Expression of the P2X₇ Receptor Increases the Ca²⁺ Content of the Endoplasmic Reticulum, Activates NFATc1, and Protects from Apoptosis

Abstract

The P2X₇ receptor is known for the cytotoxic activity because of its ability to cause opening of non-selective pores in the plasma membrane and activate apoptotic caspases. A key factor of P2X₇-dependent cytotoxicity is the massive intracellular Ca²⁺ increase triggered by its activation. Here we show that P2X₇ transfection increased the ability of the endoplasmic reticulum to accumulate, store, and release Ca²⁺. This caused a larger agonist-stimulated increase in cytosol and mitochondrial Ca²⁺ in P2X₇ transfectants than in mock transfected cells. P2X₇ transfectants survived and even proliferated in serum-free conditions and were resistant to apoptosis triggered by ceramide, staurosporin, or intracellular Zn²⁺ chelation. Finally, the nuclear factor of activated T cells complex 1 (NFATc1) was strongly activated in the P2X₇ transfectants. These observations support our previous finding that the P2X₇ receptor under tonic conditions of stimulation, i.e. those observed in response to basal ATP release, has an anti-apoptotic or even growth promoting rather than cytotoxic activity.

Cell responses to extracellular ATP are mediated by P2 receptors: ionotropic P2X and metabotropic P2Y (1). The P2X₇ receptor (P2X₇R)4 subtype stands out in the P2X subfamily for its ability to trigger a host of physiologic or pathologic responses: plasma membrane blebbing (2, 3), rapid release of interleukin-1β via microvesicle shedding (2, 4), cell fusion (5), lymphoid cell proliferation (6), cell death (7, 8), and bone formation/resorption (9). This receptor is characterized by low affinity for ATP and by two states of permeability (10, 11). At high micromolar ATP concentrations, P2X₇ behaves as a cation-selective channel, whereas a prolonged exposure to nearly millimolar concentrations triggers the transition to a nonselective pore that admits hydrophilic solutes of molecular mass up to 900 Da (12-14). A common feature of both conductance states is a strong elevation of free cytoplasmic calcium levels ([Ca²⁺]_i), a response critical for the biological role of this receptor.

Recent evidence from our group shows that basal levels of ATP, naturally present in the extracellular milieu, cause a tonic activation of the P2X₇R, which in turn triggers [Ca²⁺]_i increase and Ca²⁺ entry into the mitochondria. The increased intramitochondrial Ca²⁺ concentration ([Ca²⁺]_mt) then enhances mitochondrial potential, increases ATP synthesis, and promotes survival and proliferation in the absence of serum (15). Increased mitochondrial potential and ability to grow under serum-free conditions are hallmark of tumor transformation (16). Furthermore, overexpression of the P2X₇R is associated with several cancers or growth disturbances such as chronic lymphocytic leukemia (17), prostate (18) and breast cancer (19), squamous cell carcinoma (20), nonmelanoma skin cancer (21) neuroblastoma (22), pancreatic tumor (23), thyroid cancer (24), and renal cysts (25).

The endoplasmic reticulum (ER) is the main cellular Ca²⁺ store, characterized by a lumenal Ca²⁺ concentration several-fold higher than the cytoplasm (26). Ca⁺² is pumped against concentration gradient into the ER by ATPases of the SERCA family, and ER Ca²⁺ concentration ([Ca²⁺]_er) is kept constant by spontaneous leak via as yet poorly characterized channels. Thus, SERCA inhibition causes a slow emptying of the ER paralleled by increase in the [Ca²⁺]_i. ERCa²⁺ can also be transferred to mitochondria at dynamic sites of contact likely controlled by plasma membrane receptors, second messengers, or the metabolic state of the cell (27). ER Ca²⁺ can also be rapidly released upon activation of the IP₃ receptor/channel. The fast [Ca²⁺]_i transient caused by IP₃ receptor stimulation is a key signal for cell proliferation (28). The [Ca²⁺]_i rise activates multiple enzyme pathways, including calcineurin. Calcineurin in turn causes dephosphorylation of nuclear factor of activated T cells (NFAT) family members that translocate to the nucleus to activate gene transcription (29). Inhibition of calcineurin phosphatase activity by FK506, cyclosporin A, or new drugs such as VIVIT blocks NFAT translocation to the nucleus and DNA binding (30). The calcineurin pathway is so important for T cell proliferation that FK506 and cyclosporin are widely used as immunosuppressants (31, 32). In addition to immune cells, members of the NFAT family are also expressed in a wide range of other cell types where they are implicated in cell cycle progression, cell development and differentiation, angiogenesis, and tumorigenesis (33, 34). The NFAT family comprises four members, among which NFATc1 plays a key role in cell growth as shown by the phenotype of the NFATc1 KO mice, which exhibits serious B and T lymphocyte, cardiomyocyte, and osteoblast deficit (35). Moreover, NFATc1 is activated in pancreatic cancer overexpressing the c-myc oncogene (34) and is implicated in resistance to apoptosis (36) and in tumor invasiveness (37).

In a previous study we showed that HEK293 cells transfected with the P2X₇R (HEK293-P2X₇) have a higher resting mitochondrial Ca²⁺ ([Ca²⁺]_mt) and release a larger amount of Ca²⁺ from intracellular stores following stimulation of plasma membrane receptors coupled to IP₃ generation (15). In the present study we show that P2X₇ expression increases ER Ca²⁺ uptake and steady-state level, activates NFATc1, enhances cell survival, and protects from apoptosis.

EXPERIMENTAL PROCEDURES

Reagents and Antibodies—ATP was purchased from Roche Applied Science. Oxidized ATP (oATP), benzoylbenzoic ATP, carbachol, ceramide, staurosporin, tert-butylhydroquinone (tBHQ), thapsigargin, and TPEN were purchased from Sigma-Aldrich. fura 2/acetoxymethylester and coelenterazine were from Invitrogen-Molecular Probes (San Giuliano Milanese, Italy). VIVIT and rabbit polyclonal anti-P2X₇ antibodies were purchased from Calbiochem (Merck KGaA associate company, Darmstadt, Germany). The anti-SERCA2 mouse monoclonal antibody and horseradish peroxidase-conjugated goat anti mouse, were from AbCam (Cambridge, UK). Peroxidase-linked protein A was from Amersham Biosciences.

Cell Culture and Transfection—HEK293 and NIH3T3 cells were cultured in DMEM-F12 (Sigma-Aldrich) or DMEM (Sigma-Aldrich) medium, respectively, supplemented with 10% fetal calf serum (Invitrogen), penicillin (100 units/ml) (Euro Clone, Pero, Milano, Italy), and streptomycin (100 mg/ml) (Euro Clone). Stable clones transfected with the human P2X₇R were kept in the continuous presence of 0.2 mg/ml G418 sulfate (Geneticin, Calbiochem). The experiments, unless otherwise indicated, were performed in the following saline solution, also referred to as “standard saline” in the text: 125 mm NaCl, 5 mm KCl, 1 mm MgSO₄, 1 mm NaH₂PO₄, 20 mm HEPES, 5.5 mm glucose, 5 mm NaHCO₃, pH 7,4. When indicated, 1 mm CaCl₂ or 500 μm EGTA was added. The cells were transfected with the calcium phosphate method as described previously (15). To exclude possible effects of clonal selection, all of the experiments were performed with three different stable HEK293 clones expressing the P2X₇R (15). Although not shown in the figures, all conditions tested for mock transfected HEK293 cells (HEK293-mock) were also assayed for wild type HEK293 cells. NIH3T3 fibroblasts were transiently transfected with Lipo-fectamine 2000 (Invitrogen) as described by the manufacturer. Briefly plasmid and transfection reagent were resuspended in Opti-MEM medium (Invitrogen) at a ratio of 2 μl of Lipofectamine/1 μg of DNA, mixed and administered to cells at a final concentration of 1 μg of DNA/10⁵ cells. The experiments were performed 48 h after transfection. For [Ca²⁺]_mt measurements, double transfection was obtained at a 3:1 P2X₇R/mtAEQ plasmid ratio. Aequorins selectively targeted to the endoplasmic reticulum (erAEQ) or the mitochondria (mtAEQ) were previously engineered in Professor Rizzuto's laboratory (38).

Measurement of [Ca²⁺]_i—[Ca²⁺]_i measurements were performed in a thermostat-controlled and magnetically stirred PerkinElmer Life Sciences fluorimeter with the fluorescent indicator fura 2/acetoxymethylester. Briefly the cells were loaded with 4 μm fura 2/acetoxymethylester for 20 min in 1 mm Ca²⁺-supplemented standard saline in the presence of 250 μm sulfinpyrazone in standard saline, rinsed, and resuspended in the appropriate saline solution at a final concentration of 10⁶/ml. Excitation and emission wavelengths were 340/380 and 505 nm, respectively.

Measurement of [Ca²⁺]_er and [Ca²⁺]_mt—[Ca²⁺]_er was measured according to the protocol described previously (38). Cells were plated at a density of 5 × 10⁴ onto 13 mm coverslips and transfected with the erAEQ plasmid. To allow accurate [Ca²⁺]_er measurement, erAEQ must be reconstituted with coelenterazine in the absence of Ca²⁺. To this aim, the cells were incubated at 20 °C for 10 min in AEQ buffer: 135 mm NaCl, 5 mm KCl, 0.4 mm KH₂PO₄, 1 mm MgSO₄, 5,5 mm glucose, and 20 mm HEPES, pH 7.4, in the presence of 600 μm EGTA and 50 μm tBHQ, a reversible inhibitor of SERCA pumps, to deplete ER Ca²⁺ (38). At the end of this incubation, erAEQ was reconstituted with its cofactor coelenterazine (5 μm) for 45 min at room temperature. The coverslips were mounted in a perfused and thermostated chamber hosted in a custom-made, high sensitivity luminometer. To remove tBHQ prior to Ca²⁺ measurements, the cells were perfused for 2 min with AEQ buffer supplemented with 100 μm EGTA. Perfusion was then switched to 1 mm CaCl₂-containing AEQ buffer to measure kinetics of ER Ca²⁺ uptake and the ensuing steady-state level. The experiment was terminated by lysing the cells with 100 μm digitonin in a hypotonic Ca²⁺-rich solution (10 mm CaCl₂ in H₂O) to discharge the residual ER AEQ pool and allow calibration of the AEQ signal.

[Ca²⁺]_mt was measured with mtAEQ. Cells were plated at a density of 5 × 10⁴ onto 13-mm coverslips and transfected with the mtAEQ plasmid or, in the case of NIH3T3 cells, with mtAEQ plus P2X₇R. mtAEQ was reconstituted for 2 h at 37 °C in AEQ buffer supplemented with 1 mm CaCl₂ and 5 μm coelenterazine. Perfusion was performed with AEQ buffer plus 500 μm EGTA or, alternatively, 1 mm CaCl₂. After stabilization of the light signal, 500 μm carbachol was added to the perfusion solution.

The light signal from mtAEQ- or erAEQ-transfected cells was collected with a low noise photomultiplier with built-in amplifier discriminator. The output of the discriminator was captured by a Thorn-EMI photon counting board and stored in an IBM-compatible computer for further analyses. The AEQ luminescence data were calibrated off-line into [Ca²⁺] values using a computer algorithm based on the Ca²⁺ response curve of wild type and mutant AEQs, as described previously (39, 40). Rate of ER calcium uptake was calculated as first derivative of kinetics of ER Ca²⁺ uptake (see Fig. 3A). The total amount of released Ca²⁺ was measured as second derivative of [Ca²⁺]_i increase measured with fura-2. Origin software was used for both calculations (OriginLab, Northampton, MA). A mean cell radius of 10 μm was measured by confocal microscopy giving an approximate cell volume of 4.2 picoliters.

FIGURE 3.

Rate of ER calcium uptake is enhanced in HEK293-P2X₇ cells. A, HEK293-P2X₇ and HEK293-mock cells were pretreated with 50 μm tBHQ in AEQ buffer supplemented with 600 μm EGTA and 5 μm coelenterazine to discharge ER Ca²⁺ and reconstitute erAEQ. At the end of this incubation, they were placed into the luminometer chamber, rinsed to remove tBHQ, and perfused with 1 mm Ca²⁺-containing AEQ buffer. This procedure allows calculation of initial rate of ER Ca²⁺ uptake. Where required, prior to tBHQ incubation, the cells were treated with 600 μm oATP for 2 h. HEK293-P2X₇ (red trace), HEK293-mock (blue trace), HEK293-P2X₇ pretreated with oATP (orange trace), and HEK293-mock pretreated with oATP (cyan trace). B, summary of ER uptake rate (measured as μm increase/min) in HEK293-mock and in three different HEK293-P2X₇ cell clones (n = 6). ^*, p ≤ 0.05; ^**, p ≤ 0.01. C, SERCA2 and P2X₇ expression in mock transfected and stably transfected HEKhP2X₇ cell clones. Lane 1, HEK293-mock cells; lane 2, HEK293-P2X₇ clone A; lane 3, HEK293-P2X₇ clone B; lane 4, HEK293-P2X₇ clone C.

Cell Proliferation and Viability Assay—Cell proliferation of HEK293 cells was measured by direct count. To this purpose, 10⁵/ml cells were seeded in six-well Falcon plates (BD Biosciences, Lincoln Park, NJ) in serum-free DMEM-F12 medium in the presence or absence of various stimuli and placed in a CO₂ incubator at 37 °C. At the appropriate time cells were detached and counted in a Burker chamber with an Olympus IMT-2 (Olympus Corporation, Tokio, Japan) microscope equipped with a 40× phase contrast objective. Proliferation of NIH3T3 cells was assayed with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide test 48 h after transfection with P2X₇-containing or empty plasmid. Briefly, 5 × 10⁴ cells were incubated for 3 h in 1 mm Ca²⁺-supplemented standard saline in the presence of 0.5 mg/ml of dimethyltiazol diphenyltetrazolium bromide (Sigma-Aldrich), rinsed, and further incubated for 15 min in 90% isopropanol plus 10% DMSO. Absorbance was measured at 540 nm in a VICTOR³ spectrophotometer (PerkinElmer Life Sciences). Cell viability was also assessed with the trypan blue assay. The cells were washed and incubated in phosphate-buffered saline, 0.4% trypan blue stain solution for 5 min at room temperature and then counted. Apoptosis was evaluated with the cell death ELISA plus kit (Roche Applied Science). 10⁴ cells/sample were plated in 48-well plates (Falcon) and challenged with the different stimuli. The cells were then lysed, and histone-associated DNA fragments were measured.

Morphological Analysis—The images were acquired with a Nikon Eclipse TE300 inverted microscope (Nikon, Tokyo, Japan) equipped with a thermostatted chamber (Bioptechs, Butler, PA) and a 63× oil immersion objective.

Western Blot—The cells were lysed by repeated freeze-thawing cycles in standard saline solution supplemented with protease inhibitors. The cell lysates (15 μg of total protein) were loaded on a Bis-Tris 4-12% NuPage acrylamide gel (Invitrogen) with MOPS buffer (Invitrogen), and electrophoretically transferred to Hybond-ECL nitrocellulose membranes (Amersham Biosciences). Primary antibodies were used at a 1:2000 (SERCA) and 1:300 (P2X₇) dilutions. Secondary reagents, goat anti-mouse (AbCam) or protein A (Amersham Biosciences), were added to the blocking solution at a 1:3000 dilution.

Measurement of NFATc1 Activation—Nuclear extracts were obtained with the nuclear extract kit (Active Motif, Rixensart, Belgium). NFATc1 was measured with the TransAM NFATc1 ELISA kit (Active Motif) as described by the manufacturer.

Data Analysis—All of the data are shown as the averages ± S.E. of the mean. Tests of significance were performed by Student-Newman-Keul's test using Graphpad InStat software (Graphpad, San Diego, CA).

RESULTS

HEK293-P2X₇ Cells Have an Increased Intracellular Store Ca²⁺ Content—As previously reported (15), [Ca²⁺]_mt level was 3-4-fold higher in HEK293-P2X₇ than in mock transfected cells. Accordingly, the [Ca²⁺]_mt transient in response to an agonist coupled to IP₃ generation (carbachol) was at least twice as high. Oxidized ATP is a powerful blocker of the P2X₇R (41). Despite its low selectivity, this inhibitor is a useful tool for the investigation of P2X₇R function because it covalently modifies the receptor, thus causing an irreversible inactivation. Furthermore, countless in vitro and in vivo experiments have shown that long lasting exposure to this inhibitor has no untoward effects on cell viability (42, 43). Differences in [Ca²⁺]_mt levels depended on a functional P2X₇ receptor as preincubation in the presence of oATP (2 h) equalized basal as well as carbachol-stimulated [Ca²⁺]_mt of P2X₇- and mock-transfected HEK293 cells (Fig. 1B). Basal [Ca²⁺]_mt was closely dependent on Ca²⁺ influx because it was obliterated by chelation of extracellular Ca²⁺ (Fig. 1C). The carbachol-stimulated [Ca²⁺]_mt rise was also affected by removal of extracellular Ca²⁺ but to a lesser extent than by preincubation in the presence of oATP. This is likely due to the brief (10 min) incubation under Ca²⁺-free conditions. To exclude clonal selection artifacts, carbachol-stimulated [Ca²⁺]_mt levels were measured in three different HEK293-P2X₇ stable clones (Fig. 1D). These clones were used throughout this study. As an average, carbachol-stimulated [Ca²⁺]_mt of HEK293-P2X₇ cells was at least twice as high as in mock transfected cells. Release of Ca²⁺ into the cytosol in response to carbachol or thapsigargin was also much higher in the P2X₇-transfectants than in mock transfected cells (Fig. 2). These experiments suggested that transfection with the P2X₇R conferred a stronger ability to accumulate Ca²⁺ into the ER. In support of this suggestion we observed that rate of Ca²⁺ uptake into the ER was significantly higher in HEK293-P2X₇ than in mock transfected or wild type HEK293 cells (Fig. 3, A and B). This difference was completely obliterated by pretreatment with the P2X₇ covalent inhibitor oATP (Fig. 3A). The increased Ca²⁺ uptake is likely to depend on a higher efficiency of the SERCA pumps because the level of SERCA protein expression did not differ between P2X₇- and mock transfected cells (Fig. 3C).

FIGURE 1.

P2X₇ expression increases basal and stimulated [Ca²⁺]_mt. Luminescence emission was recorded from HEK293-P2X₇ (red trace) and HEK293-mock (blue trace) cells transfected with mtAEQ as described under “Experimental Procedures” and stimulated with 500 μm carbachol. [Ca²⁺]_mt was determined in the presence of 1 mm extracellular Ca²⁺ (A and B) or with no Ca²⁺ added and in the presence of 500 μm EGTA (C and D). In B cells were preincubated for 2 h in the presence of 600 μm oATP. Average ± S.E. carbachol-stimulated [Ca²⁺]_mt levels (0.820 + 0.006 and 0.930 + 0.04 μm for mock-and P2X₇-transfected cells, respectively) were not statistically different. D, summary data of several independent measurements (n = 6) of carbachol-stimulated [Ca²⁺]_mt increases in Ca²⁺-free-EGTA-supplemented medium from mock transfected cells (empty bar) and three different HEK293-P2X₇ clones. ^**, p < 0.01; ^***, p < 0.001.

FIGURE 2.

P2X₇ expression enhances agonist-stimulated mobilization of intracellular [Ca²⁺]_i. fura-2-loaded HEK293-P2X₇ (red trace) and HEK293-mock (blue trace) cells were incubated in a fluorimeter cuvette at a concentration of 10⁶/ml. Carbachol (A and B) or thapsigargin (C and D) were added at the concentration of 500 and 2 μm, respectively. B and D, summary data of repeated (n = 6) [Ca²⁺]_i determinations from mock transfected cells (empty bar) and three different HEK293-P2X₇ clones. The absolute amount of Ca²⁺ released by carbachol and thapsigargin was determined by measuring the area under the peak as described under “Experimental Procedures.” ^*, p < 0.05; ^***, p < 0.001.

Blockade of P2X₇R Normalizes [Ca²⁺]_er and Obliterates the Growth Advantage of HEK293-P2X₇ Cells—We then investigated the dependence of [Ca²⁺]_er on a functional P2X₇ receptor. Fig. 4 (A and B) shows that treatment of HEK293-P2X₇ and mock transfected cells with oATP prior to stimulation with thapsigargin fully obliterated the difference in ER Ca²⁺ content between the two cell populations. It is likely that tonic activation of P2X₇ was responsible for supporting a larger Ca²⁺ influx from the extracellular space and therefore also a larger ER accumulation. Because HEK293 cells constantly release ATP into the pericellular space, we tested the effect of the ATP-hydrolyzing enzyme apyrase. As shown in Fig. 4C, preincubation in the presence of apyrase obliterated the differences in ER Ca²⁺ content between P2X₇-transfected and mock transfected cells. In addition, oATP treatment abrogated the growth advantage of HEK293-P2X₇ (Fig. 4D)_. Notice that oATP treatment had no effect on ER Ca²⁺ content or survival of mock transfected cells.

FIGURE 4.

Increased [Ca²⁺]_er depends on P2X₇ function and correlates with cell survival. A, fura-2-loaded HEK293-P2X₇ (red trace) and HEK293-mock (blue trace) cells were pretreated with 600 μm oATP, incubated in a fluorimeter cuvette in standard saline solution supplemented with 500 μm EGTA at a 10⁶/ml concentration and challenged with 2 μm thapsigargin. B, summary data of thapsigargin-releasable Ca²⁺ determined as described in Fig. 2, from HEK293-mock cells (white bar), or three different HEK293-P2X₇ clones in the absence or presence oATP (n = 3). C, effect of preincubation with apyrase (10 units/ml for 30 min) on thapsigargin-releasable Ca²⁺. The cells were incubated in 500 μm EGTA-containing standard saline and stimulated with 2 μm thapsigargin. The release rate of ER Ca²⁺ in the presence of apyrase was not statistically different in HEK293-P2X₇ and HEK293-mock cells. D, summary data of 24 h survival under serum-free conditions of HEK293-mock cells (white) or different HEK293-P2X₇ clones in the absence or presence oATP are shown (n = 15). ^**, p ≤ 0.01; ^***, p ≤ 0.001.

P2X₇ Transfectants Are Resistant to Various Apoptotic Stimuli—We next investigated whether HEK293-P2X₇R cell clones were also more resistant to cell death induced by different stimuli in addition to serum deprivation. Three agents were investigated: C2 ceramide, staurosporin, and the membrane-permeant Zn²⁺-chelating agent TPEN. As shown in Fig. 5, in the presence of extracellular Ca²⁺, challenge with C2 ceramide caused shrinkage and plasma membrane blebbing in mock transfected cells. On the contrary, no such changes were induced in HEK293-P2X₇ cells. The protective effect of P2X₇ expression was abrogated by removal of extracellular Ca²⁺, suggesting that the elevated [Ca²⁺]_mt levels in resting conditions play a critical role in the inhibition of the apoptotic process. The morphological alterations typical of apoptosis were corroborated by ELISA analysis of DNA fragmentation (Fig. 6A).

FIGURE 5.

Delayed blebbing in C2-ceramide-treated HEK293-P2X₇ cells. HEK293-P2X₇ and HEK293-mock cells were plated on coverslips at a concentration of 10⁵/ml in standard saline supplemented with 1 mm Ca²⁺ or with 500 μm EGTA. The coverslips were transferred to the thermostated stage of an inverted Nikon Eclipse 300 microscope, and 50 μm C2 ceramide was added. The images were acquired every 15 min for a total of 90 min. A, B, E, and F, HEK293-mock cells; C, D, G, and H, HEK293-P2X₇. A-D, standard saline solution plus 1 mm Ca²⁺. E-H, standard saline solution plus 500 μm EGTA.

FIGURE 6.

HEK293-P2X₇ cells are protected from apoptosis. A, HEK293-P2X₇ (red) and HEK293-mock cells (blue) were suspended in serum-free DMEM medium, plated in 48 well dishes at a concentration of 5 × 10⁴/ml, and challenged with C2 ceramide (20 μm), staurosporin (STS, 10 μm), or TPEN (4 μm). Apoptosis was evaluated as described under “Experimental Procedures” after 2 h (ceramide) or 16 h (staurosporin and TPEN). B, kinetics of C2-ceramide-stimulated trypan blue uptake from HEK293-P2X₇ (red) and HEK293-mock cells (blue). The cells were suspended in standard saline, plated in 48-well dishes at a concentration of 5 × 10⁴, and challenged with C2-ceramide (20 μm). Basal trypan blue uptake at time 0 (about 15% of total) was subtracted from all values. ^**, p < 0.01; ^***, p < 0.001 (n = 9).

HEK293-P2X₇ cells were refractory to apoptosis induced by any of the three pro-apoptotic agents used. Apoptotic cells in culture undergo a terminal increase in plasma membrane permeability, which can be monitored by the usual trypan blue uptake test. Even this gross index of cell death confirmed the lower susceptibility of HEK293-P2X₇ to cytotoxic stimulation (Fig. 6B).

Enhanced Survival of HEK293-P2X₇ Cells Depends on NFATc1 Activation—Calmodulin/calcineurin/NFAT is the main intracellular survival/growth-promoting pathway activated by mobilization of ER Ca²⁺ and the associated [Ca²⁺]_i increases (34). Thus, we hypothesized that the survival/growth advantage of HEK293-P2X₇ cells might depend on NFAT activation. As shown in Fig. 7A this anticipation was fulfilled, because HEK293-P2X₇ cells displayed a level of NFATc1 activation about twice as high as mock transfected cells. To support the key role played by P2X₇ and extracellular ATP in this process, enhanced NFATc1 activation was reverted by treatment with oATP or apyrase. In addition, the selective membrane-permeant blocker VIVIT, a peptide that blocks interaction between calcineurin and NFATc1, abrogated the higher NFATc1 activation level of HEK293-P2X₇. Accordingly, incubation in the presence of VIVIT or cyclosporin also abrogated the survival/growth advantage of the P2X₇ transfectants.

FIGURE 7.

Increased NFATc1 activation in HEK293-P2X₇ cells. A, HEK293-mock and HEK293-P2X₇ (clone A) were incubated for 16 h in serum free DMEM-F12 medium at 37 °C supplemented with either oATP (600 μm), apyrase (1 unit/ml), or VIVIT (1 μm). After this incubation time, the cells were rinsed and detached, and nuclear extracts were obtained as described under “Experimental Procedures.” Nuclear translocation of NFATc1 was measured by ELISA. B, mock transfected (blue) or HEK293-P2X₇ (red) cells were incubated at 37 °C in serum free DMEM-F12 medium in the absence or presence of VIVIT or cyclosporin for 24 h. At the end of this incubation, the cells were rinsed, detached, and counted. Similar data were obtained with two other stable HEK293-P2X₇ clones. ^***, p < 0.001 (n = 6).

P2X₇ Transfection Increases NFATc1 Activation, Enhances Growth, and Increases [Ca²⁺]_mt in NIH3T3 Cells—To show that the trophic effect of P2X₇ transfection is not restricted to a single cell line (HEK293 cells) but is likely to be widespread, we also run key experiments in another widely used cell line, NIH3T3 cells. As shown in Fig. 8, even in this mouse cell line P2X₇ transfection enhanced NFATc1 nuclear levels (Fig. 8A), proliferation (Fig. 8B), and [Ca²⁺]_mt in response to a Ca²⁺-mobilizing agonist (Fig. 8, C and D). To increase [Ca²⁺]_mt, bradykinin was used instead of carbachol because NIH3T3 cells did not respond to this latter agonist. At variance with HEK293-P2X₇, P2X₇ transfection did not increase basal [Ca^{2 +}]_mt, a difference likely caused by the fact that NIH3T3 cells were transiently transfected, while HEK293 cells were stably transfected.

FIGURE 8.

P2X₇ transfection increases NFATc1 activation, proliferation and [Ca²⁺]_mt in NIH3T3 cells. Mock transfected NIH3T3 cells blue bars, P2X₇-transfected NIH3T3 cells red bars. A, levels of nuclear NFATc1. The cells were transfected with P2X₇ or empty vector, 48 h after they were plated at a concentration of 10⁶/dish in serum-free DMEM, incubated for 16 h, and assayed for NFATc1 translocation as described under “Experimental Procedures.” B, proliferation was assayed with the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) test as described under “Experimental Procedures.” C, measurement of [Ca²⁺]_mt. NIH3T3 cells were incubated in 1 mm Ca²⁺ supplemented standard saline (see “Experimental Procedures”) and challenged with 1 μm bradykinin. D, summary data of repeated independent measurements (n = 9) of bradykinin-stimulated [Ca²⁺]_mt increases. ^**, p < 0.01; ^***, p < 0.001.

DISCUSSION

P2X₇R is the prototypic cytotoxic P2 receptor. However, increasing evidence, mainly from our laboratory, suggests that the P2X₇R may also have an apparently paradoxical survival/growth promoting effect. Long ago we made the observation that heterologous expression of this receptor confers a growth advantage in the absence of serum (6). Furthermore, several tumors express high level of P2X₇R protein (15). This finding is under evaluation for the design of novel anticancer therapies based on the administration of potent P2X₇R agonists that may cause a selective deletion of tumor cells. However, given the widespread presence of ATP in tissue interstitium, we believe it unlikely that tumor cells overexpress a potentially lethal receptor unless this turns out of advantage for the tumor cells. Thus, the growth promoting activity of P2X₇R was of no surprise to us. The intracellular pathways involved in this effect are as yet largely unknown; however data shown in this and our previous work point to at least four major players: Ca²⁺, mitochondria, ER, and NFATc1.

The ability of HEK293-P2X₇ to accumulate an increased amount of Ca²⁺ from extracellular medium is a necessary feature of the growth advantage as removal of extracellular Ca²⁺ abolishes this advantage (15). [Ca²⁺]_mt seems to be more important than [Ca²⁺]_i because we were unable to find consistently elevated [Ca²⁺]_i levels in resting conditions in HEK293-P2X₇, while on the contrary a higher [Ca²⁺]_mt is a constant finding. In the present study we show that in addition to [Ca²⁺]_mt, [Ca²⁺]_er is also higher in HEK293-P2X₇ than in mock transfected cells. Thus, it is likely that the higher [Ca²⁺]_mt depends on the higher [Ca²⁺]_er level. Given the presence of contacts between mitochondria and the ER, it is possible that ER Ca²⁺ is directly transferred to the mitochondrial matrix or alternatively is first discharged into the cytoplasm and subsequently taken up by the mitochondria.

The increased ER Ca²⁺ content is due to enhanced ER accumulating ability. We initially hypothesized that this might be due to a higher expression level of SERCAs, but this is not the case because the level of expression of these enzymes is unchanged in wild type, mock transfected, or P2X₇-transfected cells. Alternatively, functional activity of SERCAs might be increased in HEK293-P2X₇ cells, but clarification of this issue requires further investigation.

The ability to withstand unfavorable environmental conditions seems to be a distinct feature of HEK293-P2X₇ because additional apoptotic stimuli, together with serum starvation, are less effective in these than in mock transfected cells. The cellular mechanism is likely to involve multiple cellular pathways, including Ca²⁺-dependent processes. The protective role of [Ca²⁺]_er is clearly shown by experiments performed in cells chronically exposed to EGTA, a condition that brings the level of [Ca²⁺]_er of HEK293-P2X₇ cells to the same level of HEK293-mock.

The relationship between [Ca²⁺]_er and cell death is rather complex. There is a substantial body of data showing that reducing [Ca²⁺]_er has a protective effect against apoptosis dependent on stimuli releasing Ca²⁺ from intracellular stores (44-46). Accordingly, procedures that lower the ER Ca²⁺ content or prevent transfer of ER Ca²⁺ to the mitochondria also prevent apoptosis (47). In principle, P2X₇R-expressing cells should be “primed” for apoptosis because resting [Ca²⁺]_er and [Ca²⁺]_mt are elevated, but this is not the case. Because enhanced transfer of ER Ca²⁺ to the mitochondria has been identified as key step by which an increased [Ca²⁺]_er may accelerate apoptosis, it might be hypothesized that P2X₇R expression reduces ER to mitochondria Ca²⁺ transfer. However, we were consistently unable to detect a reduction of ceramide-stimulated Ca²⁺ transfer from the ER to the mitochondria (data not shown); thus the site protection should be elsewhere. On the other hand, our findings suggest that the increased ER Ca²⁺ content primes P2X₇-expressing cells for resistance to apoptosis.

Our data point to a more substantial involvement of the NFAT pathway. NFAT is central for translating survival and growth signals mediated by increases in [Ca²⁺]_i (34, 48). We found that in HEK293-P2X₇ cells the level of NFATc1 nuclear translocation was at least twice as high as in control cells. This was not a mere correlation because NFATc1 blockade with the specific inhibitor peptide VIVIT or with cyclosporin fully abolished the growth advantage of HEK293-P2X₇ cells. Furthermore, VIVIT fully abolished the protective effect stemming from P2X₇R expression. This is consistent with previous data showing that pharmacological stimulation of the P2X₇R causes NFAT activation (49). VIVIT also slightly reduced survival of HEK293-mock cells, but this is not surprising because NFATc1 is basally activated also in the absence of the P2X₇R. Thus, it is possible that tonic activity of the P2X₇R, even in the absence of exogenous ATP, might drive increased nuclear localization of NFAT. This possibility is strongly suggested by the inhibitory effect of oATP or apyrase treatment on NFATc1 translocation. NFATc1 is increasingly recognized as a major transducer of growth or survival signals in normal or transformed cells; thus our data bring additional support to this role.

Our results dramatically underscore the opposite effects on cell physiology that basal versus stimulated activation of the very same receptor might have. Strong pharmacological stimulation of the P2X₇R with exogenous ATP or benzoyl ATP causes a cell catastrophe characterized by collapse of plasma membrane selectivity barrier, fragmentation of mitochondrial network, nuclear condensation, and cell death, to the point that the P2X₇R is often taken as a paradigm for cytotoxic receptors. On the contrary, basal activation has a profound trophic effect. Available data from this and previous works concordantly support the presence of basal channel activity of P2X₇R, even in the absence of added ATP: (a) removal of extracellular Ca²⁺ normalizes [Ca²⁺]_mt;(b) treatment with oATP normalizes [Ca²⁺]_er, [Ca²⁺]_mt, and proliferation rate; and (c) treatment with apyrase or oATP normalizes [Ca²⁺]_mt and proliferation rate. This body of evidence strongly suggests that under physiological resting conditions the P2X₇R undergoes a tonic level of basal activation that allows a controlled influx of Ca²⁺, which supports cell metabolism and cell growth avoiding the deleterious effect of unchecked Ca²⁺ entry because of pharmacological activation of the receptor. In this scenario, a crucial and still unsolved issue is how a low affinity ATP receptor, such as P2X₇R, is activated in conditions in which the extracellular ATP concentration should be exceedingly low. A few hypothesis can be put forward: local ATP release into protected sites close to the P2X₇R or the presence of additional factors that positively modulate P2X₇R activity. However, it is clear that additional experiments are needed to clarify this issue. In conclusion, we have provided evidence showing that basal activation of the P2X₇R has an anti-apoptotic effect and have identified NFATc1 as one of the pathways involved.