Lack of evidence for functional ADP-activated human P2X₁ receptors supports a role for ATP during hemostasis and thrombosis

Purine nucleotides acting through P2 receptors play key roles in platelet signaling and hemostasis. P2 receptors can be subdivided into P2X ionotropic and P2Y metabotropic receptors, and transgenic mice models have now firmly established roles of P2X₁, P2Y₁, and P2Y₁₂ receptors in platelet function.^1-6  P2X₁ receptors for adenosine triphosphate (ATP) mediate rapid transient increases in calcium concentration that can synergize with the larger, more sustained response through the adenosine diphosphate (ADP)-dependent P2Y₁ receptor, leading to speeding and amplification of the calcium increase.⁴  In addition, P2X₁ receptor-deficient mice have recently been shown to have a decreased aggregation response and to exhibit protection against thromboembolism.⁷  A major problem that has complicated the understanding and characterization of P2 receptors is the use of agonists to define P2 receptors because impurity or interconversion of the agonist has led to the pharmacologic misclassification of receptors. For example uridine diphosphate (UDP) appears as a weak agonist at the uridine triphosphate (UTP)-sensitive P2Y₄ receptor; however, when contaminating UTP is removed with hexokinase, agonist activity is essentially abolished.⁸  The same problem has also influenced the study of P2 receptors in platelets and characterization of the specificity of action of different purines at individual receptors. For the P2X₁ receptor we have shown that the apparent activity of ADP at the receptor can be accounted for by ATP contamination of commercially available ADP.^9,10  This has important implications for the physiologic activation of platelet purinoceptors because ATP acts as a competitive antagonist rather than an agonist at platelet P2Y₁ receptors as a result of their low level of expression.^11,12  Thus, in platelets ATP and ADP are selective stimuli at P2X₁ and P2Y₁ receptors, respectively, and synergy between these 2 receptors may be important during hemostasis. However, the inability of ADP to activate platelet P2X receptors has been questioned by a study of a deletion variant, P2X_1del. This variant, cloned from platelets and megakaryocytic cell lines, is deficient in 17 amino acids in the extracellular ligand-binding loop and has been reported to act as an ADP receptor in calcium measurement studies.¹³  Significant controversy exists as to whether the P2X_1del receptor is expressed at high levels in platelets and exerts a functional role.^4,14,15 

In this study we have (1) characterized the expression and properties of recombinantly expressed P2X_1del receptors, both as homomeric P2X_1del receptors and in heteromeric assemblies with full-length P2X₁ receptors; (2) determined in human platelets the relative concentrations of expression of P2X₁ and P2X_1del receptors; and (3) investigated whether P2X_1del receptors contribute to generating P2X receptor-mediated signals. We provide conclusive evidence that platelet P2X_1del receptors are not functional in cell lines or platelets and that ATP and ADP separately stimulate the native platelet P2X and P2Y receptors during hemostasis.

Human embryonic kidney 293 (HEK293) cells were maintained in minimal essential medium with Earle salts (with GlutaMAX I) supplemented with 10% fetal bovine serum, 1% nonessential amino acid (Invitrogen, Paisley, United Kingdom) at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. Native 1321-N1 astrocytoma cells (a gift from Dr T.E. Webb, De Montfort University, Leicester, United Kingdom) and 1321-N1 cells subcloned after transfection of either wild-type (WT) P2X₁ or P2X_1del (clone 4) (1321-N1 G P2X₁WT or 1321-N1 G4 P2X_1del, respectively; kindly provided by Dr N.J. Greco, Red Cross Research Laboratories, Rockville, MD) were maintained in Dulbecco modified Eagle medium (with GlutaMAX I, 4500 mg/L D-glucose) supplemented with 10% fetal bovine serum (Invitrogen). Stably transfected 1321-N1 cells were maintained under permanent selection in 240 μg/mL G418 (Invitrogen). HEK293 and 1321-N1 cells were transiently transfected with P2X₁WT, P2X_1del, or both receptors using LipofectAMINE 2000 Reagent/Opti-MEM (both from Invitrogen). For patch-clamp studies, as a control to identify cells that were efficiently transfected, the pEGFP-N1 (Clontech, Palo Alto, CA) vector (10% of total plasmid transfected) was cotransfected with P2X₁ plasmids, and recordings were made only from fluorescent cells expressing the pEGFP. After 24 hours, the cells were either lysed, for Western blotting or were attached to glass coverslips and kept in culture for a maximum of 3 days for use in confocal microscopy and patch-clamp experiments.

Western blots. Human platelets from 13 healthy donors were prepared as described¹⁶  and then pelleted at 10 000 rpm for 10 minutes. Platelets, human embryonic kidney 293 (HEK293), and 1321-N1 cells were homogenized in 20 mM Tris (tris[hydroxymethyl]aminomethane)-HCl (pH 8), 250 mM NaCl, 3 mM EDTA (ethylenediaminetetraacetic acid), 3 mM EGTA (ethylene glycol-bis (b-aminoethylether)-N,N,N′,N′-tetraacetic acid), and 0.5% Triton X-100 complemented with protease inhibitor cocktail (1:100) (Sigma-Aldrich, Poole, United Kingdom). Samples were separated under nonreducing conditions on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels. After electrophoretic transfer to nitrocellulose (100 V, 1 hour), the membrane was blocked overnight in TTBS (20 mM Tris-HCl [pH 7.6], 145 mM NaCl, 0.05% Tween 20) and 10% dry skim milk. The membrane was then incubated with anti-P2X₁ antibody (1:1000) (Alomone, Jerusalem, Israel) in TTBS and 10% dry skim milk for 2 hours, washed in TTBS, and incubated for 2 hours in antirabbit horseradish peroxidase secondary antibody (1:1500 dilution) (Sigma). After washing, the visualization of the protein bands was achieved with an ECL (Plus) kit (Amersham Biosciences, Little Chalfont, United Kingdom) according to the manufacturer's instructions.

Surface cell expression and immunoprecipitations. All cell surface proteins were biotinylated with Sulfo-NHS-LC-Biotin (0.5 mg/mL in phosphate-buffered saline (PBS) (Pierce, Perbio Science, Tattenhall, United Kingdom) for 30 minutes. After washing with PBS, cells were homogenized in RIPA buffer (10 mM Tris-HCl [pH 7.4], 2 mM EDTA, 160 mM NaCl, 1% Nonidet P-40, and 0.5% deoxycholic acid) for 10 minutes on ice and centrifuged at 13 000 rpm for 5 minutes. The supernatant was incubated in the presence of 300 μL anti-P2X₁ antibody (Alomone) diluted at 1: 200 in TE (10 mM Tris-HCl, 2 mM EDTA [pH 7.4]) for 2 hours on ice. Protein A Sepharose CL-4B (180 μL of 3% gel, reconstituted in 10 mM Tris-HCl, 2 mM EDTA [pH 7.4]) (Amersham Biosciences) was added, and the samples were homogenized for 15 minutes at 4°C, washed successively in RIPA buffer and TE, and resuspended in the gel sample buffer (180 mM Tris-HCl [pH 6.8], 5.7% SDS, 29% glycerol) and heated at 60°C for 3 minutes. The samples were separated on SDS-PAGE gel and were transferred onto nitrocellulose membrane as described, and the membrane was blocked overnight in TTBS and 3% bovine serum albumin (BSA). The membrane was then incubated with Immunopure streptavidin/horseradish peroxidase conjugated (0.5 μg/mL) (Pierce) in TTBS + 3% BSA for 30 minutes and washed; the visualization of the protein bands was achieved as described above.

Conventional whole-cell, patch-clamp experiments were performed at a holding potential of -60 mV at room temperature, as described previously.⁴  Agonists were rapidly applied through a U-tube.

From cell lines. 1321-N1 cells (1321-N1 G P2X₁WT or 1321-N1 G4 P2X_1del), attached to glass coverslips, were incubated with 1 to 3 μM fluo-3-AM for 30 to 45 minutes, then washed gently and added to the imaging chamber. The standard bath saline contained (in mM) 150 NaCl, 2.5 KCl, 1 MgCl₂, 2.5 CaCl₂, 10 HEPES (N-2-hydroxethylpiperazine-N'-2-ethanesulfonic acid), pH 7.3 (NaOH); CaCl₂ was omitted for experiments in nominally Ca²⁺-free saline. Fluorescence recordings (488 nm excitation, greater than 505-nm emission) were made from fields of approximately 20 to 30 cells using a Zeiss LSM 510 confocal microscope or an Olympus Fluoview FV 300 (Solent Scientific, Portsmouth, United Kingdom). Fluorescent signals were background-subtracted, and F/F₀ ratios were used to normalize fluorescence levels (F) against starting fluorescence (F₀). Figures show the Ca²⁺ responses from individual cells representative of 15 to 30 cells within the field of view. ATP and ADP were applied either through a U-tube system or by perfusion of the entire chamber. Thapsigargin (Calbiochem, Meik Biosciences, Nottingham, United Kingdom) was added by dilution directly to the chamber.

Human platelets. Blood was taken with informed consent from 13 donors, and fura-2 ratiometric fluorescence measurements of [Ca²⁺]_i in stirred, washed platelet suspensions were performed as described previously.¹⁷  Platelet saline contained (in mM) 145 NaCl, 5 KCl, 1 CaCl₂, 1 MgCl₂, 10 HEPES, 10 D-glucose, pH 7.35 (NaOH). The cuvette temperature was lowered to 13°C to slow the start of the P2Y receptor-dependent Ca²⁺ response and to more clearly distinguish P2X₁-evoked Ca²⁺ increases.⁹  The background-corrected F₃₄₀/F₃₈₀ fluorescence ratio was used as a direct indication of [Ca²⁺]_i to avoid errors in application of the standard calibration methods in platelets at this temperature (for a discussion, see Mahaut-Smith et al⁹ ). Agonists were added after insertion of Hamilton syringes into a custom-built holder to avoid artefacts introduced by opening the cuvette lid, and additions were marked electronically to allow comparison of timing between different experiments.

Data are presented throughout as mean ± SEM; n represents the number of observations. Differences between means were determined by the appropriate Student t test and were considered significant when P < .05.

αβ-Methylene ATP (αβ-meATP), ATP, and ADP were all obtained from Sigma. ADP was treated with hexokinase in a high-glucose-containing saline at pH 8, as described previously,⁹  to remove contaminating levels of ATP. Fluo-3-AM and fura-2-AM were from Molecular Probes (Leiden, the Netherlands) and made as stocks of 1 mM in 20% (wt/vol) Pluronic in dimethyl sulfoxide (DMSO).

ATP (100 μM) is a full agonist at P2X₁ WT receptors and evokes rapidly desensitizing inward currents through these receptors transiently expressed in HEK293 cells (mean amplitude, 4629 ± 700 pA; n = 18) (Figure 1A). Commercially available ADP (100 μM) evoked an inward current (mean amplitude, 1514 ± 502 pA; n = 7; data not shown); however, after removing contaminating ATP with hexokinase, the response to ADP was reduced by more than 95%, to -71 ± 44 pA (n = 10; Figure 1A). We failed to detect any ATP or ADP (both 100 μM) receptor-activated currents in HEK293 cells transfected with the P2X_1del receptor (n = 14, 18) (Figure 1B). Total expression levels of P2X₁WT and P2X_1del were determined using Western blot analysis with an anti-P2X₁ receptor antibody. The anti-P2X₁ receptor antibody identified an approximately 50-kDa band in cells transfected with the P2X₁WT plasmid and an approximately 46-kDa band in cells transfected with the P2X_1del plasmid (Figure 1D) (in addition, a nonspecific band of approximately 60 kDa was present in nontransfected HEK293 cells, but not in 1321-N1 cells; see Figure 2B). The total level of expression of the P2X_1del was approximately 20% compared with the P2X₁WT. Cell surface biotinylation studies demonstrated that the P2X₁WT receptor was trafficked to the cell surface; however, surface expression of P2X_1del receptors was below the limit of detection (Figure 1D). These results indicate that the lack of agonist-evoked membrane current in cells transfected with P2X_1del receptors is at least in part caused by poor surface expression of the receptor.

Heteromeric assembly of different P2X receptor subunits (eg, P2X₂ and P2X₃¹⁸ ) or WT and mutant P2X₁ receptor subunits can give rise to channels with altered properties.¹⁹  Both P2X₁ WT and P2X_1del receptor RNAs are made by platelets and a range of megakaryocytic cells,¹³  which raises the possibility that heteromeric channels may be formed with novel properties. We therefore cotransfected P2X₁WT and P2X_1del receptors in HEK293 cells with a DNA ratio of 1:9 to determine whether we could detect any regulation of the properties of P2X₁WT receptors. Agonist-evoked currents from cells cotransfected with P2X₁WT and P2X₁del were indistinguishable from those of WT P2X₁ receptors—that is, there was a rapidly desensitizing response to ATP, whereas ADP was ineffective as an agonist (Figure 1C; n = 15, 10) (similar results were found when the receptors were transiently cotransfected in 1321-N1 cells, ratio P2X₁WT/P2X_1del 1:9; n = 10; for Western blot see Figure 2C). Western blot analysis indicated that the total level of P2X₁WT and P2X_1del receptors was similar in HEK293 cells (Figure 1D). However, cell surface biotinylation again showed that only P2X₁WT was detected at the cell surface (Figure 1D).

In the study by Greco et al,¹³  P2X_1del receptors were stably expressed in the 1321-N1 cell line, and changes in calcium levels were monitored from populations of cells. Because Ca²⁺ increases evoked by ATP and ADP can result from both P2X and P2Y receptors, we used patch-clamp recordings to detect P2X receptor currents in individual cells from the lines used in the original study by Greco et al.¹³  Applying ATP or ADP to 1321-N1 G P2X₁ WT or 1321-N1 G4 P2X_1del cell lines gave no change in holding current, demonstrating that these cells do not express functional P2X receptors (n = 19, 8, and 15, 15 respectively) (Figure 2A). This was consistent with Western blotting results in which both P2X₁ WT and P2X_1del receptors were below the limit of detection in these cell lines (Figure 2B).

When P2X₁WT receptors were transiently transfected into a 1321-N1 cell background, ATP (100 μM) evoked inward currents (in either the 1321-N1 G P2X₁WT or the 1321-N1 G4 P2X_1del mean amplitude, -1674 ± 470 pA; n = 19) (Figure 2A), demonstrating that P2X₁WT receptors can form functional channels in this cell line as described previously.²⁰  As shown for the recombinant P2X₁ receptors expressed in HEK293 cells, purified ADP was ineffective as an agonist at activating P2X₁WT receptor-mediated inward currents in 1321-N1 cells (Figure 2Aiii, -32 ± 12 pA; n = 9). No inward current was evoked by either ATP or ADP (100 μM) when applied to native 1321-N1 cells transiently expressing P2X_1del receptors (data not shown; n = 8, 9). Western blotting was used to estimate the total levels of expression of transiently transfected P2X₁WT and P2X_1del receptors in 1321-N1 cells (Figure 2C). The anti-P2X₁ receptor antibody detected bands in total cellular lysates corresponding to the P2X₁WT and P2X_1del receptors. These results confirm that the P2X_1del receptor can be expressed but that it fails to form functional ion channels in 1321-N1 cells.

Even though we could not detect P2X₁ receptor current or protein in the cell lines provided by Greco et al,¹³  we were able to detect rapid increases in intracellular calcium in response to ATP and ADP (Figure 3A) in 50% and 67% of cells tested (70 cells from 3 different studies) (Figure 3A). These results can be explained by the presence of native P2Y receptors in the subclonal 1321-N1 G P2X₁WT and 1321-N1 G4 P2X_1del cell lines. The standard approach to assess the relative contribution of P2Y versus P2X receptors is to examine the Ca²⁺ response in the absence of external Ca²⁺. P2X receptors rely entirely on Ca²⁺ influx, whereas the response to P2Y receptors was at least partly caused by inositol 1, 4,5-triphosphate (IP₃)-dependent Ca²⁺ release. However, this proved difficult in 1321-N1 cells because of the rapid depletion of the intracellular Ca²⁺ stores in Ca²⁺-free medium. Thus, after only 2 to 3 minutes in nominally Ca²⁺-free saline, ADP failed to evoke a response, though this resulted from the depletion of the internal stores because the subsequent addition of the endomembrane Ca-ATPase inhibitor Thapsigargin had no effect on the cytosolic Ca²⁺ level (Figure 3B). The extremely rapid depletion of Ca²⁺ stores in Ca²⁺-free medium in this cell line highlights the difficulty of using this condition as a definitive test for P2X or P2Y receptors. In Ca²⁺-containing medium, when the stores are replenished after agonist exposure, Thapsigargin causes its expected effect of a substantial increase in cytosolic Ca²⁺ (Figure 3C). However, ADP did evoke a small Ca²⁺ response after shorter exposures to Ca²⁺-free medium (for example, 60 seconds; see Figure 3D) in both 1321-N1 G P2X₁WT and 1321-N1 G4 P2X_{1 del}, confirming the presence of P2Y responses in these cell lines.

The lack of functional expression of recombinant P2X_1del receptors may reflect the absence of some accessory factor necessary for efficient trafficking/processing of the receptor in HEK293 and 1321-N1 cell lines. Given that the P2X_1del receptor was originally isolated from platelets, we determined the levels of P2X_1del receptor expression in human platelets and the functional role of P2X and P2Y receptor-mediated calcium signaling. In Western blot analysis of platelet lysates (from 13 donors) the anti-P2X₁ receptor antibody detected a single band at approximately 50 kDa, corresponding to the P2X₁WT receptor, but no lower molecular weight band corresponding to the P2X_1del receptor was detected (Figure 4A). To estimate the detection threshold for the anti-P2X₁ antibody, we diluted the amount of platelet sample run on the gel. P2X₁WT receptors could still be detected at a 1:24 dilution (Figure 4B). Assuming that the antibody has the same sensitivity for P2X₁WT and P2X_1del (which is valid given that the deletion is in the extracellular portion of the receptor and the 2 variants share the identical epitope at the C-terminal), this indicates that the P2X_1del receptor accounts for less than 4% of the total protein. Given the additional reduction in efficiency of transfer to the cell membrane, this indicates that the P2X_1del accounts for a small proportion, if any, of the cell surface P2X₁ receptors. Studies of P2 receptor Ca²⁺ responses were also conducted in platelets from 13 donors, including 6 of those used for the Western blot analyses of Figure 4A. The temperature was lowered to 13°C to clearly distinguish rapid P2X₁ increases from the slower P2Y₁ response.⁹  In all donors tested (n = 13), the P2X₁ receptor agonist αβ-meATP evoked a rapid transient calcium increase (Figure 4C), whereas the P2Y₁ receptor agonist, ADP (40 μM, hexokinase purified) evoked Ca²⁺ increases with a prolonged delay of 2 to 3 seconds. The lack of evidence for an ADP-induced P2X receptor-mediated rise in calcium demonstrates that ADP-dependent P2X_1del receptors do not contribute to calcium signaling in human platelets.

In the present study we have investigated the expression and function of the P2X_1del receptor because it has been suggested that this receptor may be sensitive to ADP and may play a functional role in the activation of platelets. We have demonstrated that the P2X_1del receptor fails to form functional channels when expressed in HEK293 and 1321-N1 cell lines and that this most likely results from poor trafficking of the receptor to the cell surface. In addition, in human platelets, P2X_1del protein was below the limit of detection. These results demonstrate that the P2X_1del receptor does not play a significant role in platelet function and that the P2X₁ receptor in human platelets is activated by ATP not ADP.

P2X₁WT receptors were transiently expressed in HEK293 and 1321-N1 cells as a positive control. Activation of this channel by ATP evoked rapidly inactivating responses, and purified ADP was ineffective as an agonist, confirming results described by several independent laboratories.^{9,10,18,20-24}  It is anomalous that in the study of Greco et al,¹³  purified ADP was found to be an equi-effective agonist at the P2X₁WT receptor. Even when commercially available ADP (impure) has been used, it only acts as a weak partial agonist with more than 100-fold reduced potency compared with ATP in electrophysiologic^9,21  or calcium-imaging studies.²⁰  In the study by Greco et al,¹³  ADP evoked an intracellular calcium increase of amplitude comparable with that of ATP when these agents were added sequentially, with an interval of only 10 to 20 seconds in the P2X₁WT stable cell line (designated 1321-N1 G P2X₁ WT in this study).¹³  This is unusual because native and recombinant P2X₁ receptors show profound desensitization to agonist application, such that long wash/recovery periods between applications are required to generate reproducible responses. Furthermore, agonist desensitization is used routinely to remove “contaminating” P2X₁ receptor-mediated responses in pharmacologic studies.^18,25  For example, the role of native P2X₁ receptors in HL60 cells and platelets had been overlooked because of problems associated with receptor desensitization during cell preparation.^17,26  For these reasons the pharmacology of the P2X₁WT responses reported by Greco et al¹³  appear inconsistent with the consensus of those reported for wild-type P2X₁ receptors (for recent review, see North¹⁸ ).

The patch-clamp technique provides a highly sensitive measure of changes in membrane current and detection of functional ion channels. The single-channel amplitude of the P2X₁WT receptor under the recording conditions used is approximately 0.6 pA (R.J.E., unpublished observations, 2001), and signals of approximately 2 to 3 pA can be readily discriminated; therefore, theoretically the opening of as few as 5 channels can be detected. The P2X_1del receptor failed to produce functional ion channel currents when transiently expressed in either HEK293 or 1321-N1 cells. Using Western blot analysis (Figures 1, 2), we show that the P2X_1del receptor is produced in these cells and associates to form trimeric channels (data not shown), as for the wild type. However, the total P2X_1del receptor level was reduced compared with equivalent transfection of P2X₁WT (for a discussion, see also Oury et al¹⁴ ), indicating that the P2X_1del receptor is either inefficiently made or broken down. In addition, we showed the P2X_1del receptor is below the limit of detection at the cell surface, and this lack of receptor trafficking most likely accounts for the lack of functional P2X_1del receptor-mediated currents.

There was an approximately 3- to 4-kDa decrease in the molecular mass of the P2X_1del receptor monomer compared with the full-length P2X₁WT receptor, as described previously after in vitro translation.^13,14  This is to be expected because the deletion removes 17 amino acids that include a glycosylation site (N184, which contributes 2-3 kDa).²⁷  The poor surface expression of the P2X_1del receptor in the present study is unlikely to result solely from the removal of glycosylation at N184.^22,27  Interestingly, for the rP2X_1a deletion mutant, which has a larger deletion (amino acids 175-201 compared with 176-192 for P2X_1del), the coexpression of P2X₁WT can in part rescue cell surface channel expression in HEK293 cells.²²  This was not the case for the P2X_1del receptor. The reason the extra 10-amino acid deletion of rP2X_1a encourages cotrafficking to the cell surface remains to be determined.

We were unable to detect P2X₁ or P2X_1del receptor expression either electrophysiologically or by Western blotting using the stable cell lines expressing P2X₁ and P2X_1del receptors provided by Greco et al.¹³  When the P2X₁WT receptor was transiently transfected into these 1321-N1 stable cell lines, ATP evoked robust P2X receptor currents, and receptor protein of the appropriate molecular weight was detected, confirming that the cells can produce functional P2X₁WT receptors.²⁰  The fact that calcium responses to ADP and ATP can be recorded from these cells in the absence of detectable P2X₁ receptor currents or protein indicates that endogenous P2 receptors other than P2X₁ are expressed in these cells.¹³ 

Results following the expression of recombinant receptors indicate that P2X_1del receptors are not functional, though it is possible that HEK293 and 1321-N1 cell lines lack some accessory factor necessary for efficient trafficking/processing of these receptors. Previous estimates of relative levels of P2X₁WT and P2X_1del in platelets have been made based on reverse transcription-polymerase chain reaction (RT-PCR). In this study we measured protein levels directly with Western analysis and investigated the role of ADP-activated P2X₁ receptors in functional assays. Western blotting failed to detect the P2X_1del receptor protein in total platelet lysates from 13 donors (in assays that could detect a 24-fold dilution of the WT receptor). Given the likely additional inefficient trafficking to the cell surface, P2X_1del accounts for a small, if any, proportion of the P2X₁ receptors at the cell surface. Because the P2X_1del receptor failed to be expressed on the cell surface of HEK or 1321-N1 cells, it is still uncertain whether ADP is an agonist, as shown for the WT receptor. Nevertheless, functional studies failed to detect a rapid ADP-mediated P2X receptor Ca²⁺ response; thus, ADP-activated receptors, whether P2X_1del or other naturally occurring mutants, do not contribute to the P2X₁ receptor phenotype, as also shown in kinase assays.²⁴  Previous work has shown that the P2X₁ receptor in human platelets synergizes with functional responses through P2Y₁ and collagen receptors.^4,5  Thus, our present study supports the concept that ATP, released from damaged vascular cells or cosecreted from platelets with ADP and other agonists, is an agonist through P2X₁ receptors during hemostasis. A recent study using P2X₁^-/- mice has suggested that the main role of this receptor is to enhance platelet activation under conditions of high shear stress.⁷  This functional importance of P2X₁ receptors may be a consequence of the rapid nature of the ATP-activated Ca²⁺ influx or its ability to potentiate P2Y₁ and collagen-receptor signaling.^4,5 

In summary, we can find no evidence to support a functional role for an ADP-activated P2X_1del receptor in human platelets and conclude that ATP is the principal natural agonist at P2X receptors in platelets mediating a rapid increase in intracellular calcium. This calcium increase also synergizes with subsequent ADP-mediated P2Y receptor signaling events associated with hemostasis, and blockade of the ATP-sensitive P2X₁ receptor may provide a novel drug target for protection against thromboembolism.⁷ 

We thank Dr N. Greco for providing the P2X_1del plasmid and transfected 1321-N1 cell lines, Dr T. Webb for providing native 1321-N1 cells, ProfessorA. Goodall and Dr J.Appleby for providing samples of human platelets, and J. Holdich for technical assistance.