Alastair W. Poole, Department of Physiology and Pharmacology, School of Medical Sciences, University Walk, Bristol BS8 1TD, UK. Tel.: + 44 117 331 1435; fax: + 44 117 331 2288. E-mail: email@example.com.
Summary. Background: Adenosine diphosphate (ADP) is a critical regulator of platelet activation, mediating its actions through two G protein-coupled receptors (GPCRs), the P2Y1 and P2Y12 purinergic receptors. Recently, we demonstrated that both receptors desensitize and internalize in human platelets by differential kinase-dependent mechanisms. Objectives: To demonstrate whether responses to P2Y1 and P2Y12 purinergic receptors resensitize in human platelets and determine the role of receptor traffic in this process. Methods: These studies were undertaken either in human platelets or in cells stably expressing epitope-tagged P2Y1 and P2Y12 purinergic receptor constructs. Results: In this study we show for the first time that responses to both of these receptors can rapidly resensitize following agonist-dependent desensitization in human platelets. Further, we show that in human platelets or in 1321N1 cells stably expressing receptor constructs, the disruption of receptor internalization, dephosphorylation or subsequent receptor recycling is sufficient to block resensitization of purinergic receptor responses. We also show that, in platelets, internalization of both these receptors is dependent upon dynamin, and that this process is required for resensitization of responses. Conclusions: This study is therefore the first to show that both P2Y1 and P2Y12 receptor activities are rapidly and reversibly modulated in human platelets, and it reveals that the underlying mechanism requires receptor trafficking as an essential part of this process.
Adenosine diphosphate (ADP) plays a central role in platelet activation by acting as a cofactor in the platelet responses to physiological agonists, including thromboxane A2, collagen, and thrombin. ADP activates two surface expressed G protein-coupled receptors (GPCRs): P2Y1 and P2Y12 [1,2]. The combined stimulation of P2Y1 (coupled to Gq and PLCβ) and P2Y12 (negatively coupled to adenylyl cyclase through Gi) is necessary for the full platelet aggregation response to ADP with platelet activation initiated by the P2Y1 receptor and amplified by P2Y12.
The attenuation of receptor-stimulated signal output upon sustained or recurrent agonist stimulation, a process known as desensitization, is a crucial physiological mechanism of adaptation observed for many GPCRs . As with other GPCRs, the responsiveness of P2Y1 and P2Y12 receptors is tightly regulated. Recently, we showed that both P2Y1 and P2Y12 receptor responses desensitize in human platelets , in agreement with the observed desensitization of platelet responses following prolonged exposure to ADP [5–7]. Mechanisms underlying desensitization are complex and can involve phosphorylation of the receptor, uncoupling from G proteins, internalization, and ultimately intracellular downregulation [3,8–12]. We have recently discovered that ADP pretreatment promotes P2Y1 and P2Y12 receptor phosphorylation, desensitization, and internalization by different kinase-dependent mechanisms [4,13]. P2Y1, but not P2Y12, desensitization is mediated by protein kinase C (PKC). In contrast, agonist-induced desensitization of P2Y12, but not P2Y1, is largely dependent on G protein-coupled receptor kinases (GRKs) activity. Further, we have discovered that both receptors undergo agonist-induced internalization, interestingly through distinct populations of clathrin-coated pits .
Upon removal of agonist, the attenuated responsiveness of many GPCRs is reversed in a process known as resensitization [3,15]. There is only one report of resensitization of receptor responses taking place in platelets, in response to 5-hydroxytryptamine . Balanced homeostatic mechanisms to regulate platelet reactivity, however, would suggest that, if responses to agonists are able to desensitize, then resensitization mechanisms must also operate and be widespread. As prolonged or irreversible receptor desensitization would leave the platelet unable to respond appropriately to extracellular stimuli, and may potentially lead to impaired hemostatic function, we sought to determine if responses to ADP could be resensitized following receptor desensitization.
In order for platelet responses to agonists to be dynamically controlled, they must be able not only to lose responsiveness through the processes of desensitization, but also to regain that responsiveness through a process of resensitization. In this study, we examine if purinergic responses can resensitize following receptor desensitization. The mechanism by which the resensitization of many GPCRs is achieved is thought to be the agonist-stimulated internalization of receptors to an intracellular membrane compartment (endosomes) enriched in a GPCR-specific phosphatase activity [8,10] followed by recycling to the cell membrane. Therefore, we also examined if receptor internalization, dephosphorylation, and recycling are required for resensitization of purinergic responsiveness. This study has major implications for our understanding of how platelet responsiveness is homeostatically regulated in a balanced manner to achieve the correct responsiveness under a variety of different vascular conditions.
Materials and methods
Dulbecco’s modified Eagle’s medium (DMEM), Lipofectamine 2000, and fetal bovine serum were obtained from Life Technologies Inc. (Renfrewshire, UK). Radiochemicals were from Perkin Elmer Life Sciences (Beaconsfield, UK). Complete protease inhibitor tablets were from Roche (Welwyn Garden City, UK). Anti-HA-monoclonal antibody (HA-11), goat antimouse fluorescein-conjugated secondary antibody was purchased from Molecular Probes (Invitrogen, Paisley, UK). Dynasore was synthesized as previously described . All other reagents were from Sigma (Sigma-Aldrich, Poole, UK).
Preparation of human platelets
Human blood was drawn from healthy, drug-free volunteers on the day of the experiment under ethical approval from the Local Research Ethics Committee, United Bristol Healthcare Trust (Project E5736). Acid citrate dextrose (ACD: 120 mm sodium citrate, 110 mm glucose, 80 mm citric acid, used at 1:7 vol/vol) was used as anticoagulant. Platelet rich plasma (PRP) was prepared by centrifugation at 200 × g, for 17 min and platelets were then isolated by centrifugation for 10 min at 1000 ×g, in the presence of 0.02 U mL−1 apyrase and prostaglandin E1 (PGE1; 140 nm) for all assays other than measurement of intracellular cyclic AMP (cAMP) where PGE1 was omitted to ensure reproducible and low basal cAMP levels. PGE1 was included in all other studies to reduce platelet aggregation. The pellet was resuspended to a density of 4 × 108 platelets mL−1 in a modified Tyrodes-HEPES buffer (145 mm NaCl, 2.9 mm KCl, 10 mm HEPES, 1 mm MgCl2, 5 mm glucose, pH 7.3). To this platelet suspension, 10 μm indomethacin and 0.02 U mL−1 apyrase were added, and a 30 min resting period was allowed before stimulation.
Measurement of cytosolic free calcium ([Ca2+]i) in platelets
Measurement of cytosolic calcium was performed as previously described . Briefly, 3 μm Fura-2-AM was added to PRP, and incubated at 37 °C for 45 min in the presence of 10 μm indomethacin. Platelets were centrifuged and resuspended in modified Tyrodes. ADP (10 μm)-induced calcium responses were subsequently measured at 37 °C using a Hitachi F-4500 (Hitachi High-Technologies, Maidenhead, UK) spectrofluorimeter with fluorescence excitation made at 340 and 380 nm, and emission at 510 nm. Raw data are expressed as the ratio of emissions at excitation wavelengths 340:380 nm (fluorecence ratio F1/F2) whilst collated data are expressed as percent peak rise in cytosolic calcium concentration (as a percent of the response to 10 μm ADP alone). These calcium concentrations were determined from calibrations run with each experiment, and calculated using standard ratiometric equations . To induce receptor desensitization, a desensitizing concentration of ADP (10 μm) was added to platelets for 300 s. Subsequently a stimulating concentration of ADP (10 μm) was added, and the response monitored. Because the ADP response in platelets is often attenuated following multiple spin wash steps, desensitizing ADP was then removed by the addition of 0.2 U mL−1 apyrase (10 min), rather than a wash and spin-step, to promote receptor resensitization. Following apyrase treatment, a stimulating concentration of ADP (10 μm) was again added, and the response measured. In all experiments, non-desensitized controls were performed where no desensitizing ADP was added. Similarly, the responses of (non-desensitized) platelets to ADP were determined in the presence and absence of apyrase in order to verify that apyrase did not affect stimulations.
Measurement of cAMP levels in platelets
Platelets were stimulated in the presence of the phosphodiesterase inhibitor IBMX (100 μm) ± forskolin (1 μm) in the absence or presence of ADP (10 μm) for 5 min at 37 °C. Cyclic AMP accumulation was terminated by addition of ice cold 100% trichloroacetic acid and samples were left to lyse on ice for 1–2 h. The resulting samples were spun at 4000 ×g for 5 min and the cAMP-containing supernatant neutralized with 1m NaOH and TE buffer (50 mm Tris-HCl, 4 mm EDTA, pH 7.4). Cyclic AMP levels were subsequently determined in each sample using a binding assay as previously described . Receptor desensitization was performed as previously described . Briefly, in order to induce receptor desensitization, platelets were stimulated with ADP (10 μm; 300 s). In order to induce receptor resensitization, 0.2 U mL−1 apyrase was subsequently added to remove desensitizing ADP from platelets. As previously described, 1 mm EGTA was added 1 min prior to cAMP accumulation experiments to negate calcium dependent-apyrase activity . Cyclic AMP accumulation assays were performed on non-desensitized control, desensitized or resensitized platelets. Data are presented as percent inhibition of forskolin-stimulated adenylyl cyclase.
Radioligand binding in human platelets
Platelets were pretreated with monensin (50 μm; 15 min) or vehicle alone. To induce receptor internalization, platelets were subsequently stimulated with ADP (10 μm 15 min) or vehicle alone. ADP was then removed by the addition of 0.2 U mL−1 apyrase (15 min) to induce receptor recycling. P2Y1 and P2Y12 surface receptor expressions were subsequently determined by ligand binding in fixed platelets, as previously described .
1321N1 human astrocytoma cells stably expressing hemagglutinin (HA)-tagged human P2Y1 or P2Y12 receptors were generated as previously described . Cells were maintained in DMEM supplemented with 10% fetal bovine serum, 100 units ml−1 penicillin G, 100 μg ml−1 streptomycin sulphate, and 400 μg mL−1 geneticin at 37 °C in a humidified atmosphere of 95% air, 5% CO2.
Measurement of cytosolic free calcium ([Ca2+]i) in 1321N1 astrocytoma cells
The cytosolic free Ca2+ concentration was determined using the fluorescent Ca2+indicator fura-2-acetoxymethyl ester (fura-2/AM) as previously reported . Briefly, transfected cells were grown on poly-L-lysine coated glass coverslips and used at ∼60% confluence. Cells were washed twice with Locke’s solution (154 mm NaCl, 5.6 mm KCL, 1.2 mm MgCl2, 2.2 mm CaCl2, 5 mm HEPES, 10 mm glucose, pH 7.4) and incubated with fura-2/AM (3 μm) at 37 °C for 60 min. Glass coverslips were mounted into a quartz cuvette and placed into a thermostatically controlled cell holder at 37 °C. Using a continuous perfusion system, we were able to control the addition and subsequent washout of ADP without the need for addition of apyrase. Briefly, cells were continuously perfused with Locke’s solution. Fluorescence was measured at 340 and 380 nm excitation and 510 nm emission. ADP (1 μm) was perfused onto cell monolayers as required. As for measurement of platelet [Ca2+]i described above, raw data are expressed as the ratio of emissions at excitation wavelengths 340:380 nm (fluorescence ratio F1/F2) whilst collated data are expressed as percent peak rise in cytosolic calcium concentration, determined from ratiometric data as previously described .
Measurement of cAMP accumulation in 1321N1 astrocytoma cells
Desensitization and signaling of P2Y12 receptor responses in 1321N1 cells were measured as previously described . Cells were exposed to a desensitizing dose of ADP (10 nm; 5 or 15 min) in the presence of the phosphodiesterase inhibitor Ro201724 (250 μm). Apyrase (0.2 unit mL−1) was then added directly to each well to promote receptor resensitization and incubated at 37 °C (5, 10, and 30 min) to remove the desensitizing ADP. Cells were then washed and forskolin (1 μm) added in the absence or presence of ADP (10 nm) and plates incubated at 37 °C for 10 min. Cyclic AMP accumulation was terminated by addition of ice cold 100% trichloroacetic acid and supernatant neutralized with 1 m NaOH and TE buffer. Cyclic AMP levels were determined as previously described . Data are expressed as percent inhibition of forskolin-stimulated adenylyl cyclase.
Internalization and immunofluorescence microscopy of HA-P2Y1 and HA-P2Y12 in 1321N1cells
HA-tagged surface receptor loss was assessed by ELISA as described previously [21,22]. Briefly, cells were transiently transfected with pcDNA3 containing dominant-negative mutant dynamin (dynamin-DNM; dynaminK44A). Twenty-four hours post-transfection, cells were split into 24-well tissue culture dishes coated with 0.1 mg mL−1 poly-L-lysine. Twenty-four hours later, cells were incubated with DMEM containing apyrase (0.1 unit mL−1; 1 h) and either dynasore (80 μm; 15 min), monensin (50 μm; 15 min), okadaic acid (10 nm; 15 min) or vehicle alone. Cells were then washed and challenged with DMEM containing ADP (10 μm) for 0–15 min at 37 °C. In order to induce receptor recycling, apyrase was added (0.2 unit mL−1) to remove ADP. Changes in surface receptor expression were subsequently determined by an immunosorbent assay (ELISA) taking advantage of the HA-epitope tag [21,22], and expressed as either percent surface receptor or percent loss of surface receptor with the background signal from pcDNA3-transfected controls subtracted from all receptor-transfected values.
Cellular distribution of HA-tagged receptor in 1321N1 cells was assessed by immunofluorescence microscopy . Briefly, cells were grown on poly-L-lysine coated coverslips in six well plates. Twenty-four hours later, receptor distribution was assessed using a primary anti-HA-monoclonal antibody (HA-11;1:200) and goat antimouse fluorescein-conjugated secondary antibody (1:200). Coverslips were mounted using Slow-Fade mounting medium and examined by microscopy on an upright Leica TCS-NT confocal laser scanning microscope attached to a Leica DM IRBE epifluorescence microscope with phase-contrast and a Plan-Apo 40 × 1.40 NA oil immersion objective. All images were collected on Leica TCS-NT software for two-dimensional and three-dimensional image analysis, and processed using Adobe Photoshop 6.0.
Experimental design and statistics
Data were analyzed by the iterative fitting program GraphPAD Prism (GraphPAD Software, Inc., La Jolla, CA, USA). Log concentration–effect curves were fitted to logistic expressions for single-site analysis. Where appropriate, statistical significance was assessed by Mann–Whitney U-test or by two-way anova.
Resensitization of platelet purinergic receptor responses
Initially, we sought to determine if either P2Y1 or P2Y12 purinergic receptor responses could recover following agonist-induced desensitization. Platelets were pretreated with ADP to desensitize receptor responses, and then allowed to recover in the presence of apyrase added to remove desensitizing ADP. As expected and previously reported, pretreatment  with agonist (ADP; 10 μm; 300 s) decreased subsequent ADP-stimulated P2Y1 (Fig. 1A; showing a ratio trace and B pooled data expressed as percent of control response) and P2Y12 (Fig. 1C) receptor responsiveness. In order to confirm that, as in our previous studies , the reduction in calcium responsiveness was not a result of a depletion of intracellular stores, we determined that the calcium response to another agonist, collagen, was unaffected by pretreatment of platelets with ADP [peak calcium responses (F1/F2) to collagen (30 μg mL−1) were 0.93 ± 0.32 and 0.96 ± 0.32 in platelets pretreated with ADP (10 μm; 300 s) or non-pretreated control platelets, respectively; n = 3]. Interestingly, both P2Y1 (Figs 1A and 1B) and P2Y12 (Fig. 1C) receptor responses rapidly resensitized following ADP removal by addition of apyrase (0.2 units mL−1). Of the two receptors, responses to P2Y12 were more robustly resensitized, recovering almost completely after a period of 30 min (Fig. 1C). The P2Y1 receptor response also recovered significantly at all time points tested, although it had not fully returned 30 min after removal of ADP (Fig. 1B). In a previous study, we demonstrated that the addition of apyrase did not significantly affect subsequent ADP-stimulated P2Y1 receptor responsiveness . In these studies, we again confirmed that pretreatment with apyrase (0.2 units mL−1) for the length of the resensitization period (30 min) did not decrease subsequent P2Y1 receptor responsiveness [peak calcium response to ADP (10 μm) 107 ± 4.5% of that in control non-apyrase treated platelets], indicating that this receptor did not fully resensitize in this time frame.
Resensitization and recycling of purinergic receptors in 1321N1 cells
In order to study the molecular mechanisms underlying GPCR resensitization and to circumvent the methodological problems that this presents in platelets, we examined P2Y1 and P2Y12 receptor resensitization in 1321N1 human astrocytoma cells stably expressing either receptor. As in platelets, we initially confirmed that P2Y1 and P2Y12 purinergic receptor responses can resensitize in this cell system. As previously reported, pretreatment with ADP promoted desensitization of both P2Y1 (Figs 2A and 2B) and P2Y12 (Fig. 2C) receptor responses . As in platelets, the removal of ADP with apyrase promoted a rapid resensitization of both P2Y1 (Fig. 2A, showing a ratio trace and B pooled data expressed as peak rise in intracellular calcium) and P2Y12 (Fig. 2C) receptor responses. As in platelets, the P2Y1 receptor response had not fully returned 30 min after removal of ADP [peak calcium response to ADP (10 μm) 75.5 ± 6.4% of that in control non-pretreated cells] with control experiments confirming that pretreatment with apyrase did not significantly attenuate subsequent purinergic receptor responses (data not shown). For many GPCRs, internalization followed by subsequent recycling to the cell membrane mediates their resensitization [3,10,15]. We therefore sought to determine if either of these receptors also recycles back to the cell surface following agonist-induced internalization, and whether this may represent the mechanism underlying resensitization. As the 1321N1 cells stably express N-terminal HA-epitope tagged versions of either receptor, we were able to quantify the agonist-induced surface receptor loss and subsequent recycling back to the cell surface by ELISA [21,22] and immunofluorescence [13,14], as previously described. Interestingly, following short periods of agonist exposure, both P2Y1 and P2Y12 rapidly recycle back to the cell surface following ADP removal with apyrase (Fig. 3A). Ligand binding studies (Fig. 3B) were also performed using the non-selective P2Y receptor ligand [3H]-2MeSADP (100 nm), in the presence of A3P5P (1 mm) or AR-C69931MX (1 μm) to give an estimate of either the P2Y1 or P2Y12 surface binding sites. As with our ELISA experiments, following ADP pretreatment (15 min) there was a reduction in both P2Y1 or P2Y12 binding sites. Following apyrase exposure for 15 min (ADP/apyrase), we found that both the P2Y1 and P2Y12 surface receptor levels returned to control levels. Immunofluorescence microscopy also confirmed that both receptors rapidly internalize into endocytic vesicles following the addition of ADP, and then recycle back to the cell surface following removal of the agonist with apyrase (Fig. 4). We have previously examined the endocytic location of these internalized receptors and found that they enter a transferrin-positive compartment .
Disruption of receptor internalization, dephosphorylation and recycling attenuates receptor resensitization in 1321N1 cells. It was now critical to investigate whether disruption of receptor recycling could also disrupt resensitization, so as to provide a definitive causal link between the two events. In order to block receptor endocytosis, we expressed a dynamin-DNM (K44A), which, as we had previously shown , attenuated both P2Y1 (Fig. 3C) and P2Y12 (Fig. 3D) internalizations. In addition, we used a novel inhibitor of dynamin, dynasore, which is a cell permeant reagent that rapidly inhibits the GTPase activity of dynamin with high specificity  and has been shown to block endocytosis in cultured hippocampal neurons . Pretreatment with this compound effectively attenuated the internalization of both the P2Y1 and P2Y12 purinergic receptor vs. vehicle (DMSO) controls. To further ensure the selectivity of dynasore, we examined whether ligand binding, [3H]-2MeSADP (100 nm), was altered in the presence of this compound. Pretreatment with dynasore did not significantly alter the ability of the ligand to bind to either the P2Y1 [720 ± 34 and 741 ± 42 dpm mg−1 protein in control (DMSO) and dynasore treated cells] or P2Y12 [692 ± 29 and 681 ± 37 dpm mg−1 protein in control (DMSO) and dynasore treated cells] receptors. In order to block receptor recycling, we used a chemical inhibitor monensin, which raises the pH within endosomes and blocks recycling of receptors back to the cell surface, without first disrupting receptor internalization. A number of different groups have shown that acidification of early endosomes is important for agonist dissociation, dephosphorylation, and resensitization of receptors [24,25] and have used monensin to block receptor recycling [26–28]. In agreement with these authors, we found that pretreatment with monensin effectively blocked receptor recycling of both P2Y1 (Fig. 3C) and P2Y12 (Fig. 3D), without a significant effect upon receptor internalization, leading to accumulation of internalized receptor in recycling endosomes (Fig. 4). In order to demonstrate that internalization of either receptor was mediated by a dynamin-dependent mechanism, Figs 3C and 3D show that overexpression of dominant negative dynamin, or inhibition of dynamin by the novel inhibitor dynasore (80 μm), was able to block internalization. In addition, as receptor dephosphorylation is often required for receptor recycling, and previous data from our laboratory had shown that both of these GPCRs are phosphorylated in an agonist-dependent manner , we also examined receptor recycling in the presence of the protein phosphatase (PP) inhibitor okadaic acid. Okadaic acid selectively inhibits PP2A at the concentration (10 nm) used in our study, with the PP2B isoform inhibited at concentrations greater than 5000 nm and PP2C not inhibited by it . Okadaic acid has been shown to decrease the dephosphorylation, recycling, and resensitization of a number of GPCRs, including the β2 adrenoceptor  and P2Y2 purinergic receptor . Figures 3C and 3D show that okadaic acid does not significantly affect the internalization of the P2Y1 receptor, whilst that of the P2Y12 receptor is attenuated. Interestingly, okadaic acid did attenuate the recycling of the P2Y12 receptor (Fig. 3D) whilst that of the P2Y1 receptor (Fig. 3C) was unaffected.
It was then important to correlate the effects of these inhibitors of internalization, dephoshorylation or recycling with their effects upon desensitization and resensitization of P2Y1 and P2Y12. Importantly, the expression of dynamin-DNM, or treatment with dynasore, okadaic acid or monensin all significantly blocked P2Y1 (Fig. 5A) and P2Y12 (Fig. 5B) receptor resensitization in 1321N1 cells, but had no significant effect upon desensitization of these receptors. Expression of wild-type dynamin did not significantly affect either the desensitization or resensitization of either receptor, indicating that there is sufficient endogenous dynamin in these cells to support receptor traffic. Together, these data indicated that dynamin-dependent internalization, recycling, and dephosphorylation are all required for efficient resensitization of receptors after desensitization.
Disruption of receptor dephosphorylation and recycling attenuates receptor resensitization in human platelets
We next sought to determine if our findings in 1321N1 cells could be repeated in human platelets. We initially examined if either P2Y1 or P2Y12 could recycle back to the cell surface following agonist-induced internalization. In order to study the internalization of purinergic receptors in human platelets, as in a recent study , we used [3H]-2MeSADP (100 nm) in the presence of A3P5P (1 mm) or AR-C69931MX (1 μm) to give an estimate of either the P2Y1 or P2Y12 surface binding sites. In this previous study, ligand-binding saturation experiments revealed that [3H]-2MeSADP effectively bound to surface P2Y receptors in fixed platelets with approximately 70% displacement in the presence of unlabeled radioligand (10 μm) corresponding to 901 ± 41 [3H]-2MeSADP binding sites per platelet. Further saturation experiments using the P2Y1 receptor antagonist A3P5P (1 mm) or the P2Y12 receptor antagonist AR-C69931MX (1 μm) revealed two distinct binding populations of 184 ± 27 and 644 ± 11 [3H]-2MeSADP binding sites per platelet, which represent the P2Y1 and P2Y12 receptors respectively. Interestingly, experiments using combined P2Y1 and P2Y12 receptor antagonists estimated the number of binding sites to be 844 ± 57, a number not significantly different to that obtained with unlabeled 2MeSADP. The levels of P2Y1 and P2Y12 receptor expression were therefore significantly lower in platelets (Bmax 117 ± 41 and 470 ± 83 dpm mg−1 protein for P2Y1 and P2Y12 receptors, respectively) than in 1321N1 cells (Bmax 665 ± 45 and 685 ± 70 dpm mg−1 protein for P2Y1 and P2Y12 receptors, respectively). As in our previous studies, following pretreatment with ADP (10 μm; 10 min), its subsequent removal with apyrase (0.2 unit mL−1; 3 min) and platelet fixation, there was a clear reduction in [3H]-2MeSADP binding to both P2Y1 (Fig. 6A) and P2Y12 (Fig. 6B) compared with non-pretreated or apyrase-alone treated controls. Data shown are summarized in Fig. 6C. Interestingly, if we lengthened the period of apyrase exposure to 15 min (ADP/apyrase), we found that both the P2Y1 and P2Y12 surface receptor levels returned to control levels. We next determined if pretreatment with either dynasore, to block receptor internalization, or monensin to block receptor recycling could also disrupt the traffic of these receptors in human platelets. Dynasore (80 μm; 15 min) pretreatment effectively blocked the internalization of both the P2Y1 and P2Y12 receptors (Figs 6A and 6B). Monensin (10 nm; 15 min) pretreatment had no effect on the degree of either P2Y1 or P2Y12 receptor internalization but clearly attenuated recycling of either receptor (Figs 6A and 6B). This is even more apparent when the data are expressed as percent of the control surface receptor before the internalization and recycling experiments (Fig. 6C). Indeed, monensin pretreatment attenuates the recycling of both the P2Y1 or P2Y12 receptors by approximately 80% [level of recycling, 99% ± 5.3% (control) and 20% ± 2.8% (monensin) for P2Y1receptor and 93% ± 4.7%(control) and 10% ± 5.8% (monensin) for P2Y12 receptor]. Dynasore or monensin pretreatment did not alter the levels of [3H]-2MeSADP binding to either the P2Y1 or P2Y12 receptors (Figs 6A and 6B). As we were now able to block receptor internalization with dynasore and receptor recycling with monensin, we subsequently investigated the desensitization and resensitization of P2Y1 and P2Y12 receptor responses in the presence of these compounds. We also examined changes in receptor responsiveness in the presence of okadaic acid to determine if receptor dephosphorylation also played a role in resensitization of receptor responsiveness in human platelets. Although pretreatment with dynasore, monensin or okadaic acid did not significantly change the degree of either P2Y1 or P2Y12 receptor desensitization, each significantly attenuated resensitization of each receptor (Figs 7A and 7B). Of the three compounds, dynasore and monensin were much more effective at blocking either P2Y1 or P2Y12 receptor resensitization than okadaic acid, with the latter compound causing a small but significant attenuation of P2Y1 and P2Y12 receptor resensitization.
The activation of P2Y1 and P2Y12 purinergic receptors by ADP is critical for normal platelet function. In order to avoid inappropriate thrombosis, the sensitivity of these receptors to agonist needs to be continuously regulated. Just as the desensitization of GPCRs provides a mechanism for protecting cells against receptor overstimulation, GPCR resensitization protects cells against prolonged receptor desensitization, essential in the maintenance of cell function. In this study, we show for the first time that purinergic receptor responses in human platelets can rapidly resensitize. In addition, we determine details of the mechanism responsible for resensitization, showing that receptor internalization, dephosphorylation, and subsequent recycling to the cell surface are required to allow resensitization of both P2Y1 and P2Y12 receptor responses.
A number of groups have now shown that purinergic receptor function in human platelets undergoes rapid desensitization [5–7]. Studies in ecto-nucleoside triphosphate diphosphohydrolase1 (CD39) knockout mice, deficient in ATP/ADP hydrolysis, have also shown that purinergic receptor responses desensitize as a result of increased levels of circulating ADP . CD39 knockout mice have prolonged bleeding times whilst platelet interactions with injured mesenteric vasculature are considerably reduced in vivo. Therefore, thrombosis formation is severely blocked in CD39 knockout mice, indicating that if P2Y receptors are desensitized, then platelet activation is severely hampered. In a recent study , we demonstrated that both P2Y1 and P2Y12 desensitize in platelets by different kinase-dependent mechanisms, with GRKs regulating P2Y12 and PKC regulating P2Y1 agonist-induced desensitization in human platelets. Further detailed studies have revealed that both receptors undergo agonist-induced phosphorylation and internalization [13,14]. Upon removal of agonist, the attenuated responsiveness of many GPCRs is reversed in a process known as resensitization [8,10]. Interestingly, it is possible to resensitize purinergic receptor responses in platelets taken from CD39 knockout mice if ADP is removed by incubation with apyrase . To date, there have been minimal studies examining the resensitization of GPCR responsiveness in human platelets. Indeed, the only responses shown to resensitize in human platelets are those to the 5HT2 receptor . Clearly, in our hands both the P2Y1 and P2Y12 receptor responses can rapidly resensitize in human platelets. A recent study also showed that P2Y1 receptor responses resensitize in HEK293 cells .
The mechanism by which the resensitization of many GPCRs is achieved is thought to be the agonist-stimulated internalization of receptors to an intracellular membrane compartment (endosomes) enriched in a GPCR-specific phosphatase activity [8,10]. In order to test if this was the case for P2Y1 and P2Y12 receptors, we initially examined their signaling and trafficking in 1321N1 human astrocytoma cells. We have previously used these P2Y ‘null’ cells stably overexpressing each receptor to examine aspects of purinergic receptor regulation and trafficking [4,13]. With this system, unlike platelets we can easily alter protein expression. In addition, by making use of the epitope-tag engineered into the N-terminus of these receptors changes in surface receptor expression can be examine by either ELISA or immunofluorescence microscopy [13,14].
As with platelets, purinergic receptor responses resensitized rapidly in 1321N1 cells. P2Y12 receptor resensitization was slightly slower in platelets than in cell lines. Ligand binding studies in platelets and cell lines reveal that P2Y12 is overexpressed in cell lines vs. platelets, which may provide a greater receptor reserve speeding up receptor resensitization. Equally, differences in the endocytic machinery regulating these processes may exist between platelets and 1321N1 cells. Following agonist-induced internalization, both the P2Y1 and P2Y12 receptors did indeed rapidly recycle back to the cell surface. Interestingly, a blockade of receptor internalization by overexpression of dynamin-DNM (K44A)  receptor dephosphorylation by treatment with okadaic acid or receptor recycling with monensin all effectively blocked receptor resensitization. Our findings agree with those from a recent study in HEK293 cells, which showed that pretreatment with okadaic acid inhibited P2Y1 receptor resensitization . Pretreatment with okadaic acid altered the traffic of the purinergic receptors, specifically the P2Y12 receptor. P2Y1 receptors appear to traffic normally with both internalization and recycling unaltered by okadaic acid addition. As P2Y1 receptors did recycle back to the cell surface whilst responses did not resensitize, our results would suggest that the receptor is recycled in a non-agonist responsive state. In contrast, P2Y12 receptor internalization and recycling are attenuated by okadaic acid pretreatment. At present, we are unable to provide a molecular mechanism that may explain this phenomenon. In a recent paper, we did discover that the P2Y1 and P2Y12 receptors traffic through different clathrin-coated pits . Therefore, we hypothesize that these different populations of clathrin-coated pit may have differential sensitivity to changes in phosphatase activity, which in turn could selectively alter P2Y12 vs. P2Y1 receptor traffic.
As these studies confirmed the importance of receptor trafficking in receptor resensitization, we examined if a blockade of these processes in platelets also blocked receptor resensitization. Interestingly, ligand-binding studies revealed that both the P2Y1 and P2Y12 receptors rapidly recycled back to the cell surface following agonist-induced internalization in human platelets. Using dynasore, a cell permeant inhibitor of the GTPase dynamin , we were able to demonstrate for the first time that the internalization of both the P2Y1 and P2Y12 receptors is dynamin-dependent in human platelets. To our knowledge, this is the first time that dynamin-dependent internalization of an endogenous GPCR in human tissue has been definitively demonstrated. Interestingly, receptor desensitization was not altered if dynamin-dependent receptor internalization was blocked in either 1321N1 cells or human platelets. Therefore, at these acute time points of agonist exposure, receptor desensitization is independent of receptor internalization, with both the P2Y1 and P2Y12 receptors uncoupling from G protein following kinase specific phosphorylation [4,13,14].
Further studies examining the intracellular trafficking and subcellular location of the purinergic receptors in platelets are planned, although the absence of a commercially available antibody to either receptor of sufficient quality has hampered ongoing immunofluorescent and electron microscopy experiments. Ligand-binding studies did show, however, that pretreatment with monensin blocked receptor recycling, as was the case in 1321N1 cells. Treatment with dynasore, monensin or with the phosphatase inhibitor okadaic acid attenuated receptor resensitization. Therefore, as in 1321N1 cells, following agonist-induced desensitization and internalization, both P2Y1 and P2Y12 receptors traffic to an endosomal compartment where they are dephosphorylated by a specific phosphatase, most likely PP2A. Following this dephosphorylation, the resensitized receptor is recycled back to the cell membrane.
Clearly, an important question that now needs to be addressed is the identification of endocytic proteins that may regulate the traffic of these receptors. For example, studies in cell models have identified that NHERF proteins, which can regulate GPCR sorting , can specifically bind to P2Y1 [35,36]. Studies in our laboratory are now underway to identify if this or other putative endocytic proteins [10,11] may regulate purinergic receptor traffic in our cell line model. It should be noted, however, that it is currently technically too challenging to address these questions in human platelets.
In summary, this and previous studies from our laboratory now show that platelet P2Y1 and P2Y12 receptor responses are able to rapidly respond to changes in circulating ADP levels. An interesting question to address is the concentration of ADP at the site of a thrombus and to determine when receptor desensitization maybe initiated. Platelet P2Y responses in vitro desensitize to ADP concentrations in the high nanomolar range and above . Most work so far has concentrated on mechanisms of desensitization of responses, and we have recently shown differential kinase-dependent processes operating for P2Y1 and P2Y12 in platelets to regulate their sensitivity status and internalization . If these processes operated in isolation, however, it is clear that platelets would rapidly become permanently unresponsive to ADP, and other agonists, leading to a state of platelet hypo-responsiveness and associated bleeding disorders such as that seen in CD39 knockout mice where ADP levels are continuously raised . In a homeostatic system, counterbalancing processes are likely to operate to achieve equilibrium levels of sensitivity to stimuli, and this is clearly very important in a system such as the platelet where rapid responsiveness is essential physiologically. Following decreases in circulating ADP levels therefore, platelet purinergic receptor responses subsequently resensitize. As a consequence of this rapid resensitization, platelets are able to retain their reactivity at sites of vessel injury even if they have already been exposed previously to ADP, thereby sustaining hemostatic function. Recent studies have suggested that continuous ADP signaling is required for thrombus stability [37,38]. It is unclear at present how the processes and timing of P2Y desensitization and subsequent resensitization link with those of thrombus formation and subsequent stability. Further study is required to determine how blockade of either of these processes will alter thrombus stability.
As ADP performs such a pivotal role in the formation of stable platelet aggregates, receptor desensitization and then subsequent resensitization represent key mechanisms by which the delicate balance between rest and activation of platelets in the circulation is maintained.
The authors would like to thank AstraZeneca for the kind gift of AR-C69931MX. We thank D. Jane, Department of Physiology and Pharmacology, University of Bristol for assistance with the synthesis of dynasore. We thank M. Jepson and A. Leard for their assistance within the School of Medical Sciences Cell Imaging Facility. The work was supported by the British Heart Foundation (grant nos. RG/05/015, FS/04/023 & FS/05/017). A. W. Poole is a BBSRC Research Development Fellow and S. J. Mundell is a British Heart Foundation Basic Science Lecturer.
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.