Extracellular ATP Induces Cytokine Expression and Apoptosis through P2X₇ Receptor in Murine Mast Cells

Mast cells are derived from CD34⁺ hemopoietic progenitor cells and circulate in blood and the lymphatic system before homing to tissues and acquiring their final effector characteristics under the influence of local tissue microenvironmental factors (1, 2, 3). The expansion, homing, and maturation of mast cell progenitors are influenced by several cytokines, most important of which are stem cell factor and IL-3 (1). Mast cells are recognized as the key cells of allergic inflammatory reactions, but they are also implicated in the pathogenesis of a number of chronic inflammatory diseases, in wound healing, in fibrosis, and in native immunity (2). They are described as long-living cells that keep relatively constant numbers in tissues under physiologic conditions, which depends both on the rate of production of mast cell precursors from the bone marrow and the length of survival of mature mast cells within tissues (2). Given, however, their pivotal role in the acute allergic reaction, mast cell numbers need to be tightly controlled by a balance between cell proliferation, development, and death. A link between mast cell activation during early stages of the allergic process and triggering of antiapoptotic signaling pathways has been suggested as an important factor that contributes to the extended life of mast cells (2, 3). Antiapoptotic mechanisms limit the initiation of programmed cell death, thereby contributing to the multiple pathological conditions that involve mast cell activities.

Mast cells contain cytoplasmic granules that store several preformed mediators, such as TNF-α, IL-4, histamine, heparin, serotonin, kinins, and proteases, which are released immediately upon activation of mast cell after cross-linkage of FcεRI (1, 2). Newly synthesized mediators include IL-1 through IL-8, TNF-α, IL-12, IL-13, IL-15, and IL-16, chemokines, prostaglandins and leukotrienes, and growth and angiogenesis factors, such as vascular endothelial factor and platelet-derived growth factor (1, 2, 3). Despite the well-characterized role of FcεRI in mast cell activation, a variety of other agonists can activate mast cells. These include complement, lipid mediators, neuropeptides, cytokines, chemokines, microbial products, and extracellular nucleotides (2, 3, 4, 5, 6). In particular, rat mast cells migrate toward adenine nucleotides (7), whereas extracellular ATP has been demonstrated to induce pore formation in the plasma membrane of rat peritoneal mast cells (8). Furthermore, ATP induced release of hexosaminidase, an elevation in Ca²⁺ level and protein tyrosine phosphorylation in MC/9 mast cell line (9), and stimulated release of histamine and leukotriene from murine bone marrow-derived mast cells (BMMCs)³ (10, 11, 12). In contrast, no histamine release in response to ATP or adenosine was observed in human lung mast cells, although ATP was able to modulate anti-IgE-induced release of this compound (13), thus emphasizing functional heterogeneity of mast cells from different sources.

Extracellular ATP and other nucleotides act through specific cell surface receptors and can regulate a broad range of cellular responses, such as platelet aggregation, smooth muscle contractility, neurotransmission, vascular tone, mucociliary clearance, mitogenic stimulation, or induction of cell death (reviewed in Refs. 14 and 15). ATP is known to be released in a Ca²⁺-dependent manner from storage compartments in nerve terminals, chromaffin cells, circulating platelets, and mast cells (6). Relatively large amounts of ATP and UTP are liberated from epithelial and endothelial cells, smooth muscle, glial cells, fibroblasts, and hepatocytes upon mechanical stimulation, including shear stress, hypotonic swelling, stretch, and vascular injury (14). In addition, ATP can be released from exercising muscle, ischemic cells, inflammatory cells, and necrotic/apoptotic cells (6, 14). The presence of ATP in the extracellular space leads to a robust activation of purinergic receptors (6, 14, 15). Two primary classes of purinoreceptors, P1 and P2, mediate biological effects of extracellular nucleotides. The P2 receptors are subdivided in two mechanistically distinct subclasses, the metabotropic G protein-coupled P2Y receptors and the ionotropic ligand-gated channel P2X receptors (6, 14). Each purinoreceptor is defined by its relative response to different purinergic ligands. Adenosine has been demonstrated to be a selective agonist of P1 receptors (13). UTP serves as a high potency agonist for human P2Y₂ and P2Y₄ receptors, whereas at P2Y₁ and P2Y₁₁ it is inactive (14, 15). ADP activates P2Y₁₂ and P2Y₁₃, and was reported to be equipotent or even more potent as ATP for P2Y₁, while for P2Y₁₁ ATP is more potent than ADP (14). UDP selectively activates P2Y₆ (14). The unique naturally occurring agonist of P2X receptors is ATP (14, 15, 16, 17).

P2X₇ receptor shares similar to other P2X receptors overall membrane topology of two membrane-spanning domains, a large extracellular loop, and intracellular N- and C-terminal domains, but its COOH-terminal intracellular chain is ∼200 amino acids longer (17, 18). P2X₇ has a pharmacological profile similar to the receptor previously designated as P2Z, with prominent expression in many immune cells, particularly lymphocytes, monocytes, macrophages, bone marrow, dendritic, mast, mesangial, and microglial cells (6, 14, 17), as well as on a limited number of other cell types including parotid acinar cells, testis, and fibroblasts (14). P2X₇ receptor requires millimolar levels of ATP in the presence of divalent cations to achieve activation (19, 20), which leads to the formation of a nonselective cationic channel with low affinity for ATP and increased permeability to Ca²⁺, intracellular depolarization, and equilibration of sodium and potassium gradients (21, 22). In addition, P2X₇ receptor may also induce the opening of a nonselective pore permeable to large organic molecules up to 900 Da (6, 14). Continuous activation of the receptor and the formation of a large transmembrane pore can cause perturbations in ion homeostasis and finally result in cell death (14).

A number of studies have implicated P2X₇ in mediating ATP-induced apoptosis in macrophages, mesangial, dendritic, and microglial cells (14, 22, 23, 24, 25). Zanovello et al. (26) demonstrated that extracellular ATP caused apoptosis in P815 mastocytoma cell line. However, the ability of extracellular ATP to induce apoptosis in BMMCs as well as in MC/9 and P815 mast cell lines through P2X₇ receptor has not been investigated yet. In this report, we provide experimental evidence that extracellular ATP in millimolar range induces the P2X₇-mediated apoptosis in BMMCs and MC/9 mast cells. In agreement with earlier findings (26), ATP triggers apoptosis in P815 mastocytoma cell line, and this effect is also P2X₇-dependent. Finally, in the time lag between the commitment to apoptosis and actual cell death, extracellular ATP stimulates the phosphorylation of ERK1/2, Jak2, and STAT6 in mast cells, and transitorily up-regulates expression of several proinflammatory cytokines, such as IL-4, IL-6, IL-13, and TNF-α.

ATP; 1[N,O-bis(5-isoquinolinesulphonyl)-N-methyl-l-tyrosyl]-4-phenylpiperazine (KN-62), and oxidized ATP (oATP), both antagonists of P2X₇ receptor (27, 28); 3-O-(4′-benzoyl)-benzoyl-benzoyl-ATP (BzATP), agonist of P2X₇ receptor (14); and EGTA were purchased from Sigma-Aldrich. Anti-phosphotyrosine (anti-pTyr) Abs (RC20H) were obtained from BD Transduction Laboratories. Abs against STAT6 (M-20), phospho-STAT6 (pSTAT6 (Tyr⁶⁴¹)), and P2X₇ (l-20) were purchased from Santa Cruz Biotechnology, Abs against Jak2 phosphorylated at Tyr^1007/1008 were from BioSource International, and Abs against intact poly(ADP-ribose) polymerase (PARP) and cleaved PARP (7C9) from Cell Signaling Technology. Concentration of IL-4, IL-6, IL-13, and TNF-α in cell supernatants was detected by standard ELISA procedure using DuoSet kits from R&D Systems.

BMMCs were obtained from femoral bone marrow of 6-wk-old C57BL/6 mice as previously described (29), and cultured in IMDM supplemented with 10% FCS (PAA Laboratories), 50 μM 2-ME, 2 mM l-glutamine (Sigma-Aldrich), 0.1 mM nonessential amino acids, 100 U/ml penicillin, 100 μg/ml streptomycin, 1 mM sodium pyruvate (all from Invitrogen Life Technologies) and 5 ng/ml IL-3 (PeproTech). Cells were collected after 4 wk of culture. The purity of mast cells was assessed by morphological (toluidine blue staining) and FACS analysis (98% of the cells were c-Kit- and FcεRI positive). MC/9 (American Type Culture Collection (ATCC) CRL 8306) is a murine mast cell line derived from fetal liver cells of a (B6 × A/J)F₁ mouse. MC/9 cells were cultured in DMEM supplemented with 10% FCS, 4.5 g/L glucose, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 1 mM sodium pyruvate, 1.5 g/L sodium carbonate, 116 mg/L l-arginine, 36 mg/L l-asparagine, 6 mg/L folic acid, and 0.1 mM nonessential amino acids (all components from Invitrogen Life Technologies), 10% WEHI-3 conditioned medium, and 40% conditioned medium from Con A-activated splenocytes obtained from BALB/c mice. The mastocytoma cell line P815 (ATCC TIB-64) derived from DBA/2 mice was cultured in RPMI 1640 supplemented with 10% FCS, 2 mM l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin.

Before stimulation, cells were washed twice with Dulbecco’s PBS. For each assay, 5 × 10⁶ cells/ml were stimulated with ATP (100 μM, 1 or 3 mM) or BzATP (100 μM) for different time intervals at 37°C. For inhibitory experiments, cells were pretreated with 1 μM KN-62 for 5 min or 300 μM oATP for 30 min followed by ATP stimulation.

RNA was extracted from cells using TRIzol reagent (Invitrogen Life Technologies). cDNA was synthesized from 5 μg of total RNA using random oligonucleotides and SuperScriptII kit (Invitrogen Life Technologies). Sequences of the primers used are shown in Table I. All primers were purchased from Metabion. cDNA was amplified by standard PCR procedure as described previously (30). To evaluate mRNA expression semiquantitatively, in addition to the PCR product from 35 cycles, 15 μl of the PCR product from the 25 cycles and the 30 cycles were run simultaneously. β-Actin message was used to normalize the cDNA amount to be used. A mock PCR (without cDNA) was included to exclude contamination in all experiments.

Cell pellets were lysed for 15 min on ice in 1% Nonidet P-40 cell extraction buffer: 20 mM Tris-HCl (pH 8.0), 15 mM NaCl, 2 mM EDTA, 10 mM sodium fluoride, 1 μg/ml pepstatin A, 1 μg/ml leupeptin, 10 mM PMSF, and 100 μM sodium vanadate (all reagents from Sigma-Aldrich). The detergent-insoluble material was removed by centrifugation at 13,000 rpm for 15 min at 4°C. Samples were resuspended in SDS-PAGE loading buffer, boiled for 5 min, and analyzed by 10% SDS-PAGE. The resolved proteins were transferred onto nitrocellulose (Bio-Rad). Blots were blocked for 1 h in PBS with 0.05% Tween 20 and 3% BSA (Sigma-Aldrich). After incubations with first and second Abs and washing with PBS with 0.05% Tween 20, visualization of specific proteins was conducted by an ECL method using ECL Western blotting detection reagents (Amersham Pharmacia) according to the manufacturer’s instructions.

For apoptosis induction cells were cultured at 1 × 10⁶ cells/ml in the presence or absence of ATP or BzATP. The percentage of apoptotic cells was evaluated by ApoTarget Annexin-V^FITC Apoptosis kit (BioSource International) according to the manufacturer’s protocol. Cell viability was determined by propidium iodide (PI) exclusion. Cells were analyzed by flow cytometry using FACSCalibur (BD Biosciences) and CellQuest software.

ATP-dependent increase in the plasma membrane permeability was measured using the extracellular fluorescent tracer Lucifer yellow (Molecular Probes) as described earlier (31). For Lucifer yellow uptake, cells were incubated for 15 min at 37°C in PBS containing 250 μM sulfinpyrazone and 1 mg/ml Lucifer yellow, and stimulated with 3 mM ATP. After several washings to remove the extracellular dye, cells were analyzed with an inverted fluorescence microscope (Nikon Diaphot 300) using a ×40 objective and a fluorescein filter.

Cells were preincubated with 2 μM membrane-permeable fura 2-AM (Molecular Probes) for 30 min at 37°C and Ca²⁺ influx was measured in PTI-RF-M2001 spectrophotometer (Photon Technology International) using 340/380 excitation filters and emission at 510 nm. After recording of background for 20 s, cells were stimulated with 3 mM ATP. Maximal and minimal fluorescence ratios were determined by lysing the cells in 0.05% reduced Triton X-100 and subsequent addition of EGTA (final concentration 10 mM) and concentration of Ca²⁺ was calculated using equation of Grynkiewicz et al. (32).

The enzymatic activity of caspase-3 and -8 was measured essentially as described (33). Cells were collected and lysed in a buffer containing 10 mM HEPES (pH 7.4), 142 mM KCl, 5 mM MgCl₂, 1 mM EGTA, 0.2% v/v Nonidet P-40, 1 mM DTT, and 2 mM Pefabloc. To measure caspase activity, 100 μl of caspase buffer (20 mM PIPES, 100 mM NaCl, 10 mM DTT, 1 mM EDTA, 0.1% w/v CHAPS, 10% w/v sucrose (pH 7.2)) containing 100 μM zDEVD-7-amido-4(trifluoromethyl)coumarin (AFC) or zIETD-AFC were added to cytosolic extracts (25 μg protein) and incubated at 37°C. The release of AFC was measured as emission at 505 nm upon excitation at 405 nm using a Labsystems Fluoroscan II fluorometer equipped with a thermostat plate reader.

All experiments were performed in at least three independent assays, which yielded highly comparable results. Data are summarized as mean ± SD. Statistical analysis of the results was performed by Student’s t test for unpaired samples. A p-value of <0.05 was considered statistically significant.

Extracellular nucleotides can reportedly induce a number of biologically relevant responses in mast cells, including the release of histamine, morphological changes, and chemotaxis (6, 7). To analyze the expression pattern of P2 receptors in BMMCs, MC/9, and P815 cells, a panel of mouse purinoreceptors was tested by RT-PCR using specific primers. The amplified bands were purified from gel and sequenced to prove the identity of the products (data not shown). This approach revealed that primary mast cells and mast cell lines express varying amounts of mRNA coding for several members of the P2X family (P2X₁, P2X₂, P2X₃, P2X₄, P2X₆, and P2X₇). Notably, the expression of P2X₃ was absent in MC/9 cells, whereas message for P2X₅ was detected only in P815 cells (Fig. 1,A). Interestingly, all mast cells exhibited a prominent expression of P2X₇ receptor. This fact was further corroborated by Western blotting analysis using Abs directed against P2X₇ receptor, which confirmed the presence of P2X₇ protein in the cell lysates (Fig. 1,B). Furthermore, mast cells expressed varying amounts of mRNA for P2Y₁, P2Y₄, P2Y₆, P2Y₁₂, P2Y₁₃, and P2Y₁₄. However, P2Y₄ and P2Y₁₂ were absent in MC/9 cells, whereas P2Y₂ was found only in P815 (Fig. 1 A). Thus, primary mast cells and mast cell lines express several P2 receptors subtypes, including P2X₇.

It has been repeatedly reported that signal transduction through P2X₇ receptor is associated with Ca²⁺ influx across the cellular membrane (14, 31), and may also open a nonselective pore capable of allowing uptake of low molecular mass hydrophilic solutes (up to 900 Da), such as Lucifer yellow and ethidium bromide (14). In addition, the ability of ATP to induce pore formation in the plasma membrane of rat peritoneal mast cells or mobilization of calcium in IL-dependent cultured mast cells has already been documented (8, 12). Thus, we tested whether these effects could be observed in BMMCs, MC/9, and P815 cells treated with various concentrations of ATP (0.1–3 mM). Indeed, concentrations of ATP in the range of 1–3 mM rendered the cells permeable for Lucifer yellow (Fig. 2,A) or ethidium bromide (data not shown). The fluorescent dyes exhibited diffuse distribution in the cytoplasm of ATP-treated mast cells, indicating membrane permeabilization. Phase-contrast microscopy analysis demonstrated that mast cells did not become permeable to trypan blue (molecular mass 961 Da) after ATP stimulation, showing a molecular mass limitation for the phenomenon of permeabilization (data not shown). Furthermore, we also observed an increase in intracellular Ca²⁺, although this effect was rather weak in P815 cells (Fig. 2 B). The stimulation of the cells in Ca²⁺-free medium in the presence of Ca²⁺ chelator EGTA (2 mM) abrogated ATP- or BzATP-induced Ca²⁺ uptake (data not shown), indicating that changes in the amount of intracellular Ca²⁺ are due to Ca²⁺ influx across the cellular membrane rather than release from the intracellular stores. The fact that only rather high concentrations of ATP (1–3 mM) were able to induce the cell membrane permeability and influx of calcium ions supports the involvement of P2X₇ receptor (20, 22).

To assess the role of P2X₇ receptor more rigorously, BMMCs and mast cell lines were treated with BzATP, KN-62, and oATP. BzATP is a more potent agonist for P2X₇ receptor than ATP and can induce the P2X₇-mediated responses at lower concentrations (14, 22). In contrast, an isoquinoline derivative KN-62 is widely used as a most potent and selective antagonist of both human and mouse P2X₇ receptor due to its species-specific action, because it is inactive at rat receptor (27, 34). In addition, oATP is considered as an effective antagonist of P2X₇ because it covalently and irreversibly binds to the receptor and inhibits its effects (28), although data are available indicating that this agent also blocks currents at P2X₁ and P2X₂ receptors (22). In fact, treatment of mast cells with BzATP (100 μM) mimicked ATP action (Fig. 2,A). In contrast, the pretreatment of cells with KN-62 (1 μM) for 5 min before ATP stimulation was sufficient to prevent the permeabilization of the cell membrane in BMMCs and P815 cells, and significantly attenuate this response in MC/9 cells (Fig. 2,A). Furthermore, the ATP-induced Ca²⁺ influx was significantly inhibited in the presence of 1 μM KN-62 (Fig. 2 B), which is in agreement with earlier findings (34). In addition, the pretreatment of mast cells with oATP at concentrations >300 μM for 30 min at 37°C before ATP stimulation also prevented the permeabilization of the cell membrane and Ca²⁺ uptake, further corroborating the involvement of P2X₇ (data not shown). The fact that KN-62 was not able to fully abrogate membrane permeability in MC/9 cells may reflect differences in the amount and/or kinetics of activation of functional P2X₇ receptors upon their surface. Taken together, these results show that ATP in millimolar range induces Ca²⁺ influx and permeabilization of the cell membrane in murine mast cells, whereas these effects are inhibited by KN-62 or oATP.

The ability of extracellular ATP to trigger apoptosis of mouse cell lines was documented in P815 mastocytoma and YAC lymphoid cells, and subsequently extended to macrophages and dendritic cells (22, 25, 26). However, the molecular mechanism of the ATP-induced apoptosis in P815 mastocytoma cell line is still unclear (26). Thus, we investigated the effects of ATP upon the viability of mast cells by incubating BMMCs, MC/9, and P815 cells with various amounts of extracellular ATP for different time intervals at 37°C. Then, the ATP-containing supernatant was replaced by fresh medium. After 18 h, the cells were collected and analyzed by Annexin V/PI staining and flow cytometry. In fact, ATP at millimolar concentrations (1–3 mM) induced apoptosis in BMMCs, and P815 and MC/9 cell lines, as confirmed by the presence of Annexin V/PI double-positive cells. The percentage of dead cells is summarized in Fig. 3,A, which shows that ∼45% of BMMCs died after ATP treatment. The number of dead cells was rather similar in MC/9 cell line, resulting in killing of ∼40% of cells, whereas death-inducing properties of ATP were clearly less prominent in P815 cell line where only 25–30% of cells died (Fig. 3,A). Conversely, extracellular ATP at concentrations below 1 mM (0.1–0.5 mM) had little, if any, effect upon the viability of BMMCs (Fig. 3,B). To confirm that the ATP-dependent cell death is specifically induced through P2X₇ receptor, we pretreated mast cells with KN-62 or oATP. As expected, these pharmacological agents significantly reduced apoptosis of mast cells (Fig. 3,A, and data not shown). KN-62 completely blocked the ATP-induced cell death in P815 cells, and substantially diminished the numbers of apoptotic cells in BMMCs and MC/9 cells, as compared with controls. Notably, the ability of KN-62 to inhibit the ATP-induced apoptosis in BMMCs was dose-dependent, and almost undetectable at concentration 0.1 μM (Fig. 3,C). Furthermore, the stimulation with BzATP (100 μM) induced apoptosis more potently than ATP, resulting in killing of ∼60% of BMMCs (Fig. 3,D). It should be noted that the action of BzATP was dose-dependent, and this agent was not able to induce apoptosis in BMMCs and MC/9 cells at concentrations below 10 μM (data not shown). Importantly, KN-62 at concentration 1–3 μM was able to rescue BMMCs from the BzATP-induced cell death only partially, resulting in the survival of ∼30% of the cells, whereas lower doses of this isoquinoline had little effect (Fig. 3 D). The partial antagonism of KN-62 at BzATP-activated human P2X₇ was already reported (34). Interestingly, the kinetic experiments demonstrated that stimulation with high doses of ATP induced apoptosis of mast cells already within the first minutes, and the removal of ATP did not result in cell recovery (data not shown), thereby indicating that some internal irreversible apoptotic mechanisms are activated by such treatment. The requirement of millimolar concentrations of ATP to induce cell death and the ability of KN-62 or oATP to inhibit this process strongly support the idea that apoptosis of mast cells is mediated through P2X₇ receptor.

Caspases are ubiquitously expressed cysteine proteases involved in the activation and execution of apoptosis (35). A number of studies reported that P2X₇ mediates apoptosis in various cells through caspase-dependent mechanism (36, 37, 38, 39). Stimulation of P2X₇ has been associated with activation of caspase-1, -3, and -8, and subsequent cleavage of the caspase substrates (39). Because caspases have been implicated as the principal mediators of apoptosis, we wanted to test whether these proteases are involved in the ATP-induced apoptosis in mast cells. As a first step, the cleavage of known caspase-3 substrate PARP (37), a DNA repair enzyme, was assessed. In fact, millimolar concentrations of extracellular ATP induced a characteristic cleavage of 117-kDa PARP protein into the 89-kDa fragment both in BMMCs and MC/9 cells, while pretreatment with KN-62 abrogated this process (Fig. 3,E). The cleavage of PARP was also observed in P815 cells (data not shown). Furthermore, ATP induced activation of caspase-3 and -8 in BMMCs after 1 h of treatment, increasing the basal activity of caspase-3 to 134% and caspase-8 to 162% (Fig. 3 F). Similar results were observed in MC/9 and P815 cells (data not shown). Importantly, lower doses of ATP (below 1 mM) or BzATP (below 10 μM) were not able to induce caspase activation (data not shown). Thus, ATP induces the P2X₇-mediated apoptosis of mast cells, which involves activation of caspase-3 and -8.

One of the early effects of P2X₇ receptor activation is Ca²⁺ influx across the plasma membrane and the equilibration of membrane K⁺ and Na⁺ gradients (14). Such ionic perturbations can trigger activation of a number of intracellular signaling molecules, including MAPKs (40). It has been shown that ATP can induce phosphorylation of ERK in PC12 cells and fetal astrocytes (40, 41), and of JNK in BAC1 murine macrophages (22). In addition, we have recently demonstrated that stimulation with extracellular ATP mediates activation of p56^lck, ERK, and JNK in Jurkat cells (42). Based on these findings, our next goal was to determine whether ATP is able to induce activation of intracellular signaling molecules in mast cells. For this purpose, BMMCs, MC/9, and P815 cells were pulse-stimulated with extracellular ATP for relatively brief periods ranging from 5 to 15–30 min. We considered such pulses to represent a more “physiologic” stimulus, as mast cells are likely exposed to high ATP concentrations only for short time intervals due to the ubiquitous expression of ecto-ATPases/ectonucleotidases (43) which enzymatically degrade extracellular ATP. In the experiments illustrated in Fig. 4, mast cells were stimulated with 3 mM ATP. At the indicated time points, the cells were collected, lysed, and assessed for patterns of tyrosine phosphorylation by Western blotting. Immunoprecipitation and probing with anti-pTyr Abs showed that a number of proteins were phosphorylated upon the exposure of cells to extracellular ATP. These included 10–18, 20–22, 27, 30, 33, 38, 45, 55–60, and 80–85 kDa proteins (Fig. 4, upper panel). Such phosphorylation pattern was already detectable within the first 5 min of ATP administration, reaching its maximum after 15 min, and returning to near basal level within 30 min (Fig. 4, and data not shown).

Notwithstanding, ATP stimulation induced rather weak changes in the pattern of tyrosine phosphorylation in P815 cells, as compared with untreated control cells (Fig. 4). The probing of the membranes with phosphospecific Abs allowed us to identify among the phosphorylated proteins STAT6, ERK1 (p44), ERK2 (p42), as well as weak phosphorylation of Jak2 (Fig. 4,A). Kinetic experiments confirmed that ATP induced a rapid and transient phosphorylation of these signaling molecules. In contrast, the phosphorylation state of other members of Jak (Jak1, Jak3, Tyk2) and STAT family (STAT1, STAT3, STAT5), as well as p38 kinase, JNK1 (p54), and JNK2 (p52) was not affected, remaining at the same level throughout the time course of the experiment (data not shown). For loading control, membranes were stripped and reprobed with anti-STAT5, anti-ERK, or anti-Jak2 Abs, respectively. In contrast, these effects were abolished when cells were pretreated with KN-62 (Fig. 4,A) or oATP (Fig. 4 B) before ATP stimulation, suggesting that the observed phosphorylation is mediated through P2X₇ receptor. Taken together, these results show that extracellular ATP induces activation of multiple signaling molecules via P2X₇ receptor in murine mast cells.

Mast cells are efficient producers of many key inflammatory cytokines in response to stimulation with a variety of stimuli (1, 2, 3). To understand the biological significance of the observed activation of signaling molecules after ATP stimulation, a number of cytokines (e.g., IL-1β, IL-2, IL-4, IL-6, IL-7, IL-13, IL-15, IL-18, and TNF-α) were tested by semiquantitative RT-PCR using specific primers to detect changes in the level of transcription. To this end, BMMCs or MC/9 cells were stimulated with ATP at concentrations ranging from 100 μM to 3 mM, harvested for RNA preparation, and assessed for the level of cytokine transcription. Fig. 5, A and B, show that ATP in millimolar range was a potent stimulus for the transcription of IL-4, IL-6, IL-13, and TNF-α in BMMCs. However, only messages for IL-4 and IL-13 were up-regulated in MC/9 cells (Fig. 5,C). Notably, the expression of IL-6 and TNF-α was already rather high in MC/9 cells, which might account for the inability of ATP to further enhance mRNA level of these cytokines. The increase in the transcription was most prominent after 3 h of ATP stimulation and undetectable after 12 h (data not shown), presumably due to the cell apoptosis. Furthermore, the stimulation of BMMCs with BzATP (100 μM) mimicked ATP action (3 mM), whereas the increase in the cytokine transcription was significantly inhibited by the pretreatment of cells with 1 μM KN-62 (Fig. 5 B) or oATP (data not shown), further supporting the involvement of P2X₇. Notably, KN-62 at concentrations below 0.5 μM had little, if any, effect on the transcription of these cytokines (data not shown).

Next, we wanted to test whether mast cells may release these cytokines to the culture medium in response to the stimulation with ATP or BzATP. To this end, BMMCs and MC/9 cells were incubated with these agents essentially as described above, and the cell supernatants were harvested for ELISA after 24 h. These experiments showed the ability of ATP or BzATP to up-regulate production of IL-6 and IL-13 in BMMCs and MC/9 cells, whereas KN-62 inhibited most of these effects (Fig. 5 D). Interestingly, KN-62 had almost no effect upon the BzATP-induced release of IL-13 in MC/9 cells. In contrast, no changes in the release of TNF-α and IL-4 from BMMCs and MC/9 cells were observed (data not shown). Taken together, these experiments demonstrate the ability of ATP or BzATP to up-regulate transcription and production of several proinflammatory cytokines in murine mast cells.

The present work demonstrates that murine mast cells express a functional P2X₇ receptor capable of initiating apoptosis after stimulation with millimolar concentrations of extracellular ATP. Concomitantly, we observed a rapid but transient phosphorylation of multiple intracellular proteins, such as ERK1/2, Jak2, and STAT6. Notwithstanding the development of apoptosis, an up-regulation in the expression level of several proinflammatory cytokines was detected by RT-PCR and ELISA in BMMCs and MC/9 cells, including IL-4, IL-6, IL-13, and TNF-α. The identification of P2X₇ as a receptor subtype responsible for the observed effects of ATP was supported by several lines of experimental evidence: 1) only millimolar concentrations of ATP were effective in inducing apoptosis, whereas BzATP mimicked ATP action at submillimolar doses; 2) the treatment with ATP or BzATP triggered a rapid transmembrane Ca²⁺ influx through channels permeable to small cations and also opened a nonselective pore capable of allowing an uptake of low molecular mass hydrophilic solutes; 3) the effects of ATP or BzATP on mast cells were abrogated by pretreatment with KN-62 or oATP; and 4) RT-PCR and Western blotting analysis confirmed the presence of P2X₇ mRNA and protein in mast cells. Thus, the requirement of ATP in millimolar range and the ability of submillimolar amounts of BzATP to mimic its effects, taken together with inhibitory properties of KN-62 and oATP, are typical for the pharmacological profile of P2X₇, strongly suggesting a role for this receptor as the mediator of the ATP-induced effects in murine mast cells.

Although the best characterized pathophysiologic stimulus for the activation of mast cells is provided by cross-linking of FcεR on the cell surface by allergens complexed to IgE molecules, a variety of other stimuli are capable of inducing mast cell activation, including complement, lipid mediators, neuropeptides, cytokines, chemokines, and extracellular nucleotides (2, 3). Because ATP can be released to the extracellular environment in many physiologic or pathologic conditions, it represents a good candidate to the role of nonimmune activation stimulus for mast cells. However, such stimuli may induce mast cells to respond in a restricted way, which could result in a release of a limited subset of mediators or cytokines or affect the amount and/or kinetics of the production of mast cell-specific bioactive substances (44). In our hands, short exposures of murine mast cells to millimolar concentrations of extracellular ATP were sufficient to initiate a number of the P2X₇-mediated biological responses, which finally culminated in the apoptotic cell death. The P2X₇-mediated apoptosis of mast cells involved the activation of caspases-3 and -8, and the cleavage of caspase-3 substrate PARP. The ability of P2X₇ to mediate activation of caspases is in agreement with recent findings in other cell systems (37, 38, 39). It remains to be elucidated whether the caspase-dependent apoptosis represents the main mechanism of the ATP-induced mast cell death, or other caspase-independent mechanisms may also be operative. Currently, we perform experiments to assess how broad-spectrum or specific protease inhibitors may affect the death-inducing properties of ATP in BMMCs and MC/9 cells. In addition, the irreversibility of the ATP-induced apoptosis in mast cells is intriguing and deserves a more profound investigation to define its biological significance.

The P2X₇-initiated cascade of signaling events reportedly involves a defined sequence of phenotypic changes that result in apoptosis only at least several hours after the transient exposure to ATP (22). It has been suggested that in the time period between commitment to apoptosis and actual cell death, agonistic stimulation of P2X₇ receptor also activates additional signaling pathways, which may lead to cytokine production in case of proteolytic processing and release of IL-1β from LPS-primed macrophages that precedes cell death (45), and activation of various transcription factors like NFAT or NF-κB (46), thus modulating the overall response to ATP. Accordingly, a rapid ATP-induced phosphorylation of ERK1/2, Jak2, and STAT6 was observed in murine BMMCs and MC/9 cells, which returned to nearly basal levels after 30 min, whereas the phosphorylation state of other members of Jak and STAT families, p38 kinase, and JNK was not altered (data not shown). Additional experiments are underway to determine the identity of other tyrosine-phosphorylated proteins in BMMCs and MC/9 cells upon ATP stimulation. Furthermore, we observed that ATP induced in a dose-dependent manner a significant increase in the transcription level of IL-4 and IL-13, and to a lesser extent, of IL-6 and TNF-α. Furthermore, ATP stimulation enhanced the release of IL-6 and IL-13 to the culture medium from BMMCs and MC/9 cells. Moreover, BzATP, which is a more potent agonist for P2X₇ receptor than ATP, mimicked the effects of ATP at lower concentration. Notwithstanding, only ∼40–45% of BMMCs or MC/9 cells died via apoptosis after 18 h of ATP treatment. Thus, not all mast cells are equally susceptible to ATP, and the observed phosphorylation pattern as well as the increase in the cytokine expression may predominantly take place in the apoptosis-resistant mast cells, although additional experiments are required to confirm or reject this suggestion.

Gargett and Wiley (27) first reported the fact that an isoquinoline derivative KN-62 can act as a potent antagonist of P2X₇ receptor. Although KN-62 is also a potent inhibitor of Ca²⁺/calmodulin-dependent protein kinase II (CaMKII), an inactive analog of KN-62, KN-04, also blocked the P2X₇ responses, indicating that the inhibition of P2X₇ receptor by KN-62 is not mediated by CaMKII (27). Importantly, the action of KN-62 upon P2X₇ does not involve its inhibitory properties on CaMKII in short-term studies, whereas prolonged exposures to KN-62 require cautious interpretation because of concomitant inhibition of this kinase (14, 27). Humphreys and colleagues (22, 34) have demonstrated the high selectivity of KN-62, which inhibits both human and mouse P2X₇ but is inactive at rat receptor, thus providing a very useful tool for identifying the P2X₇-mediated functional responses. However, the molecular mechanism(s) that underlie the ability of KN-62 to inhibit human and mouse P2X₇ receptor remain unclear (22, 34). It has been suggested that KN-62 elicits its inhibitory action on human P2X₇ receptor through direct binding to the amino-terminal half, which contains the large extracellular loop. Correspondingly, the introduction of the first 335 amino acids of the human receptor sequence conferred KN-62 sensitivity to rat P2X₇ receptor (34). The pretreatment of mast cells with KN-62 or another irreversible P2X₇ antagonist, oATP, inhibited the effects of extracellular ATP such as membrane permeabilization, protein phosphorylation, and cytokine expression. Moreover, these chemical agents significantly reduced the numbers of apoptotic BMMCs, MC/9, and P815 cells as compared with untreated controls. Notably, the antagonists were more effective in preventing the ATP-induced apoptosis in P815 cells, whereas KN-62 was able to antagonize the BzATP-induced effects only partially, which is in accord with previous findings (34). It is noteworthy that the amount and/or kinetics of activation of functional P2X₇ receptors upon the cell surface of BMMCs and these two mast cell lines may presumably vary, contributing to slightly different functional responses to ATP and the ability of the antagonists to inhibit them. Importantly, ATP induced transmembrane Ca²⁺ influx, which was most prominent in BMMCs and MC/9 but rather weak in P815 cells, whereas KN-62 or oATP abrogated this effect. The ability of P2X₇ receptor to trigger a long-lasting transmembrane Ca²⁺ influx is in agreement with earlier studies (14, 31).

The fact that brief exposures to millimolar concentrations of extracellular ATP result in the irreversible P2X₇ receptor-mediated apoptosis raises the issue of the relevance of this process to mast cell physiology. Mast cells play critical roles in a variety of allergic, autoimmune, and inflammatory diseases (1, 3, 6, 47). The expression of an active P2X₇ receptor capable of mediating apoptosis offers a possibility to quickly eliminate unwanted mast cells under circumstances which favor accumulation of extracellular ATP in rather high concentrations. Although it is still not clear whether such high levels of ATP could be achieved in the extracellular space, amounts of ATP in the protected compartments at the level of the cell membrane could easily reach concentrations sufficient to activate the low-affinity P2X₇ receptor (14). Recent findings implicate mast cells in a variety of neuroinflammatory diseases, especially those worsened by stress (6), thereby questioning the role of neuronal mast cell activation in the development of migraines and multiple sclerosis (6). It remains to be elucidated whether extracellular ATP can contribute through the P2X₇-mediated apoptosis to the elimination of unwanted mast cells in such pathological conditions, and how an activation of antiapoptotic signaling pathways may oppose its action.

Given that many cell types including mast cells can release ATP, this chemical agent might be able to alter function of the same or nearby cells by autocrine/paracrine mechanism. In fact, the amount of ATP released from one mast cell was shown to be sufficient to diffuse several tens of micrometers and elicit rises in intracellular Ca²⁺ in surrounding cells (6, 48). It has been suggested that the operation of an ATP-based autocrine/paracrine loop can support the P2X₇-mediated lymphoid cell growth in the absence of serum-derived growth factors (49). Many studies have documented that P2X₇ receptor/pore participates in diverse monocyte, macrophage, microglia, lymphocyte, and dendritic cells responses such as cell membrane permeabilization, cytokine release, multinucleated giant cell formation, cell proliferation or apoptosis (6, 14, 15). For example, stimulation of P2X₇ receptor by ATP induces posttranslational processing of IL-1β and IL-18 precursors in human monocytes (50, 51), altered cytokine production in mice lacking this receptor (52), stimulation of JNK activity and induction of apoptosis in murine macrophages (22), inhibition of osteoclastic resorption (53), and activation of transcription factor NF-κB (54) and NFAT (46) in microglial cell lines. Data from our laboratory suggest that ATP can trigger proliferation of Jurkat T lymphoblastoid cell line and mediate p56^lck-dependent phosphorylation of ERK and JNK, and activation of AP-1 transcription factor, simultaneously down-regulating p50/p65 NF-κB heterodimers (42). It is believed that in T and B lymphocytes, ATP acting over P2X₇ generates a pore smaller than that seen in other cell types, such as macrophages (14). In regard to mast cell function, extracellular ATP has been demonstrated to stimulate degranulation and release of histamine from these cells (10, 11), and to induce cell membrane permeabilization (9, 12). Notably, ATP was shown to trigger apoptosis in P815 mastocytoma cell line, although the mechanism of the ATP-mediated cell death remained obscure (26). Further, ATP stimulated the release of hexosaminidase from MC/9 mast cell line through a mechanism that was distinct from the activation induced by the cross-linking of Fc receptors (9), and induced the release of histamine and leukotriene from BMMCs (10, 11, 12). Several studies have reported a direct Ca²⁺-dependent histamine release and a potentiation of Ag- or ionophore A23187-induced histamine release from rat peritoneal mast cells in response to ATP stimulation (55, 56). The ability of extracellular ATP to induce degranulation of murine mast cells is in accordance with our findings (data not shown). Conversely, ATP only modulated anti-IgE-induced histamine release from human lung mast cells without having the ability to stimulate the release of this compound (13). Thus, mast cells derived from different body districts or distinct species exhibit a functional heterogeneity, which invites a cautious interpretation of mast cell responses in vitro with respect to their relevance in vivo.

ATP is found at a concentration 5–10 mM in the cytosol of most eukaryotic cell types, whereas its amount in platelets is significantly higher and can reach a concentration of ∼1 M (14). Release of ATP to extracellular space by both lytic and nonlytic mechanisms has been observed from virtually all cell types and tissues under conditions of hypoxia, ischemia, inflammation, injury, and cell necrosis or apoptosis (14, 15). However, extracellular ATP concentrations are generally maintained at extremely low levels by a sequential action of ubiquitous cell surface enzymes such as ecto-ATPases/ectonucleotidases, which rapidly degrade it to ADP, AMP, and adenosine (6, 14). Conversely, ATP can be copackaged with serotonin in platelet granules and released locally in significant amounts during platelet activation (14, 57). Thus, platelets themselves are a major source of ATP, whereas cytosolic ATP stores can also be liberated by the sudden breakage of intact cells, as might occur during the rupture of blood vessels and other tissue injury (57). These two sources suggest that significant amounts of extracellular ATP may be generated at vascular sites of thrombus formation and infection/inflammation. The key role of extracellular ATP and P2 receptors in hemostasis has already been emphasized (14). On the other hand, an emerging concept implicates mast cells as repair cells, which provide antithrombotic and/or profibrinolytic mediators to prevent thrombus formation or help to dissolve thrombotic material in the course of vascular repair process (58, 59). The fibrinolytic potential of a cell is determined by the relative abundance of either generated plasminogen activators or plasminogen activator inhibitors. Mast cells act fibrinolytic through expression of a key fibrinolytic enzyme, tissue-type plasminogen activator, without coexpression of plasminogen activator inhibitors, and supernatants as well as cell lysates from cultured mast cells are capable of lysing a fibrin clot in a manner similar to recombinant tissue-type plasminogen activator (58). The recognition of mast cell as a profibrinolytic cell raises a question of its potential pathophysiologic role in the prevention or repair of vascular thrombosis and tissue repair following fibrinogen extravasation. Mast cells can accumulate around thrombosed vessels and provide a number of important repair molecules including antithrombotic heparin, and other profibrinolytic substances. Thus, mast cell recruitment and activation in the sites of vascular injury may result in local thrombolysis and prevention of coagulation, which may interfere with the healing of a vascular rupture at the initial stages of injury. Therefore, the apoptosis of mast cells in response to rather high concentrations of extracellular ATP which may be achieved at the sites of acute vascular injury due to its release from damaged blood cells, endothelial cells, and activated platelets appears to be physiologically relevant, because it presumably prevents their function as antithrombotic and/or profibrinolytic cells at initial stages of vascular rupture.

In summary, our results demonstrate that ATP induces a number of the P2X₇-mediated cell responses in murine mast cells, including the phosphorylation of signaling molecules, transcription of cytokines, membrane permeabilization, and induction of apoptosis. The high sensitivity of mast cells to the cytotoxic effects of extracellular ATP suggests that this process might have a biological relevance in certain physiologic or pathologic conditions. The fact that extracellular ATP can play an important role in maintaining mast cell numbers through the P2X₇-mediated apoptosis may provide helpful insights to the development of new therapeutic approaches to intervene and limit excessive and/or undesired mast cell activities.

The authors have no financial conflict of interest.

We are grateful to Manuel Fohlmeister and Katrin Streeck for excellent technical assistance and Renate Bergmann for the help with ELISA.