Introduction

Top Summary Introduction Procedures Results Discussion References

Purinergic responses of the P2 type are mediated by two classes of membrane receptor: ionotropic P2X receptors and metabotropic P2Y receptors (Abbracchio and Burnstock, 1994; Harden et al., 1995; Burnstock and King, 1996; Weissman et al., 1996). The structures of P2Y₁ receptors from chicken (Webb et al., 1993), turkey (Filtz et al., 1994), bovine (Henderson et al., 1995), rat, mouse (Tokuyama et al., 1995), and human (Ayyanathan et al., 1996; Léon et al., 1996) species are known. ATP, 2-MeSATP, 2-ClATP, 2-MeSADP, and ADP usually are considered to be potent agonists of P2Y₁ receptors (Boyer et al., 1993; Vigne et al., 1994; Boyer et al., 1996; Schachter et al., 1996), although ATP has been reported in some studies to be only a partial agonist (Feolde et al., 1995; Henderson et al., 1995; Léon et al., 1997). Characterization of the pharmacological properties of purinoceptors is, however, difficult because nucleotides from commercial sources do not have the desired purity and the purity of their aqueous solutions decreases considerably during storage. In addition, nucleotides can be degraded by cellular ectonucleotidases. These difficulties can be circumvented by using freshly purified nucleotides (Léon et al., 1997) or enzymatic systems that regenerate degraded nucleotides (Nicholas et al., 1996). In a previous article, we described the pharmacological characteristics of the human P2Y₁ receptor transfected into Jurkat cells. Results showed that (1) ADP was a selective agonist of this receptor, (2) freshly purified ATP and ATP derivatives were ineffective, and (3) ATP antagonized the effects of ADP (Léon et al., 1997). On the other hand, purified 2-MeSATP and 2-ClATP were found to be full agonists but with delayed responses compared with the corresponding diphosphate derivatives. It was suggested that the triphosphate analogues were metabolized into diphosphates by ectoenzymes, thus explaining their apparent agonistic effect. These results led us to put forward the idea that the P2Y₁ receptor is similar to the platelet ADP receptor, the elusive P2T receptor (Gachet et al., 1997).

The first aim of the current study was to determine the conditions that allow control of the purity of ATP nucleotides by means of a CP/CPK ATP-regenerating system. This technique then was used to define more precisely the pharmacological properties of the human P2Y₁ receptor expressed in Jurkat cells and the endogenous P2Y₁ receptor expressed in rat brain capillary endothelial cells (Webb et al., 1996). These latter cells were found previously to express specific ADP receptors at which ATP behaved as a partial agonist (Frelin et al., 1993; Feolde et al., 1995). Results showed the rat and human P2Y₁ receptors to be ADP-specific receptors antagonized by ATP and its derivatives, 2-ClATP and 2-MeSATP, thus pointing to the existence of strong pharmacological similarities between the endothelial P2Y₁ receptor and the platelet ADP receptor.

	Experimental Procedures

Top Summary Introduction Procedures Results Discussion References

Materials. ATP, ADP, iodoacetamide, creatine, CP, and CPK (Type III from bovine heart) were from Sigma Chemical (St. Quentin-Fallavier, France). Indo-1/AM was from Calbiochem (Meudon, France), whereas 2-ClATP, 2-MeSADP, and 2-MeSATP were from Research Biochemicals (Natick, MA).

Cell cultures and intracellular Ca²⁺ measurements. Rat brain capillary endothelial cells of the B10 clone (Feolde et al., 1995) and Jurkat cells stably transfected with the human P2Y₁ receptor (Léon et al., 1997) have been described previously. The human P2Y₁ receptor sequence differs from that originally described (Léon et al., 1996) by the presence of an additional 3 base pairs leading to an additional serine residue in transmembrane domain 3, between residues 137 and 138 (Léon et al., 1997). An identical sequence was reported by Schachter et al. (1996).

Purification of ATP and ATP derivative solutions. The purity of nucleotide solutions was checked by HPLC analysis (Léon et al., 1997) on a Partisil 10-µm SAX column (Interchrom; Interchim, Montluçon, France) eluted with a linear gradient from 0% to 100% of 1 M ammonium phosphate buffer, pH 3.8, after 20 min with a flow rate of 1 ml/min. Absorbance was recorded at 260 nm for ATP and 274 nm for 2-ClATP and 2-MeSATP. As discussed previously (Nicholas et al., 1996; Léon et al., 1997), commercial nucleotide powders often are contaminated by degradation products that can be the cause of misleading results. Contamination was usually 1% for ATP and 10% for 2-MeSATP and 2-ClATP (see Results). Nucleotides also may be degraded by cell ectonucleotidases. Problems arising from the contamination and degradation of ATP solutions were circumvented by using an ATP-regenerating system. ATP solutions (1 mM) were treated at room temperature with 20 units/ml CPK and 10 mM CP, and the entire mixture was added to cell suspensions. The same procedure was used to purify 2-ClATP and 2-MeSATP solutions from contaminating ADP derivatives. Time course experiments indicated, however, that exposition to CPK had to be increased to remove contaminants of 2-MeSATP and 2-ClATP solutions. A 30-min incubation period was necessary to purify 2-ClATP solutions. This time was 90 min for 2-MeSATP. These differences probably reflect different catalytic efficiencies of CPK toward ATP, 2-ClATP, and 2-MeSATP. Contaminant 2-MeSADP and 2-ClADP were identified by mass spectrometry using a Bio-Q triple quadrupole mass spectrometer (Micromass Ltd, Altrincham, UK).

Platelet aggregation. Aggregation was measured at 37° by a turbidimetric method in a dual-channel Payton aggregometer (Payton Associates, Scarborough, Ontario, Canada). A 450-µl aliquot of citrated platelet-rich plasma was stirred at 1100 rpm and activated by the addition of agonists to a final volume of 500 µl (Cazenave et al., 1983).

	Results

Top Summary Introduction Procedures Results Discussion References

Partial agonism of ATP is abolished in the presence of an ATP-regenerating system. Purinergic responses were analyzed in rat brain capillary endothelial cells of the B10 clone that specifically express mRNA sequences corresponding to the cloned P2Y₁ receptor (Webb et al., 1996) and on P2Y₁ receptor-transfected Jurkat cells. Fig. 1 shows dose-response curves for the actions of ADP on Jurkat cells and B10 cells. EC₅₀ values for the actions of ADP were 0.88 ± 0.15 µM (10 experiments) in B10 cells and 0.20 ± 0.06 µM (3 experiments) in Jurkat cells. Similar results were obtained with ADPS, which behaved as a full agonist. EC₅₀ values for the action of ADPS were 0.90 ± 0.05 µM (3 experiments) in B10 cells and 0.90 ± 0.3 µM (3 experiments) in Jurkat cells (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   An ATP-regenerating system abolishes agonistic actions of ATP in Jurkat -and B10 cells. Dose-response curves for the actions of ATP (, ) and ADP (, ) in Jurkat (A) and B10 (B) cells. Actions of ATP were analyzed in the absence () or presence () of CP and CPK. Actions of ADP were analyzed in the absence () or presence () of creatine and CPK. ATP solutions (1 mM) were treated for 10 min with 10 mM CP and 20 units/ml CPK. ADP solutions (1 mM) were treated for 10 min with 10 mM creatine and 20 units/ml CPK. Appropriate dilutions were added to the cells, and [Ca2+]i was measured after 15 sec. Data are from a single experiments with values determined in triplicate and are representative of the results of three similar experiments.

ATP inhibits ADP-induced intracellular Ca²⁺ mobilization. In B10 cells, ATP (0.1 mM) inhibited the action of ADP by shifting the dose-response curve to higher concentrations (Fig. 2A). ATP did not, however, modify the maximum efficacy of ADP. EC₅₀ values for ADP were 0.88 ± 0.12, 6.4 ± 0.5, and 49 ± 9 µM in the presence of no ATP, 100 µM ATP, and 1 mM ATP, respectively. The K_i value for ATP estimated from these data was 20 µM. ATP (100 µM) also inhibited the effects of 2-MeSADP on [Ca²⁺]_i in B10 cells, raising the EC₅₀ value of this agonist from 19 ± 3 to 93 ± 25 nM. This shift corresponded to a K_i value for ATP of 25 µM. Finally, Fig. 2A (inset) shows the dose-response curve for the inhibition by ATP of responses induced in B10 cells by 300 nM ADP. In seven independent experiments, the K_i value for ATP was estimated to be 23.8 ± 3.3 µM. The mean K_i value for ATP estimated from these three types of experiments was 23.0 ± 1.5 µM.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   ATP inhibits ADP-induced [Ca2+]i responses. A, Dose-response curves for the action of ADP on B10 cells in the absence of ATP () and in the presence of 0.1 mM purified ATP (). Inset, dose-response curve for the inhibition by purified ATP of ADP (0.3 µM) responses in B10 cells. B, Dose-response curves for the action of ADP on Jurkat cells in the absence of ATP () and in the presence of 0.1 mM purified ATP (). Inset, plot of EC50 values for the action of ADP on Jurkat cells as a function of ATP concentration. EC50 values for ADP actions were 100 ± 20 nM (control), 230 ± 30 nM (10 µM ATP), 300 ± 40 nM (30 µM ATP), 530 ± 80 nM (50 µM ATP), and 850 ± 70 nM (100 µM ATP). The slope corresponded to a Ki value for ATP of 14 µM. ATP solutions were treated for 10 min with 10 mM CP and 20 units/ml CPK. After the addition of 10 mM iodoacetamide, the mixture was added to the cells. [Ca2+]i was measured 15 sec after the addition of ADP.

2-MeSATP inhibits ADP-induced intracellular Ca²⁺ mobilization. Fig. 3 (inset) shows the HPLC profile of a freshly prepared 2-MeSATP solution that contained >10% of a major contaminant (peak 1). This contaminant had the same retention time as 2-MeSADP. It induced platelet aggregation and was identified as 2-MeSADP by mass spectrometry analysis. Fig. 3 further shows that a 90-min treatment of a 2-MeSATP solution with 20 units/ml CPK and 10 mM CP almost completely removed contaminating 2-MeSADP from the solution. The biological activity of purified 2-MeSATP was checked in a platelet aggregation assay. It is well known that although 2-MeSADP is a potent agonist of platelet aggregation, 2-MeSATP is an antagonist (Hall and Hourani, 1993). In Fig. 3, it can be seen that purification of 2-MeSATP solutions with CP/CPK suppressed the aggregative effect of 2-MeSATP (traces a and b, left) and further that as expected 50 µM purified 2-MeSATP abolished the action of 5 µM ADP (traces a and b, right).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Purification of 2-MeSATP solutions with CP/CPK. Left: trace a, human platelet aggregation was induced by unpurified 100 µM 2-MeSATP. Trace b, purified 2-MeSATP did not induce platelet aggregation. Middle, HPLC profiles of unpurified 2-MeSATP (top inset) and of 2-MeSATP that had been treated for 90 min with 20 units/ml CPK and 10 mM CP (bottom inset). Peak 1, 2-MeSADP. Peak 2, 2-MeSATP. Right, aggregation induced by 5 µM ADP (trace a) was prevented in the presence of 50 µM purified 2-MeSATP (trace b).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   2-MeSATP does not activate P2Y1 receptors in Jurkat and B10 cells. Dose-response curves for the actions of 2-MeSATP in Jurkat (A) and B10 (B) cells. Experiments were performed using unpurified 2-MeSATP () and purified 2-MeSATP (). 2-MeSATP solutions (1 mM) were treated for 90 min with 10 mM CP and 20 units/ml CPK. Appropriate dilutions then were added to the cells, and [Ca2+]i was measured after 15 sec. Data are from a single experiment with values determined in triplicate and are representative of the results of three similar experiments.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   2-MeSATP inhibits ADP-induced [Ca2+]i responses. A, Dose-response curves for the action of ADP on B10 cells in the absence of 2-MeSATP () and in the presence of 0.1 mM purified 2-MeSATP (). Inset, dose-response curve for the inhibition by purified 2-MeSATP of ADP (1 µM) responses in B10 cells. B, Dose-response curves for the action of ADP on Jurkat cells in the absence of 2-MeSATP () and in the presence of 0.1 mM purified 2-MeSATP (). Inset, plot of EC50 values for the action of ADP on Jurkat cells as a function of 2-MeSATP concentration. EC50 values for the action of ADP were 98 ± 12 nM (control), 279 ± 72 nM (10 µM 2-MeSATP), 332 ± 41 nM (30 µM 2-MeSATP), 554 ± 81 nM (50 µM 2-MeSATP), and 2.1 ± 0.3 µM (100 µM 2-MeSATP). The slope corresponded to a Ki value for 2-MeSATP of 6.5 µM. 2-MeSATP solutions were treated for 90 min with 10 mM CP and 20 units/ml CPK. After the addition of 10 mM iodoacetamide, the mixture was added to the cells. [Ca2+]i was measured 15 sec after the addition of ADP.

2-ClATP inhibits ADP-induced intracellular Ca²⁺ mobilization. Fig. 6 (left, inset) shows the HPLC profile of a freshly prepared 2-ClATP solution that contained >10% of a major contaminant (peak 1). This contaminant induced platelet aggregation and was identified as 2-ClADP in mass spectrometry experiments. It was completely removed by 30-min treatment of the solution with 20 units/ml CPK and 10 mM CP (Fig. 6, inset). Purified 2-ClATP did not induce platelet aggregation (Fig. 6, left) and inhibited the action of ADP (Fig. 6, right), in accordance with the known antagonism of platelet ADP receptors by 2-ClATP (Hall and Hourani, 1993).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Purification of 2-ClATP solutions with CP/CPK. Left: trace a, human platelet aggregation was induced by unpurified 100 µM 2-ClATP. Trace b, purified 2-ClATP did not induce platelet aggregation. Middle, HPLC profiles of unpurified 2-ClATP (top inset) and of 2-ClATP that had been treated for 60 min with 20 U·ml CPK and 10 mM CP (bottom inset). Peak 1, 2-ClADP. Peak 2, 2-ClATP. Right, aggregation induced by 5 µM ADP (trace a) was prevented in the presence of 50 µM purified 2-ClATP (trace b).

View larger version (14K): [in this window] [in a new window]   Fig. 7.   2-ClATP does not activate P2Y1 receptors in Jurkat and B10 cells. Dose-response curves for the actions of 2-ClATP in Jurkat (A) and B10 (B) cells. Experiments were performed using unpurified 2-ClATP () and purified 2-ClATP (). 2-ClATP solutions (1 mM) were treated for 30 min with 10 mM CP and 20 units/ml CPK. Appropriate dilutions were added to the cells, and [Ca2+]i was measured after 15 sec. Data are from a single experiment with values measured in triplicate and are representative of the results of three similar experiments.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   2-ClATP inhibits ADP-induced [Ca2+]i responses. A, Dose-response curves for the action of ADP on B10 cells in the absence of 2-ClATP () and in the presence of 0.1 mM purified 2-ClATP (). Inset, dose-response curve for the inhibition by purified 2-ClATP of ADP (1 µM) responses in B10 cells. B, Dose-response curves for the action of ADP on Jurkat cells in the absence of 2-ClATP () and in the presence of 0.1 mM purified 2-ClATP (). Inset, plot of EC50 values for the action of ADP on Jurkat cells as a function of 2-ClATP concentration. EC50 values for ADP actions were 139 ± 60 nM (control), 376 ± 37 nM (10 µM 2-ClATP), 1.09 ± 0.27 µM (30 µM 2-ClATP), 1.78 ± 0.16 µM (50 µM 2-ClATP), and 3.08 ± 0.64 µM (100 µM 2-ClATP). The slope corresponded to a Ki value for 2-ClATP of 2.5 µM. 2-ClATP solutions were treated for 30 min with 10 mM CP and 20 units/ml CPK. After the addition of 10 mM iodoacetamide, the mixture was added to the cells. [Ca2+]i was measured 15 sec after the addition of ADP.

EDTA potentiates the effects of ADP and ATP. In physiological buffers, ATP and ADP exist as mixtures of several species, including forms that are complexed with monovalent and divalent cations and free ADP³ and ATP⁴ forms. The species constituting the preferential ligand of a receptor can be determined by investigating the effects of divalent cations on the action of ADP or ATP. Fig. 9A compares dose-response curves for the action of ADP on [Ca²⁺]_i in B10 cells in experiments performed in the presence or absence of 5 mM EDTA. The addition of EDTA decreased the EC₅₀ value for ADP action from 0.86 ± 0.08 µM (7 experiments) to 0.35 ± 0.03 µM (4 experiments). Thus, chelation of extracellular divalent cations with EDTA increased by 2.4-fold the potency of ADP. EDTA also potentiated the inhibitory action of ATP on ADP-induced [Ca²⁺]_i increase in B10 cells (Fig. 9B). The mean K_i value for ATP action decreased from 23.8 ± 3.3 µM (7 experiments) in the absence of EDTA to 6.5 ± 1.3 µM (13 experiments) in the presence of 5 mM EDTA. These values were calculated by using EC₅₀ values for ADP actions measured in the presence of EDTA. EDTA therefore increased 3.6-fold the apparent affinity of ATP for the P2Y₁ receptor of B10 cells.

View larger version (16K): [in this window] [in a new window]   Fig. 9.   EDTA potentiates the effects of ADP and ATP on B10 cells. A, Dose-response curves for the action of ADP on B10 cells in the absence of EDTA () and in the presence of 5 mM EDTA (). B, Dose-response curves for the inhibition by ATP of ADP (0.3 µM) responses in B10 cells. Experiments were performed in the absence of EDTA () or in the presence () of 5 mM EDTA. Experiments were performed as described in the legends of Figs. 1 and 2.

	Discussion

Top Summary Introduction Procedures Results Discussion References

ATP, 2-ClATP, and 2-MeSATP usually are considered to be agonists of P2Y₁ receptors (Boyer et al., 1993; Vigne et al., 1994; Boyer et al., 1996; Burnstock and King, 1996; Schachter et al., 1996). In the current study, the pharmacological properties of rat and human P2Y₁ receptors were studied under conditions in which care was taken to purify the agonist solutions of contaminating ADP or ADP derivatives by treatment with CP/CPK. The agonistic action of ATP, 2-MeSATP, and 2-ClATP was suppressed in the presence of this ATP-regenerating system. Any possibility that CP/CPK inhibited purinergic responses through a nonspecific mechanism was ruled out by the following observations: (1) CPK or CP alone was inactive, (2) creatine/CPK did not modify ADP responses, and (3) iodoacetamide-inactivated CPK and CP did not alter ADP responses. These findings suggested that the previously reported agonistic effects of ATP and its derivatives were most likely due to contaminating ADP or ADP derivatives, present in the unpurified agonist solutions or produced at the cell surface by ectoenzymes, or both. Thus, the human P2Y₁ receptor expressed in Jurkat cells and the endogenous P2Y₁ receptor of B10 cells appear to be ADP-specific receptors. This, together with a previous report (Nicholas et al., 1996), points to a remarkable specificity of P2Y receptors. Mammalian P2Y receptors can be classified as an ADP-specific receptor (P2Y₁ receptor), a UDP-specific receptor (P2Y₆ receptor), a UTP-selective receptor (P2Y₄ receptor), or a mixed ATP/UTP receptor (P2Y₂ receptor) (Nicholas et al., 1996).

This study also showed that purified solutions of ATP, 2-ClATP, or 2-MeSATP shifted ADP dose-response curves to higher concentrations in both B10 cells and P2Y₁ receptor-transfected Jurkat cells. The maximum efficacy of ADP was not modified, indicating that ATP and its derivatives are competitive antagonists of ADP at P2Y₁ receptors. K_i values were estimated as 23 µM for ATP, 31 µM for 2-MeSATP, and 27 µM for 2-ClATP for the rat P2Y₁ receptor and as 5.6 µM for 2-MeSATP, 2.3 µM for 2-ClATP, and 15 µM for ATP for the human receptor. Hence, ATP, 2-ClATP, and 2-MeSATP are weak antagonists of P2Y₁ receptors, whereas ADP (K_app = 0.2 µM in Jurkat cells and 0.9 µM in B10 cells) and 2-MeSADP (K_app = 49 nM in Jurkat cells and 19 nM in B10 cells) are potent agonists.

Different results were reported by Schachter et al.(1996), who reported full agonistic activity of 2-MeSADP, 2-MeSATP, ADP, and ATP. This pharmacological profile resulted from measurement of inositol phosphate production after 10-min incubation of the human P2Y₁ receptor-transfected 1321N1 cells with nonpurified nucleotides. These experimental conditions, under which metabolic transformation of the nucleotides could occur, may explain the discrepancies with our results. More recently, at the "Purine and Their Receptors" meeting in New Orleans, Harden et al. reported a new pharmacological profile, with 2-MeSADP and ADP being the strongest agonists and 2-MeSATP and ATP being the weakest. In their study, [Ca²⁺]_i increase was measured on few cells adhering to a coverslip perfused with HPLC-purified nucleotides but in the absence of an ATP-regenerating system. These new data are fully consistent with the results presented in the current report.

The potencies of ADP and ATP at the rat P2Y₁ receptor were increased 2.4- and 3.6-fold, respectively, when extracellular divalent cations were chelated with EDTA (Fig. 6). This could imply that the P2Y₁ receptor is more responsive to charged forms of nucleotides than to forms that are complexed with divalent cations. However, the situation probably is not so simple because a larger shift in affinity for ATP would be expected if only ATP⁴ were acting on receptors; although ATP⁴ is the preferred ligand of the P2Y₁ receptor, it also may be recognized by other forms of ATP.

Despite spectacular advances in recent years in the fields of ionotropic P2X and metabotropic P2Y receptors, the putative platelet-specific ADP receptor (P2T), which was the first purinoceptor to be defined pharmacologically, has not yet been cloned (Hourani and Hall, 1996; Mills, 1996; Gachet et al., 1997). In a previous publication, we suggested that the P2Y₁ receptor could be this elusive ADP receptor of platelets (Léon et al., 1997). The results presented here provide further evidence of close similarities in the pharmacological profiles of the P2Y₁ receptor and the platelet ADP receptor.

First, ATP, 2-ClATP, and 2-MeSATP are antagonists of the ADP receptor of platelets (Hall and Hourani, 1993) and of the rat and human P2Y₁ receptors. The K_i value of ATP for the human P2Y₁ receptor in Jurkat cells (16 µM) is similar to that reported for the ADP receptor of platelets (10 µM) (Hall and Hourani, 1993). The K_i value of 2-ClATP in Jurkat cells (2.3 µM) likewise is close to that reported for the platelet ADP receptor (2.5 µM) (Hall and Hourani, 1993). The only difference between the human P2Y₁ receptor in Jurkat cells and the platelet ADP receptor relates to 2-MeSATP. 2-MeSATP is a competitive antagonist of ADP actions in Jurkat cells (Fig. 4B), whereas it was found to be an insurmountable antagonist of ADP actions in platelets (Hall and Hourani, 1993).

Second, ADP³ is a 2.4 times more potent activator of the rat P2Y₁ receptor than ADP (Fig. 6A) and a 1.9 times more potent activator of the ADP receptor of platelets (Hall et al., 1994). Conversely, chelation of divalent cations with EDTA increased by 3.6-fold the capacity of ATP to inhibit the rat P2Y₁ receptor, whereas the ability of ATP to inhibit the platelet ADP receptor increased 10-fold when experiments were performed in the absence of divalent cations (Hall et al., 1994).

On the other hand, the only form of P2Y receptor detected to date in platelets is the P2Y₁ receptor (Léon et al., 1997), as also is true in B10 cells (Webb et al., 1996). Another striking similarity is the fact that although ADP induces the mobilization of intracellular Ca²⁺ stores in platelets and B10 cells, in both cases inositol-1,4,5-trisphosphate seems to be only poorly involved in this response (Feolde et al., 1995; Hourani and Hall, 1996; Mills, 1996; Gachet et al., 1997).

In summary, our results and data from the literature strongly suggest that platelets and endothelial cells share a common P2Y₁ receptor that mediates platelet aggregation and vasodilation. Whether P2T receptors and P2Y₁ receptors are encoded by the same gene remains to be established through knock-out experiments.