase-3 and results in LOS-induced apoptosis of endothelial cells.

ACKNOWLEDGMENTS The work was supported by funding from the University of Wisconsin School of Veterinary Medicine, the United States Department of Agriculture National Research Initiative (0035204-9212 for C. J. C., and 99-35204-7670 for T. J. I., the Wisconsin Agricultural Experiment Station, and the University-Industry Research fund from the University of Wisconsin. We thank M. Howard (Virginia Polytechnic Institute and State University) for preparation of H. somnus LOS, C. J. Johnson (UW-Madison) for assistance with ERK1/2 immunoblots, and D. S. McVey (Pfizer Inc.) for critical review of the manuscript.

REFERENCES 1. Bannerman, D. D., Goldblum, S. E. (2003) Mechanisms of bacterial lipopolysaccharide-induced endothelial apoptosis. Am. J. Physiol. Lung Cell. Mol. Physiol. 284, L899 –L914. 2. van Deuren, M., Dofferhoff, A. S., van der Meer, J. W. (1992) Cytokines and the response to infection. J. Pathol. 168, 349 –356. 3. Pober, J. S., Gimbrone Jr., M. A., Lapierre, L. A., Mendrick, D. L., Fiers, W., Rothlein, R., Springer, T. A. (1986) Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J. Immunol. 137, 1893–1896. 4. Sylte, M. J., Corbeil, L. B., Inzana, T. J., Czuprynski, C. J. (2001) Haemophilus somnus induces apoptosis in bovine endothelial cells in vitro. Infect. Immun. 69, 1650 –1660. 5. Sylte, M. J., Leite, F. P., Kuckleburg, C. J., Inzana, T. J., Czuprynski, C. J. (2003) Caspase activation during Haemophilus somnus lipooligosaccharide-mediated apoptosis of bovine endothelial cells. Microb. Pathog. 35, 285–291. 6. Bodin, P., Burnstock, G. (1998) Increased release of ATP from endothelial cells during acute inflammation. Inflamm. Res. 47, 351–354. 7. Imai, M., Goepfert, C., Kaczmarek, E., Robson, S. C. (2000) CD39 modulates IL-1 release from activated endothelial cells. Biochem. Biophys. Res. Commun. 270, 272–278. 8. Wilson, H. L., Francis, S. E., Dower, S. K., Crossman, D. C. (2004) Secretion of intracellular IL-1 receptor antagonist (type 1) is dependent on P2X7 receptor activation. J. Immunol. 173, 1202–1208. 9. Surprenant, A., Rassendren, F., Kawashima, E., North, R. A., Buell, G. (1996) The cytolytic P2Z receptor for extracellular ATP identified as a P2X receptor (P2X7). Science 272, 735–738. 10. Fredholm, B. B., Ijzerman, A. P., Jacobson, K. A., Klotz, K. N., Linden, J. (2001) International Union of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol. Rev. 53, 527–552. 11. North, R. A. (2002) Molecular physiology of P2X receptors. Physiol. Rev. 82, 1013–1067. 12. Sak, K., Webb, T. E. (2002) A retrospective of recombinant P2Y receptor subtypes and their pharmacology. Arch. Biochem. Biophys. 397, 131–136. 13. North, R. A., Surprenant, A. (2000) Pharmacology of cloned P2X receptors. Annu. Rev. Pharmacol. Toxicol. 40, 563–580. 14. North, R. A. (2002) Molecular physiology of P2X receptors. Physiol. Rev. 82, 1013–1067. 15. Evans, R. J., Lewis, C., Buell, G., Valera, S., North, R. A., Surprenant, A. (1995) Pharmacological characterization of heterologously expressed ATP-gated cation channels (P2X purinoceptors). Mol. Pharmacol. 48, 178 –183. 16. Bianchi, B. R., Lynch, K. J., Touma, E., Niforatos, W., Burgard, E. C., Alexander, K. M., Park, H. S., Yu, H., Metzger, R., Kowaluk, E., Jarvis, M. F., van Biesen, T. (1999) Pharmacological characterization of recombinant human and rat P2X receptor subtypes. Eur. J. Pharmacol. 376, 127–138. 17. Denlinger, L. C., Fisette, P. L., Sommer, J. A., Watters, J. J., Prabhu, U., Dubyak, G. R., Proctor, R. A., Bertics, P. J. (2001) Cutting edge: the nucleotide receptor P2X7 contains multiple protein- and lipid-interaction

http://www.jleukbio.org

18. 19.

20. 21.

22.

23.

24.

25.

26. 27.

28.

29.

30.

31. 32.

33.

34. 35.

36. 37.

motifs including a potential binding site for bacterial lipopolysaccharide. J. Immunol. 167, 1871–1876. Collo, G., Neidhart, S., Kawashima, E., Kosco-Vilbois, M., North, R. A., Buell, G. (1997) Tissue distribution of the P2X7 receptor. Neuropharmacology 36, 1277–1283. Coutinho-Silva, R., Persechini, P. M., Bisaggio, R. D., Perfettini, J. L., Neto, A. C., Kanellopoulos, J. M., Motta-Ly, I., Dautry-Varsat, A., Ojcius, D. M. (1999) P2Z/P2X7 receptor-dependent apoptosis of dendritic cells. Am. J. Physiol. 276, C1139 –C1147. Ramirez, A. N., Kunze, D. L. (2002) P2X purinergic receptor channel expression and function in bovine aortic endothelium. Am. J. Physiol. Heart Circ. Physiol. 282, H2106 –H2116. Ferrari, D., Los, M., Bauer, M. K., Vandenabeele, P., Wesselborg, S., Schulze-Osthoff, K. (1999) P2Z purinoreceptor ligation induces activation of caspases with distinct roles in apoptotic and necrotic alterations of cell death. FEBS Lett. 447, 71–75. Humphreys, B. D., Rice, J., Kertesy, S. B., Dubyak, G. R. (2000) Stressactivated protein kinase/JNK activation and apoptotic induction by the macrophage P2X7 nucleotide receptor. J. Biol. Chem. 275, 26792– 26798. Pizzo, P., Murgia, M., Zambon, A., Zanovello, P., Bronte, V., Pietrobon, D., Di Virgilio, F. (1992) Role of P2Z purinergic receptors in ATPmediated killing of tumor necrosis factor (TNF)-sensitive and TNF-resistant L929 fibroblasts. J. Immunol. 149, 3372–3378. Harada, H., Chan, C. M., Loesch, A., Unwin, R., Burnstock, G. (2000) Induction of proliferation and apoptotic cell death via P2Y and P2X receptors, respectively, in rat glomerular mesangial cells. Kidney Int. 57, 949 –958. Schulze-Lohoff, E., Hugo, C., Rost, S., Arnold, S., Gruber, A., Brune, B., Sterzel, R. B. (1998) Extracellular ATP causes apoptosis and necrosis of cultured mesangial cells via P2Z/P2X7 receptors. Am. J. Physiol. 275, F962–F971. Ferrari, D., Chiozzi, P., Falzoni, S., Dal Susino, M., Collo, G., Buell, G., Di Virgilio, F. (1997) ATP-mediated cytotoxicity in microglial cells. Neuropharmacology 36, 1295–1301. Rounds, S., Yee, W. L., Dawicki, D. D., Harrington, E., Parks, N., Cutaia, M. V. (1998) Mechanism of extracellular ATP- and adenosine-induced apoptosis of cultured pulmonary artery endothelial cells. Am. J. Physiol. 275, L379 –L388. von Albertini, M., Palmetshofer, A., Kaczmarek, E., Koziak, K., Stroka, D., Grey, S. T., Stuhlmeier, K. M., Robson, S. C. (1998) Extracellular ATP and ADP activate transcription factor NF-␬ B and induce endothelial cell apoptosis. Biochem. Biophys. Res. Commun. 248, 822– 829. Hu, Y., Fisette, P. L., Denlinger, L. C., Guadarrama, A. G., Sommer, J. A., Proctor, R. A., Bertics, P. J. (1998) Purinergic receptor modulation of lipopolysaccharide signaling and inducible nitric-oxide synthase expression in RAW 264.7 macrophages. J. Biol. Chem. 273, 27170 –27175. Sommer, J. A., Fisette, P. L., Hu, Y., Denlinger, L. C., Guerra, A. N., Bertics, P. J., Proctor, R. A. (1999) Purinergic receptor modulation of LPS-stimulated signaling events and nitric oxide release in RAW 264.7 macrophages. J. Endotoxin Res. 5, 70 –74. Ray, F. R., Huang, W., Slater, M., Barden, J. A. (2002) Purinergic receptor distribution in endothelial cells in blood vessels: a basis for selection of coronary artery grafts. Atherosclerosis 162, 55– 61. Glass, R., Townsend-Nicholson, A., Burnstock, G. (2000) P2 receptors in the thymus: expression of P2X and P2Y receptors in adult rats, an immunohistochemical and in situ hybridization study. Cell Tissue Res. 300, 295–306. Schwiebert, L. M., Rice, W. C., Kudlow, B. A., Taylor, A. L., Schwiebert, E. M. (2002) Extracellular ATP signaling and P2X nucleotide receptors in monolayers of primary human vascular endothelial cells. Am. J. Physiol. Cell Physiol. 282, C289 –C301. Ryan, U. S., Clements, E., Habliston, D., Ryan, J. W. (1978) Isolation and culture of pulmonary artery endothelial cells. Tissue Cell 10, 535–554. Widders, P. R., Paisley, L. G., Gogolewski, R. P., Evermann, J. F., Smith, J. W., Corbeil, L. B. (1986) Experimental abortion and the systemic immune response to “Haemophilus somnus” in cattle. Infect. Immun. 54, 555–560. Inzana, T. J., Iritani, B., Gogolewski, R. P., Kania, S. A., Corbeil, L. B. (1988) Purification and characterization of lipooligosaccharides from four strains of “Haemophilus somnus”. Infect. Immun. 56, 2830 –2837. Fuse, T., Yoon, K. W., Kato, T., Yamada, K. (1998) Heat-induced apoptosis in human glioblastoma cell line A172. Neurosurgery 42, 843– 849.

38. Page, T. J., O’Brien, S., Jefcoate, C. R., Czuprynski, C. J. (2002) 7,12Dimethylbenz[a]anthracene induces apoptosis in murine pre-B cells through a caspase-8-dependent pathway. Mol. Pharmacol. 62, 313–319. 39. Murgia, M., Hanau, S., Pizzo, P., Rippa, M., Di Virgilio, F. (1993) Oxidized ATP. An irreversible inhibitor of the macrophage purinergic P2Z receptor. J. Biol. Chem. 268, 8199 – 8203. 40. Dell’Antonio, G., Quattrini, A., Cin, E. D., Fulgenzi, A., Ferrero, M. E. (2002) Relief of inflammatory pain in rats by local use of the selective P2X7 ATP receptor inhibitor, oxidized ATP. Arthritis Rheum. 46, 3378 – 3385. 41. Brown, C., Tanna, B., Boarder, M. R. (1995) PPADS: an antagonist at endothelial P2Y-purinoceptors but not P2U-purinoceptors. Br. J. Pharmacol. 116, 2413–2416. 42. Lambrecht, G. (1996) Design and pharmacology of selective P2-purinoceptor antagonists. J. Auton. Pharmacol. 16, 341–344. 43. Ferrari, D., Chiozzi, P., Falzoni, S., Hanau, S., Di Virgilio, F. (1997) Purinergic modulation of interleukin-1 ␤ release from microglial cells stimulated with bacterial endotoxin. J. Exp. Med. 185, 579 –582. 44. Borbiev, T., Verin, A. D., Birukova, A., Liu, F., Crow, M. T., Garcia, J. G. (2003) Role of CaM kinase II and ERK activation in thrombin-induced endothelial cell barrier dysfunction. Am. J. Physiol. Lung Cell. Mol. Physiol. 285, L43–L54. 45. Beigi, R. D., Kertesy, S. B., Aquilina, G., Dubyak, G. R. (2003) Oxidized ATP (oATP) attenuates proinflammatory signaling via P2 receptor-independent mechanisms. Br. J. Pharmacol. 140, 507–519. 46. Di Virgilio, F. (2003) Novel data point to a broader mechanism of action of oxidized ATP: the P2X7 receptor is not the only target. Br. J. Pharmacol. 140, 441– 443. 47. Jones, C. A., Chessell, I. P., Simon, J., Barnard, E. A., Miller, K. J., Michel, A. D., Humphrey, P. P. (2000) Functional characterization of the P2X(4) receptor orthologues. Br. J. Pharmacol. 129, 388 –394. 48. Lambrecht, G., Friebe, T., Grimm, U., Windscheif, U., Bungardt, E., Hildebrandt, C., Baumert, H. G., Spatz-Kumbel, G., Mutschler, E. (1992) PPADS, a novel functionally selective antagonist of P2 purinoceptormediated responses. Eur. J. Pharmacol. 217, 217–219. 49. Abbracchio, M. P., Burnstock, G. (1994) Purinoceptors: are there families of P2X and P2Y purinoceptors? Pharmacol. Ther. 64, 445– 475. 50. Communi, D., Robaye, B., Boeynaems, J. M. (1999) Pharmacological characterization of the human P2Y11 receptor. Br. J. Pharmacol. 128, 1199 –1206. 51. North, R. A., Surprenant, A. (2000) Pharmacology of cloned P2X receptors. Annu. Rev. Pharmacol. Toxicol. 40, 563–580. 52. Wang, L., Karlsson, L., Moses, S., Hultgardh-Nilsson, A., Andersson, M., Borna, C., Gudbjartsson, T., Jern, S., Erlinge, D. (2002) P2 receptor expression profiles in human vascular smooth muscle and endothelial cells. J. Cardiovasc. Pharmacol. 40, 841– 853. 53. Janssens, R., Boeynaems, J. M. (2001) Effects of extracellular nucleotides and nucleosides on prostate carcinoma cells. Br. J. Pharmacol. 132, 536 –546. 54. Tapia-Vieyra, J. V., Mas-Oliva, J. (2001) Apoptosis and cell death channels in prostate cancer. Arch. Med. Res. 32, 175–185. 55. Communi, D., Raspe, E., Pirotton, S., Boeynaems, J. M. (1995) Coexpression of P2Y and P2U receptors on aortic endothelial cells. Comparison of cell localization and signaling pathways. Circ. Res. 76, 191–198. 56. Simon, J., Filippov, A. K., Goransson, S., Wong, Y. H., Frelin, C., Michel, A. D., Brown, D. A., Barnard, E. A. (2002) Characterization and channel coupling of the P2Y(12) nucleotide receptor of brain capillary endothelial cells. J. Biol. Chem. 277, 31390 –31400. 57. Pirotton, S., Communi, D., Motte, S., Janssens, R., Boeynaems, J. M. (1996) Endothelial P2-purinoceptors: subtypes and signal transduction. J. Auton. Pharmacol. 16, 353–356. 58. Michel, A. D., Xing, M., Humphrey, P. P. (2001) Serum constituents can affect 2⬘-& 3⬘-O-(4-benzoylbenzoyl)-ATP potency at P2X(7) receptors. Br. J. Pharmacol. 132, 1501–1508. 59. Wang, Q., Wang, L., Feng, Y. H., Li, X., Zeng, R., Gorodeski, G. I. (2004) P2X7 receptor-mediated apoptosis of human cervical epithelial cells. Am. J. Physiol. Cell Physiol. 287, C1349 –C1358. 60. Kuckleburg, C. J., Sylte, M. J., Inzana, T. J., Corbeil, L. B., Darien, B. J., Czuprynski, C. J. (2005) Bovine platelets activated by Haemophilus somnus and its lipooligosaccharide induce apoptosis in bovine endothelial cells. Microb. Pathog. 38, 23–32. 61. Imai, M., Goepfert, C., Kaczmarek, E., Robson, S. C. (2000) CD39 modulates IL-1 release from activated endothelial cells. Biochem. Biophys. Res. Commun. 270, 272–278.


965