P2Y₁ and P2Y₁₂ receptor cross-talk in calcium signalling: Evidence from nonstarved and long-term serum-deprived glioma C6 cells

Abstract

The current work presents results of experiments on the calcium response evoked by the stimulation by extracellular nucleotides occurring in control, nonstarved glioma C6 cells and in cells after long-term (96 h) serum starvation. Three nucleotide receptors were studied: P2Y₁, P2Y₂ and P2Y₁₂. Two of them, P2Y₁ and P2Y₂, directly stimulate calcium response. The protein level of the P2Y₂ receptor did not change during the serum starvation, while P2Y₁ protein level fell dramatically. Observed changes in the calcium response generated by P2Y₁ are directly correlated with the receptor protein level as well as with the amount of calcium present in the intracellular calcium stores, partially depleted during starvation process. The third receptor, P2Y₁₂, did not directly evoke calcium response, however it is activated by the same ligand as P2Y₁. The experiments with AR-C69941MX, the P2Y₁₂-specific antagonist, indicated that in control and serum-starved cells, calcium response evoked by P2Y₁ receptor is potentiated by the activity of P2Y₁₂-dependent signaling pathways. This potentiation may be mediated by P2Y₁₂ inhibitory effect on the plasma membrane calcium pump. The calcium influx enhanced by the cooperation of P2Y₁ and P2Y₁₂ receptor activity directly depends on the capacitative calcium entrance mechanism.

Introduction

In most cell types, a large number of cell functions are regulated by the level of cytoplasmic Ca²⁺. Efficient calcium signalling requires a Ca²⁺ source as well as mechanisms for Ca²⁺ depletion after a period of activation. Studies on the Ca²⁺ source have led to division of all cells into two types, excitable and nonexcitable, based on the presence of voltage-dependent calcium channels in their plasma membranes [1, 2]. Transformed glial cells, such as glioma C6, belong to the nonexcitable cells [3, 4]. We have reported that an increase in KCl concentration in the extracellular medium to 140 from 2.7 mM does not produce any change in the intracellular Ca²⁺ level in glioma C6 cells [3]. On the contrary, under such conditions, excitable cells open voltage-dependent Ca²⁺ channels via membrane depolarization, and massive influx of Ca²⁺ is observed [1, 2].

Nonexcitable cells are characterized by capacitative Ca²⁺ entry, also known as store-operated Ca²⁺ entry, regulated by the concentration of calcium ions in the endoplasmic reticulum (ER) pool [5–8]. The depletion of this pool (the first phase of calcium response) causes the opening of voltage-independent Ca²⁺ channels in the plasma membrane (PM), which permits the Ca²⁺ entry (second phase of calcium response). First messengers mediating inositol-1,4,5-trisphosphate (IP₃) formation cause release of Ca²⁺ from the ER via IP₃ receptors followed by Ca²⁺ entry into the cell as a consequence of the depletion of the ER store [5–8].

We have previously reported that, in glioma C6 cells, stimulation of nucleotide receptors by agonists ATP, UTP, ADP, or ADP-analogue 2-methylthioadenosine-5’-diphosphate (2MeSADP) initiates a biphasic Ca²⁺ response compatible with the typical capacitative model of Ca²⁺ influx [9–12]. Among the large family of metabotropic P2Y nucleotide receptors, P2Y₂ responds to ATP and UTP, whereas P2Y₁ and P2Y₁₂ respond to ADP and 2MeSADP. P2Y₁ and P2Y₂ receptors are both coupled to phospholipase C (PLC) and are responsible for Ca²⁺ mobilization from intracellular stores, while the P2Y₁₂ receptor is negatively coupled to adenylate cyclase [13–17]. We have also shown that differences in the cell culture conditions, i.e. the presence or absence of serum in the cell culture medium, have a strong effect on P2Y₁ receptor mRNA expression, which strongly decreases in the absence of serum [9, 10].

This effect is so strong that Slegers group [18], growing glioma C6 cells in serum-free chemically-defined medium, could not evoke P2Y₁ receptor-dependent PI turnover and suggested that, in these cells, the P2Y₁₂ receptor was not only coupled to inhibition of adenyl cyclase but also to activation of PLC-independent Ca²⁺ influx. According to the authors, the mechanism of P2Y₁₂-dependent calcium influx remained to be determined, however, in Fig. 8 of [19] and Fig. 1 of [20], they proposed the interesting idea that stimulation of the P2Y₁₂ receptor might generate, via Gβγ subunits of G_i protein, a direct entry of extracellular calcium [19, 20].

The aim of the present study was to check whether the P2Y₁₂ receptor may be directly responsible for the calcium entry. Our results indicate a full correlation between the P2Y₁ receptor protein level and the extent of Ca²⁺ response in glioma C6 cells. Moreover, although Ca²⁺ response evoked by the P2Y₁ receptor may be potentiated by P2Y₁₂-dependent signalling pathways, the hypothesis proposed by other authors [18–20], that in this cell line P2Y₁₂ is directly responsible for Ca²⁺ influx, seems to be doubtful. Thus, data presented herein clarify the problem of the roles that both receptors play in the calcium response.

Materials and methods

Materials

Dulbbeco’s Modified Eagle’s Medium (DMEM) and newborn calf serum (NCS) were obtained from Gibco BRL. ADP, 2MeSADP, UTP, EGTA, phosphate-buffered solutions (PBS), penicillin and streptomycin were purchased from Sigma Aldrich Chemical. Fura-2/AM was from Molecular Probes. AR-C69931MX was a kind gift from AstraZeneca (Wilmington, DE, USA). MRS 2179 tetraammonium salt was from Tocris. Antibodies recognizing P2Y₁, P2Y₂, and P2Y₁₂ were purchased from Alomone and Sigma Aldrich Chemical labs. Horseradish peroxidase-conjugated anti-rabbit IgG was from Cell Signalling. Nitrocellulose membrane and enhanced chemiluminescence detection system (ECL) were from Amersham Pharmacia Biotech. Medical X-ray films were from Foton Trading Poland. All other reagents were purchased from Sigma Chemical.

Cell culture

Rat glioma C6 cells (passages 40–60) were obtained from American Tissue Culture Collection and cultured in DMEM with high glucose (4,500 g/dm³) and GlutaMAX I, supplemented with 10% (v/v) NCS, penicillin (100 UI/ml) and streptomycin (50 μg/ml) under humidified atmosphere of 5% CO₂ at 37°C. For experiments, control cells were cultivated in DMEM supplemented with 10% NCS to reach 90% confluence in 60-mm dishes (for Western blot analysis) or on 22-mm glass coverslips in 35-mm dishes (for calcium measurement). In case of serum-starved cells, the medium was changed to DMEM without NCS 96 h before the experiment.

Measurement of intracellular calcium

Thirty minutes before the calcium measurements, cells on coverslips were washed once with PBS and once with solution containing 137 mM NaCl, 2.7 mM KCl, 1 mM Na₂HPO₄, 25 mM glucose, 20 mM HEPES (pH 7.4), 1 mM MgCl₂, 1% bovine serum albumin and 2 mM CaCl₂ (standard buffer). In experiments performed in the absence of external Ca²⁺, 500 μM EGTA was added instead of 2 mM CaCl₂. The cells were then incubated at 37°C for 30 min in the standard buffer with 2 μM Fura-2 AM. Thereafter, the coverslips were mounted in a chamber on a Nikon Diaphot inverted-stage microscope equipped with a fluo ×40/1.3 NA oil-immersion objective lens. Fura-2 digital fluorescence microscopy was used to determine the changes in intracellular calcium levels ${\left( {{\left[ {Ca^{{2 + }} } \right]}_{i} } \right)}$ [21].

Ludl Lep MAC 5000 filter wheel system loaded with a Chroma Fura-2 filter set was used for illumination of specimens. Images were acquired using Retiga 1300 chilled digital CCD camera (QImaging). Data processing was carried out using AQM Advance 6 (Kinetic Imaging Inc) and MS Excel software. All data are expressed as 340/380 nm-induced fluorescence of Fura-2 ratio changes against time (Δ340/380). Each experiment was repeated at least three times, and data are expressed as means.

For evoking the calcium response, 30 μM MeSADP, 100 μM UTP or 10 μM ADP was used as described by Sabala et al. [12]. For inhibition of P2Y₁₂, 10 μM AR-C69931MX was used, and 30 μM MRS 2179 was used for inhibition of P2Y₁ receptors.

Western blot analysis

For the experiments, cells were grown up to 85% confluency on 100-mm dishes in DMEM supplemented with 10% NCS. In order to cause serum deprivation, the medium was replaced by fresh serum-free medium for 96 h. After this time, cells were washed in PBS, detached with a 3-min CDS (cell dissociation solution) treatment, centrifuged at 3,000 rpm for 5 min and then resuspended in lysis buffer containing 1% Nonidet P-40, 120 mM NaCl, 50 mM Tris/HCl, pH 7.5, and freshly added proteinase inhibitors. Separation of proteins was performed on 12% polyacrylamide SDS/PAGE. Proteins were transferred to nitrocellulose membranes (Hybond C, Amersham Pharmacia Biotech), blocked for 1 h at room temperature with 5% milk in PBS-T (phosphate-buffered saline pH 7.6/0.005% Tween 20) and incubated overnight at 4°C with the antisera against β-actin (Sigma) and P2Y₁, P2Y₁₂, and P2Y₂ (all 1:2,500, Alomone and Sigma Adrich) diluted in 1% nonfat milk in PBS-T.

The primary antibody reaction was followed by 2 h incubation with relevant secondary (1:2,500) antibody conjugated to HRP. Immunocomplexes were detected using the ECL-enhanced chemiluminescence detection system and membrane exposure to X-ray film. The molecular weight of proteins was estimated with prestained protein markers (Fermentas protein ladder). Band intensities were determined by densitometry analysis in an Ingenious station using provided programs. Only statistically significant immunoblot band intensity data are reported.

Statistical analysis

Nonparametric Mann-Whitney U-test was used to discriminate differences between calcium responses. Differences with P < 0.01 were considered highly significant and marked with two asterisks on the plots, P < 0.001 were marked three asterisks, the lack of statistically significant difference was marked with a minus sign.

Results

We have previously shown the presence of P2Y₁, P2Y₂ and P2Y₁₂ nucleotide receptors in glioma C6 cells using polymerase chain reaction on reverse-transcribed total mRNA [11, 12]. We have also shown that the mRNA expression level, as well as Ca²⁺ responses to extracellular nucleotides, may depend on the presence of serum in the cell-cultured medium. In cells that have been starved of serum for 48 h, the expression of the P2Y₁ receptor mRNA decreases and P2Y₁₂ predominates [9, 10]. This change reflects the lower intensity of Ca²⁺ response [9]. Furthermore, we have presented experimental data that the serum-free chemically defined medium used by other authors [18] causes similar changes, decreasing the expression of P2Y₁ receptor mRNA [10]. These data show that observed changes are present at the gene expression level.

In the present study, we explored the effects of prolonged serum deprivaton, up to 96 h, on the expression of the described receptors and measured changes in their protein level using Western blot analysis. Figure 1 shows that upon such treatment, the level of the P2Y₁ receptor protein was very low. Densitometry analysis of blots from three independent experiments revealed about a 20-fold decrease in this receptor protein level and a distinct, 3-fold increase in the P2Y₁₂ receptor.

Fig. 1

The effect of long-term (96 h) serum deprivation on the relative expression levels of P2Y₁, P2Y₂ and P2Y₁₂ receptors. Total protein isolated from glioma C6 cells was subjected to SDS-PAGE on 12% acrylamide gel and transferred onto nitrocellulose membrane. Filters were probed with the indicated P2Y antisera. Protein bands were detected with secondary antisera coupled to HRP by ECL chemiluminescences. Band intensity was determined by densitometry analysis. The data represent the means ± SD from three independent experiments

When the level of P2Y protein receptors in the control C6 cells, cultivated in the medium supplemented with 10% new born calf serum, was taken as 100%, the level of P2Y₁ receptor protein in the 96-h-deprived serum cells was 6 ± 2%, and P2Y₁₂ was 306 ± 60% (Fig. 1). Additionally, we have shown that serum deprivation had no effect either on P2Y₂ mRNA expression [9] or on total receptor protein level (Fig. 1). Multiple bands observed for P2Y receptors are probably the effect of protein glycosylation and were shown to disappear after total receptor deglycosylation [22]. We investigated such phenomena in glioma C6 cells and recently showed that preincubation of lysates of cells with N-glycolidase F resulted in reduction of bands detected by P2Y antibodies [23].

Since under long-term serum starvation the P2Y₁₂ receptor expression strongly predominates, one could suppose that this receptor might be primarily involved in ADP- or 2MeSADP-evoked signal transduction. Slegers group [18] suggested the existence of an alternative, IP₃-independent, calcium influx pathway caused by P2Y₁₂ receptor activation [19, 20].

The fact that in long-term serum-deprived C6 glioma cells the P2Y₁₂ receptor is massively expressed and P2Y₁ strongly diminished provides a good model system to study functional properties of both receptors. Therefore, in our subsequent experiments, Ca²⁺ responses to nucleotides in the control cells were compared to those in the cells growing 96 h in the medium without serum.

Figure 2 shows the effect of 30 μM 2MeSADP (Fig. 2a–d) and 10 μM ADP (Fig. 2e–h) on Ca²⁺ response in the control, nonstarved cells, and in the 96-h serum-starved cells. Figure 2a, b, e, f shows cells studied in the extracellular medium containing 2 mM CaCl₂, while Fig. 2c, d, g, h shows calcium transients in cells tested in the medium containing 500 μM EGTA. Statistical analysis (mean value ± SD from at least three separate experiments for the indicated number of cells; data were standardized to obtain stationary calcium levels equal 1 arbitrary unit, AU) revealed that in the control cells the initial Ca²⁺ peak evoked by 2MeSADP (Fig. 2a) and ADP (Fig. 2e) were 1.50 ± 0.15 AU for 2MeSADP (n = 71) and 1.47 ± 0.37 AU for ADP (n = 51), whereas the values in serum-starved cells were 1.19 ± 0.21 AU for 2MeSADP (n = 79) and 1.17 ± 0.25 AU for ADP (n = 134). The difference was statistically significant with P < 0.001.

Fig. 2 a–h

The effect of the long-term, 96-h serum deprivation on 2MeSADP-evoked calcium signals. The glioma C6 cells growing in the medium supplemented with 10% newborn calf serum (control cells) and those cultivated without serum for 96 h were loaded with Fura-2 and treated with 30 μM 2MeSADP or 10 μM ADP, as indicated by arrows. a, b, e, f Experiments performed in the standard buffer containing 2 mM CaCl₂. c, d, g, h Experiments conducted in the absence of extracellular Ca²⁺. Each trace represents the mean ratio value of the responses of the indicated number of cells (n), recorded in five separate experiments. a Black line Control cells induced by 2MeSADP, n = 71; gray line 96-h starved cells, n = 79. b Ca²⁺ responses in 96-h serum-deprived cells induced by 2MeSADP, n = 79, divided into groups according to the extent of response (light gray: cells with strong response, medium gray: cells with medium response, dark gray: cells with weak response). Black line Mean ratio value. c Control cells induced by 2MeSADP, n = 51. d 96-h starved cells induced by 2MeSADP, n = 49. e Black line Control cells induced by ADP, n = 51; gray line 96-h starved cells, n = 134. f Ca²⁺ responses in 96-h serum-deprived cells induced by ADP, n = 134, divided into groups according to the extent of response (light gray cells with strong response, medium gray cells with medium response, dark gray cells with weak response). Black line Mean ratio value. g Control cells induced by ADP, n = 97. h 96-h starved cells induced by ADP, n = 235. ***P < 0.001

The calcium signal strength thus fell to ~0.2 AU from ~0.5 AU indicating 60% inhibition of the signal. The initial calcium transient was associated with the depletion of intracellular stores (the first phase of the calcium response), followed by sustained elevation of Ca²⁺ (the second phase of the calcium response), which was a result of a subsequent influx of extracellular Ca²⁺ to the cells.

It is worth adding that Ca²⁺ responses are often analyzed only in cells that respond to agonists. We have previously shown that 70–95% of glioma C6 cells, growing in the medium supplemented with serum responded to extracellular nucleotides [12]. Similarly, of the control cells presented in Fig. 2a and e, 80% responded to agonist. In contrast, among serum-deprived cells, only 8.8% (7 out of 79) responded to 2MeSADP, and 16.4% responded to ADP (22 out of 134). Therefore, the traces shown in Fig. 2a and e represent mean values of all individual cell responses, including those responding and those not responding to the agonist.

Figure 2b and f shows Ca²⁺ responses in 96-h serum-deprived cells, divided into groups according to the extent of the response. It is shown that in the cells with the highest response, a rapid rise in intracellular Ca²⁺ level associated with the depletion of the ER stores is emphasized (the first phase), whereas the second phase is diminished. Figure 2c, d, g and h shows Ca²⁺ responses in the absence of extracellular Ca²⁺. It is shown that addition of agonist resulted only in the initial transient rise in intracellular Ca²⁺ concentration, which declined to the basal level (the first phase of the calcium response) indicating that this phase of the cytosolic Ca²⁺ increase was indeed caused by mobilization of intracellular Ca²⁺ stores. The mean values of Ca²⁺ elevation in control cells were 1.30 ± 0.17 AU (n = 51, Fig. 2c) for 2MeSADP and 1.25 ± 0.56 AU (n = 97, Fig. 2g) for ADP; the values in serum-starved cells were 1.05 ± 0.13 AU (n = 49, Fig. 2d) for 2MeSADP and 1.03 ± 0.27 AU (n = 235, Fig. 2h) for ADP. Among serum-deprived cells, 10% (5 out of 49) responded to 2MeSADP and 7% (17 out of 235) to ADP.

Thus, the very low expression of the P2Y₁ receptor in long-term serum-deprived cells strictly corresponds to the low intensity of the functional effects. When, in serum-starved cells, the P2Y₁ receptor protein level decreases to 6% of that in the control cells, this change is reflected by approximately 10% of the cells strongly responding to 2MeSADP or ADP. Since in serum-starved cells the expression of P2Y₁₂ receptor protein distinctly increases, low calcium responses reflect stimulation of P2Y₁ and not of P2Y₁₂ receptor. Nevertheless, the average calcium signal reduction is much weaker than can be expected from 94% reduction of P2Y₁ receptor protein level.

To check the possibility that P2Y₁₂ may still play a role in contributing to the agonist-induced Ca²⁺ response, the extent of inhibition of this response by the P2Y₁₂ selective antagonist AR-C69931MX (10 μM) was determined. Figure 3 shows that both in the control, nonstarved (Fig. 3a, c) and in the serum-deprived (Fig. 3b, d) cells, AR-C69931MX reduced the strength of the calcium response (measured as the signal integral) to 2MeSADP by 42.3% (Fig. 3a) and 69.8% (Fig. 3b), respectively, and to ADP by 31.9% (Fig. 3c) and 57.3% (Fig. 3d). In both cases the difference was statistically significant with P < 0.005. It is important to notice that, as shown in Fig. 3a and c, the observed difference in calcium signal integral caused by AR-C69931MX affected the second phase of calcium response. Since the plots presented in Fig. 3 show the overlay, we can not precisely state whether or not the fall of the initial peak in Fig. 3a is the result of the strong second-phase reduction.

Fig. 3

The effect of AR-C69931MX, the P2Y₁₂ receptor antagonist, on Ca²⁺ signals evoked by 30 μM 2MeSADP (a, b), 10 μM ADP (c, d), and 100 μM UTP (e, f). Black lines represent mean ratio values of Ca²⁺ responses of cells untreated with P2Y₁₂ antagonist. a, c, e Nonstarved cells, respectively, n = 71, n = 51, n = 73. b, d 96-h serum-starved cells, respectively, n = 79, n = 134. Gray lines in the same panels represent mean ratio value of Ca²⁺ responses of cells pre-treated for 3 min with 10 μM AR-C69931MX (a, n  =  95; b, n = 80; c, n = 19 ; d, n = 175; e, n = 79) and then, while still in its presence, stimulated by agonist, as indicated by arrows. f 100 μM UTP-evoked Ca²⁺ signals in the control cells (n = 73, black line); and 96-h serum-starved cells, (n = 120, gray line). Each trace represents mean ratio value from five experiments. ***P < 0.001. Statistically insignificant differences are indicated with a minus sign

In order to demonstrate that AR-C69931MX has no effect upon other P2Y receptors, Ca²⁺ responses were evoked by UTP (100 μM), an agonist of the P2Y₂ receptor. In these cases, there was no statistically significant change in the peak response either in the presence or absence of this antagonist (Fig. 3e). It is worth adding that MRS2179, the P2Y₁ receptor antagonist, led to a 96% reduction in calcium response induced by ADP [10, 11], or 75% reduction if induced by 2MeSADP (data not shown). Figure 3f also shows that the UTP-generated Ca²⁺ response is diminished in the 96-h serum-starved cells. In the control cells, the initial peak of Ca²⁺ elevation was 2.50 ± 0.23 AU (n = 73), whereas in the serum-deprived cells it was 1.58 ± 0.16 AU (n = 120). The difference was statistically significant with P < 0.001. Thus, 96-h, prolonged cell cultivation in the medium without serum blocked the effect of UTP by approximately 47.4%. On the other hand, serum-deprivation did not decrease the expression level of P2Y₂ receptor protein (Fig. 1).

Figure 4a shows two phases of calcium response in glioma C6 cells. The signal was evoked by UTP in the Ca²⁺-free medium. Under such conditions, the rise in ${\left[ {Ca^{{2 + }} } \right]}_{i}$ was the effect of depletion of intracellular stores (ER). When the transient rise in calcium returned to the basal level, a standard, calcium-containing buffer was added. The addition of Ca²⁺ to the extracellular medium generated a high elevation in the cytosolic Ca²⁺ concentration. These data demonstrate that the Ca²⁺ entry is solely activated by the depletion of the ER Ca²⁺ store.

Fig. 4 a–c

The effect of intracellular store depletion evoked by UTP, 2MeSADP, or by incubation in the calcium-free buffer on Ca²⁺ influx in glioma C6 cells. a 100 μM UTP was added to the control cells (black line, n = 73), or to 96-h serum-starved cells (gray line, n = 120), placed in a buffer without Ca²⁺ and with 500 μM EGTA. Next, 5 min after the addition of UTP, the buffer was changed for the buffer containing 2 mM CaCl₂ (Ca²⁺). b The control cells (black line, n = 30) and 96-h serum-starved cells (gray line, n = 114) were placed in a buffer without Ca²⁺ and with 500 μM EGTA, and after 8 min incubation in the Ca²⁺-free buffer, new buffer containing 2 mM CaCl₂ (Ca²⁺) was added. c 30 μM 2MeSADP was added to the control cells (black line, n = 78) or to 96-h serum-starved cells (gray line, n = 92) placed in a buffer without Ca²⁺ and with 500 μM EGTA; 5 min after the addition of 2MeSADP, the buffer was changed for the buffer containing 2 mM CaCl₂ (Ca²⁺). Each trace in a, b, and c represents the mean value for the indicated number of cells tested in three to five experiments. *** P < 0.001, **P < 0.01. Statistically insignificant differences are marked with a minus sign

However, when we compared UTP-induced Ca²⁺ responses in the control (1.95 ± 0.27 AU, n = 95) and serum-starved cells (0.52 ± 0.18 AU, n = 83), tested in Ca²⁺-free medium, there was a visible difference in the response of both groups. The response to UTP in 96-h serum-deprived cells was much lower, and the difference was statistically significant with P < 0.001. On the other hand, after exchanging that medium for one containing Ca²⁺, the strong and rapid elevation of the cytosolic Ca²⁺ level was the same in both control (1.92 ± 0.28 AU, n = 95) and serum-deprived (1.83 ± 0.24 AU, n = 83) cells (Fig. 4a); the difference was statistically insignificant.

Please note that the data suggested that the different response to UTP in Ca²⁺-free medium observed in the two groups of cells (see also Fig. 3d) not only depended on the expression of physiologically active receptors (which in this case was on the same level in both starved and nonstarved groups of cells) but also on the amount of Ca²⁺ present in the intracellular store. To check this assumption, 96-h serum-starved cells were placed in a buffer without Ca²⁺ for about 8 min, and then the buffer was changed for that containing 2 mM CaCl₂. This change resulted in a rapid increase in [Ca²⁺]_i caused by extracellular Ca²⁺ influx (1.28 ± 0.14 AU, n = 114) (Fig. 4b). Thus, prolonged, 96-h serum deprivation constitutes inhospitable conditions, manifested not only in the decrease in the P2Y₁ receptor expression but also in a partial, autonomous depletion of ER stores from Ca²⁺. Nonstarved cells did not respond by increasing [Ca²⁺]_i when there was an exchange from calcium-free medium with that containing 2 mM CaCl₂ (Fig. 4b, n = 36).

Therefore, in the long-term serum-starved glioma C6 cells tested in Ca²⁺-free buffer, there was a very small, but statistically significant (P < 0.001) increase in ${\left[ {Ca^{{2 + }} } \right]}_{i}$ (0.31 ± 0.11 AU, n = 111, Fig. 4c) induced by 2MeSADP. It was caused by both the low expression of the P2Y₁ receptor protein level and by a partial depletion of ER stores from Ca²⁺ (Figs. 2d and 4c). The change in the extracellular medium for the buffer containing Ca²⁺ caused Ca²⁺ influx to the cells (Fig. 4c). The influx after 96 h of serum starvation was even stronger (1.93 ± 0.31 AU, n = 107) than in control cells (1.43 ± 0.24 AU, n = 111); the difference was statistically significant with P < 0.01. It is the depletion of ER stores that gives a signal to voltage-independent Ca²⁺ channels and Ca²⁺ entry across plasma membranes, according to the mechanism of the capacitative Ca²⁺ entry.

In subsequent experiments, the role of P2Y₁₂ on the capacitative Ca²⁺ entry was examined in the nonstarved cells. In such experiments the cells were incubated in the medium without Ca²⁺ (Fig. 5a). After addition of ADP, the first phase of calcium response was observed. Thereafter the medium was replaced by standard buffer containing 2 mM CaCl₂ (with no agonist).

To distinguish if the P2Y₁₂ receptor plays an active role in the calcium signal formation or if it only modulates the result of P2Y₁ receptor activity, we used AR-C69931MX in two separate experimental setups. In the first experiment, P2Y₁₂ receptor competitive antagonist was used before addition of agonist to inhibit both hypothetical P2Y₁₂ direct calcium signalling as well as regulation of P2Y₁ activity by P2Y₁₂ receptor. In the second experiment, the use of antagonist well after agonist addition but before medium replacement with that containing calcium should affect only regulatory functions of P2Y₁₂ receptor but not its ability to directly form the calcium signal.

As has been shown, AR-C69931MX has an inhibitory effect on the second phase of Ca²⁺ response, however this effect does not depend on the moment of the antagonist addition (cells treated with AR-C69931MX before ADP addition: 1.43 ± 0.44 AU, n = 12; cells treated with AR-C69931MX after ADP addition: 1.55 ± 0.42 AU, n = 18; the difference was not statistically significant) (Fig. 5b). Figure 5b also shows that AR-C69931MX has no statistically significant effect on the first phase of the calcium response. The same results were observed when 2MeSADP was used as an agonist (data not shown).

Fig. 5 a, b

The effect of AR-C69931MX, the P2Y₁₂ receptor antagonist, on the first and second phase of calcium response in cells grown in the presence of serum. a Cells in Ca²⁺-free buffer were stimulated by 10 μM ADP (arrow). Two cell groups were studied: the first experimental group was treated with 10 μM AR-C69931MX for 3 min before ADP addition and release of calcium from the ER (light gray line, ARC - ADP), while the second experimental group was treated for 3 min with 10 μM AR-C69931MX just after the end of calcium transient caused by ADP-induced release from the ER (dark gray line, ADP - ARC). The control cell group was induced with ADP, without use of AR-C69931MX (black line, Ctrl). Then, in all cell groups, the medium was exchanged with fresh medium containing 2 mM CaCl₂ (arrow). b Strength of two phases of calcium response. Colour of bars as described above. Difference between bars marked with a minus sign is statistically insignificant

Discussion

It is now well-documented that in the G_q-dependent signalling initiated by ADP or 2MeSADP, the P2Y₁ receptor stimulation triggers PLCβ activation and ${\left[ {Ca^{{2 + }} } \right]}_{i}$ increase [9–17]. On the other hand, the same agonists, via the P2Y₁₂ receptor, activate the G_i pathway and inhibit adenylate cyclase in various animal cells [11, 12, 18, 24, 25]. The cross-talk between those two receptors is extremely complex [9, 26]. In human platelets, Sage et al. [27] and Fox et al. [28] suggested that P2Y₁₂ may enhance P2Y₁-induced cytosolic Ca²⁺ rise, whereas Daniel et al. [29] presented evidence that this receptor is not involved in such response. Hardy et al. [30] have explained this conflicting evidence as the different conditions used during platelets preparation. Similarly in glioma C6 cells, there is conflicting evidence regarding the role of P2Y₁ in ADP-mediated calcium response that can also be explained by the differences in the culture conditions [10]. Presence or absence of serum in the culture medium provides conclusions on functional activity [9–11] or inactivity [18] of this receptor.

Hardy et al. [30], as well as Sage et al. [27], suggested the modulatory role of P2Y₁₂, positively regulating P2Y₁-induced Ca²⁺ response. It has been suggested that this potentiation is mediated by P2Y₁₂-induced inhibition of adenylate cyclase and activation of phosphatidylinositol 3-kinase (PI3-K), whereas the effect of P2Y₁ on PI3-K is inhibitory [30]. Our previous study concerning cross-talk between nucleotide receptor-induced signalling pathways in glioma C6 cells also revealed P2Y₁ inhibitory and P2Y₁₂ stimulatory effects on PI3-K signalling [9, 10]. Thus, since stimulation of the P2Y₁₂ receptor in glioma C6 inhibits adenylate cyclase [11, 12] and stimulates PI3-K [9, 10], its modulatory effect on the P2Y₁-induced Ca²⁺ responses in this cell line may occur via a similar mechanism to the one suggested in platelets [30]. It has been proposed that in this process the cAMP-dependent pathway has a stimulatory effect on PM calcium pumps, thereby limiting the strength of the calcium response. The P2Y₁₂ receptor reduced this effect by inhibition of adenylate cyclase activity. Hardy et al. [30] suggested that PI3-K, activated by the P2Y₁₂ receptor stimulation, may also activate PLCγ, leading to the rise in PIP₃ and enhancement of the calcium signal. It has been reported that receptors coupled to PI3-K may activate PLCγ indirectly in the absence of PLCγ-tyrosine phosphorylation [31, 32].

Our study confirms the conclusions of Hardy et al. [30] concerning the positive regulatory role of P2Y₁₂ potentiating the P2Y₁-induced rise in intracellular Ca²⁺ level. We have shown here that, in glioma C6, the calcium response to ADP or its analogue 2MeSADP is directly dependent upon activation of the P2Y₁ receptor. The modulatory effect of P2Y₁₂ receptor on intracellular free calcium level seems to be a result of the PM calcium pump inhibition [26]. The pump activity is enhanced by cAMP. The constant activation of P2Y₁₂ receptor lowers the level of cAMP and inhibits the PMCA pump. Thus, inhibition of this receptor activity may enhance the pump efficiency and lower the calcium signal as shown in the present study when we observed decreased calcium influx under the influence of the P2Y₁₂ competitive antagonist AR-C69931MX (Fig. 5). A similar phenomenon of calcium-removal inhibition was described in platelets, where the adenylate cyclase inhibitor SQ22536 reduced the strength of calcium signal by approximately 50% [26].

Thus, the suggestion that Ca²⁺ influx in this cell line may involve a channel, activated directly or indirectly by protein Gβγ subunits after the P2Y₁₂ receptor stimulation (see figures in [19, 20]), seems to be doubtful. If such a channel were to exist it should be opened by an agonist when P2Y₁₂ antagonist is added after addition of the agonist and remain closed if P2Y₁₂ antagonist is added before this agonist (Fig. 5). Since we did not observe such a difference in the calcium signal strength we may exclude the existence of a P2Y₁₂-dependent calcium channel in glioma C6 cells. Even if there is some evidence that the stimulation of P2Y receptors may affect N-type Ca²⁺currents [33], glioma C6 cells belong to the type of nonexcitable cells and do not possess voltage-dependent Ca²⁺ channels [3].

In conclusion, the present study shows that Ca²⁺ influx in these cells occurs exclusively via the mechanism of capacitative calcium entry. Thus any activation of calcium-dependent signalling pathways resulting from the presence of extracellular ADP must be the result of P2Y₁-P2Y₁₂ cross-talk, not solely P2Y₁₂ receptor activity. Nevertheless, the P2Y₁₂ modulatory effect on glioma C6 calcium signalling seems to be very important.