Satya P. Kunapuli, Department of Physiology, Temple University, Department of Physiology, Rm. 224, OMS, 3420 N. Broad Street, Philadelphia, PA 19140, USA. Tel.: +1 215 707 4615; fax: +1 215 707 4003; e-mail: email@example.com
Summary. Alpha granule release plays an important role in propagating a hemostatic response upon platelet activation. We evaluated the ability of various agonists to cause α granule release in platelets. Alpha granule release was measured by determining P-selectin surface expression in aspirin-treated washed platelets. ADP-induced P-selectin expression was inhibited both by MRS 2179 (a P2Y1 selective antagonist) and AR-C69931MX (a P2Y12 selective antagonist), suggesting a role for both Gαq and Gαi pathways in ADP-mediated α granule release. Consistent with these observations, the combination of serotonin (a Gαq pathway stimulator) and epinephrine (a Gαz pathway stimulator) also caused α granule release. Furthermore, U46619-induced P-selectin expression was unaffected by MRS 2179 but was dramatically inhibited by AR-C69931, indicating a dominant role for P2Y12 in U46619-mediated α granule release. Additionally, the Gα12/13-stimulating peptide YFLLRNP potentiated α granule secretion in combination with either ADP or serotonin/epinephrine costimulation but was unable to induce secretion by itself. Finally, costimulation of the Gαi and Gα12/13 pathways resulted in a significant dose-dependent increase in α granule release. We conclude that ADP-induced α granule release in aspirin-treated platelets occurs through costimulation of Gαq and Gαi signaling pathways. The P2Y12 receptor plays an important role in thromboxane A2-mediated α granule release, and furthermore activation of Gα12/13 and Gαq signaling pathway can cause α granule release.
Platelet activation plays an important role in thrombosis and hemostasis . When platelets are stimulated with agonists such as ADP, thromboxane A2 or thrombin, platelets aggregate, release their granule contents, and generate thromboxane A2. In addition to dense granules, platelets also contain α granules that contain homologs of plasma proteins and platelet-specific proteins . In response to platelet activation, these proteins are secreted and/or shuttled to the plasma membrane where they aid in the formation of platelet aggregates (αIIbβ3 and fibrinogen), wound healing (platelet-derived growth factor), acceleration of the clotting cascade (prothrombin) and mediation of platelet–leukocyte interactions (P-selectin). As physiologically relevant agonists, ADP and thromboxane A2 mediate their effects on platelets by utilizing G protein-coupled receptors found on plasma membranes. Costimulation of the Gαq-coupled P2Y1 receptor and the Gαi-coupled P2Y12 receptor is needed for ADP-induced platelet activation [4–7]. The thromboxane A2 (TPα) receptor couples to both Gαq and Gα12/13, but depends on ADP from dense granule contents for Gαi signaling (through the P2Y12 receptor) and subsequent platelet activation . We have previously shown that ADP- and thromboxane A2-induced platelet aggregation is accomplished by concomitant signaling from both Gαq and Gαi pathways [9,10].
Although it has been previously shown that ADP is capable of inducing α granule release [11–13], the contribution of Gαq-coupled P2Y1 and Gαi-coupled P2Y12 receptors in ADP-mediated α granule release has not been studied yet. It is well established that ADP does not cause dense granule secretion in aspirin-treated platelets, even though signaling through the P2Y12 receptor can potentiate dense granule secretion mediated by other platelet agonists such as thromboxane A2, thrombin and collagen . As far as ADP-mediated α granule release is concerned, the main questions that have to be addressed include: can the Gαq-coupled P2Y1 receptor by itself cause α granule release independently of P2Y12 signaling with the P2Y12 receptor playing just a potentiating role? Or is signaling by both P2Y1 and P2Y12 receptors essential for ADP-induced α granule release? In this study, one of our primary aims was to investigate the role of each ADP receptor on α granule release using specific receptor antagonists in human platelets. While thromboxane A2-mediated dense granule secretion occurs primarily downstream of the Gαq-coupled TPα receptor, signaling downstream of ADP-induced P2Y12 receptor signaling is essential for thromboxane A2-mediated platelet aggregation. The receptors that play an important role in thromboxane A2-mediated α granule release have not yet been studied. In particular, the roles of TPα, P2Y12 and P2Y1 receptors in this process are addressed in this study.
In addition, it has been shown that costimulation of the Gαq-coupled serotonin receptor and Gαz-coupled α2A adrenergic receptor can result in platelet aggregation . Recently, studies performed in our laboratory and others have shown that stimulation of Gα12/13 and Gαi can cause fibrinogen receptor activation and subsequent platelet aggregation [15,16]. Given the interplay between the different G proteins in causing platelet aggregation, it would be interesting to determine if these redundant G protein pathways could also result in α granule secretion in human platelets and hence substitute for Gαq and Gαi signaling pathways.
In this study, we show that signaling through both P2Y12 and P2Y1 receptors are important for ADP-mediated α granule release in human platelets. Also, thromboxane A2-mediated α granule release requires signaling through Gαi pathways through either P2Y12 or α2A-adrenergic receptors. Finally, Gα12/13 signaling, in addition to potentiating ADP-induced granule release, can independently synergize with Gαi signaling and cause α granule release in human platelets.
Materials and methods
Human blood was collected from informed healthy volunteers, all of whom are students or staff at Temple University School of Medicine. Antihuman CD62P monoclonal antibodies were purchased from BD-Pharmingen (San Jose, CA, USA). The stable thromboxane analog 15(S)-hydroxy-9,11-epoxymethanoprosta-5Z,13E-dienoic acid (U46619) was purchased from Biomol (Plymouth Meeting, PA, USA). AR-C69931MX was a gift from Astra Zeneca Research Laboratories (Charnwood, Loughborough, UK). ADP, MRS 2179 and all other chemicals were purchased from Sigma Chemical Co. (St Louis, MO, USA).
Human blood was collected into a one-sixth volume of ACD (acid/citrate/dextrose; 2.5 g sodium citrate, 1.5 g citric acid, 2 g glucose in 100 mL deionized H2O). Human platelet-rich plasma (PRP) was isolated by centrifugation of citrated blood at 180 × g for 15 min at room temperature. PRP was incubated for 30 min at 37 °C in the presence of 1 mm acetylsalicylic acid, followed by centrifugation at 1000 × g for 10 min at room temperature. The platelet pellet was resuspended in HEPES-buffered Tyrode's solution (138 mm NaCl, 2.7 mm KCl, 1 mm MgCl2, 3 mm NaH2PO4, 5 mm glucose, 10 mm HEPES pH 7.4) supplemented with 0.2% bovine serum albumin and 0.05 U mL−1 apyrase.
Determination of α granule release
Platelet concentrations were adjusted to 1.5 × 107 cells mL−1 with FITC-conjugated anti-CD62P (P-selectin) antibody at a final concentration of 14 µg mL−1. Platelets were incubated with various agonists without stirring for 1 h. Samples were then diluted to 2 × 106 cells mL−1 with HEPES-buffered Tyrode's solution and analyzed with a Becton Dickinson FACScan benchtop analyzer (San Jose, CA, USA). Results are expressed as a percentage of maximal fluorescence with the response to thrombin set at 100%. Results are averages of at least three separate experiments.
ADP stimulation and α granule release in human platelets
Before investigating the role of P2Y receptors in α granule release, we wanted to confirm the results from earlier studies that showed α granule release could occur downstream of ADP stimulation in human platelets. The extent of α granule release was measured by determining the amount of P-selectin exposed on the platelet plasma membrane. Thrombin-mediated α granule release was used as a positive control. Stimulation of aspirin-treated washed platelets with 10 µm ADP resulted in a significant shift to the right, by approximately 35% of that due to thrombin stimulation (not shown). This indicated that ADP increased the expression of P-selectin on the membrane surface, suggesting α granule release. In order to determine if the α granule release response to ADP was saturable, we established the dose–response relation between ADP concentration and P-selectin expression. With increasing concentrations of ADP from 0.3 to 10 µm, there was a corresponding increase in P-selectin expression that appeared to plateau at 10 µm (Fig. 1). The above results confirmed the results of the previous studies and also established our experimental conditions.
Role of P2Y receptors in platelet α granule release
To determine the contribution of each ADP receptor in mediating P-selectin expression, we used either the P2Y12 antagonist AR-C69931MX or the P2Y1 antagonist MRS 2179 so that the non-antagonized receptor would be preferentially stimulated by ADP. ADP-mediated P-selectin expression and α granule release was affected under either P2Y12 or P2Y1 receptor antagonism (Fig. 2A). Furthermore, the effect of antagonizing the P2Y12 receptor had a more significant effect on ADP-induced α granule release compared with that of blocking the P2Y1 receptor. This observation suggests that the Gαi-coupled P2Y12 receptor plays a more important role in ADP-mediated α granule release compared with the Gαq-coupled P2Y1 receptor. Finally, blocking both ADP receptors resulted in complete inhibition of P-selectin expression and α granule release (Fig. 2B). These results showed that both P2Y1 and P2Y12 receptors are important for ADP-mediated α granule secretion, with the P2Y12 receptor playing a more dominant role compared with the P2Y1 receptor.
Contribution of ADP in thromboxane-mediated α granule release
The ability of thromboxane A2 to fully activate platelets requires the action of secreted ADP and epinephrine to mediate the Gαi signaling component necessary for platelet activation [9,14]. In particular, the role of the P2Y12 receptor in thromboxane A2-mediated platelet dense granule secretion and fibrinogen receptor activation is well established [9,14]. In order to determine the extent of ADP involvement in thromboxane-mediated α granule release, we examined P-selectin expression in response to increasing concentrations of the synthetic thromboxane A2 analog U46619 in the presence and absence of the ADP receptor antagonists. Blocking P2Y1 signaling with MRS 2179 had little effect on U46619-mediated P-selectin expression in comparison with untreated controls (Fig. 3). In contrast, blocking P2Y12 signaling with AR-C69931MX resulted in nearly a 3-fold decrease in U46619-mediated P-selectin expression. Furthermore, blocking both the ADP receptors inhibited P-selectin expression to the same extent as that of blocking P2Y12 receptor with AR-C69931MX alone, which was to a level of approximately 15% (Fig. 3).
Effect of Gαq, Gαi and Gα12/13 pathways in agonist-induced α granule release
Recent studies have demonstrated that Gα12/13 and Gαi pathways can synergize to activate platelets [15,16]. Since the TP thromboxane receptor couples to Gα12/13 as well as Gαq, we examined the possible synergy between Gαi and Gα12/13 signaling in platelet α granule release. The Gα12/13 signaling pathway was stimulated with either the PAR1 partial agonist YFLLRNP (60 µm) or U46619 (10 nm) at concentrations that selectively stimulate the Gα12/13 signaling without any contribution from Gαq. At the same time, varying concentrations of epinephrine were added along with the Gα12/13 stimulatory agonists to initiate signaling via the Gαi family member Gαz. As shown in Fig. 4, in comparison with Gαz stimulation alone, the combination of Gα12/13 and Gαz signaling caused significant increases in P-selectin expression under either conditions of Gα12/13 stimulation with YFLLRNP or U46619.
If concomitant stimulation of the Gαq and Gαi signaling pathways is needed for α granule release as implied by the results in Fig. 2, then the simultaneous stimulation of both signaling pathways with separate unrelated agonists should produce the same result. With regard to G protein signaling not mediated by nucleotides, the combination of both serotonin (to activate Gαq via the 5HT2A receptor) and epinephrine (Gαz) resulted in a significant increase in α granule release compared with treatment with either agonist alone (Fig. 5); it should be noted that treatment with epinephrine alone produced a significant increase in P-selectin expression over control and was comparable to what is seen in Fig. 4.
Effect of Gα12/13 pathways on α granule release mediated by Gαq and Gαi pathways
As seen from the previous results, Gα12/13 or Gαq can synergize with Gαi pathways and cause α granule release. Under physiological conditions, all three pathways are activated due to the action of various agonists. Hence we wanted to investigate the role of Gα12/13 in Gαq- and Gαi-mediated α granule release. Washed human platelets were treated either with ADP (Gαq plus Gαi) or simultaneously with serotonin (Gαq) and epinephrine (Gαz), in the absence or presence of YFLLRNP (Gα12/13). When added alone, YFLLRNP did not significantly increase P-selectin expression over control values. However, when YFLLRNP was added in combination with ADP, P-selectin expression was potentiated compared with ADP treatment alone (Fig. 6A). Furthermore, the combination of YFLLRNP and ADP-mediated Gαi signaling (i.e. in the presence of MRS 2179) resulted in increased P-selectin expression over that from ADP-mediated signaling alone. Platelets treated with serotonin, epinephrine and YFLLRNP demonstrated potentiated P-selectin expression in reference to the combination of serotonin and epinephrine (Fig. 6B).
Role of PI 3-kinase in α granule release downstream of Gαi signaling
The Gαi family of G proteins seem to be important among the different G protein pathways that mediate α granule secretion in platelets. Given this predominance of the Gαi family signaling pathway in α granule release, we wanted to investigate the downstream effectors of Gαi-mediated α granule release. A widely studied downstream target of Gαi signaling is phosphatidyl inositol 3-kinase (PI 3-K), which phosphorylates PI-4,5-P2 to PI-3,4,5-P3, and is in turn an important signaling molecule in cytoskeletal rearrangement [18,19]. In an effort to identify the role of PI 3-K in Gαi-mediated α granule release, we used the PI 3-K-specific inhibitors wortmannin and LY294002 and observed their effects on ADP-mediated P-selectin expression. In comparison with ADP stimulation alone, pretreatment of platelets with either wortmannin or LY294002 inhibited ADP-mediated P-selectin expression to basal levels (Fig. 7). These results showed that PI-3 kinase plays an important role in Gαi-mediated α granule release.
The reliance upon both P2Y receptors for ADP-mediated α granule release belies a common emerging theme in platelet activation, namely the coactivation of Gαq and Gαi signaling pathways to achieve full activation [10,20–22]. Of the two ADP-induced G protein-coupled signaling pathways, Gαi stimulation through activation of P2Y12 appears to be more involved than P2Y1-mediated Gαq activation in α granule release. The sharper decline in P-selectin expression in the presence of the P2Y12 antagonist AR-C69931MX compared with that with the P2Y1 antagonist MRS 2179 affirms the importance of P2Y12 in ADP-mediated α granule release as reported in previous studies [23–25]. Each antagonist was included at increasing concentrations to levels that have been shown previously to be effective in antagonizing the respective ADP receptors [26,27].
ADP can cause the formation of thromboxane A2 in platelets , and yet the secretion of additional ADP from platelet dense granules does not occur when thromboxane A2-mediated signaling is blocked . ATP, another nucleotide component of the dense granules, activates the P2X1 calcium channel and has been shown to be involved in platelet aggregation and secretion induced by low levels of collagen . It should be noted, however, that we have ruled out the participation of endogenously produced thromboxane, and hence any resulting dense granule secretion, in our studies by treating the platelets with aspirin to inhibit cyclooxygenase activity prior to experimentation.
The TP thromboxane receptor does not couple to Gαi directly; instead, it relies upon activation of the P2Y12 and α2A receptors for Gαi and Gαz activation by ADP and epinephrine, respectively, via dense granule secretion. The G protein signaling coupled to stimulation by ADP and epinephrine then synergizes with thromboxane-mediated Gαq stimulation to effect platelet activation . From the results shown in Fig. 3, it appears that thromboxane-induced α granule release is highly dependent upon secreted ADP. Blocking the activation of P2Y12 with AR-C69931MX reduces U46619-mediated P-selectin expression from about 40% to approximately 15% of maximal levels. However, antagonism of P2Y1 with MRS 2179 has little if any effect on U46619-mediated α granule release.
In addition to coupling with Gαq, the TP receptor also couples with Gα12/13, which is important in mediating platelet shape change in response to low concentrations of thromboxane [8,31]. Stimulation of Gα12/13 with either YFLLRNP or low concentrations of U46619, in combination with Gαz stimulation from epinephrine, results in a level of P-selectin expression that is comparable to that shown in response to higher concentrations of U46619 in the presence of both ADP receptor antagonists (Fig. 4). Indeed, the costimulation of Gαi and Gα12/13 has been previously shown to cause integrin αIIbβ3 activation and platelet aggregation [15,16]. Therefore, the extent of U46619-mediated P-selectin expression that remains without the contribution of ADP appears to be due to synergy between the thromboxane-induced Gα12/13 and epinephrine-induced Gαz signaling pathways. It should be noted that ADP is also able to stimulate the Gα12/13 pathway, although it occurs downstream of Gαq stimulation via activation of P2Y1 .
PI 3-K is an important signaling molecule involved in reorganization of the cytoskeleton, which in turn is implicated in platelet secretion [33,34]. The complete loss of ADP-mediated P-selectin expression upon inhibition of PI 3-K is somewhat surprising given that only partial inhibition of ADP- and thromboxane-mediated platelet activation occurs in the absence of PI 3-K activity [14,35,36]. While ADP causes α granule secretion in platelets, it does not cause dense granule secretion. However, thromboxane-mediated platelet signaling is highly dependent upon further signaling from ADP secreted from dense granules . A more complete role for PI 3-K in platelet activation awaits additional work in this area.
In conclusion, this paper demonstrates that concomitant stimulation of the Gαq and Gαi signaling pathways is required for α granule release, while the Gα12/13 signaling pathway potentiates this response. The molecular mechanism by which α granule secretion occurs is incompletely understood and represents an active field of study. Of particular interest will be the identification of other potential pathways downstream of Gαi stimulation given the apparent predominant role of Gαi signaling in this process. The signaling mechanisms downstream of the P2Y12 activation and Gαi stimulation as they relate to platelet activation are still largely unknown and under active study. Future studies involving downstream signaling molecules are needed to help resolve the participation of Gαi in α granule release.
The authors thank R. Dorsam for his critical review of the manuscript. This work was supported by Research Grants HL64943 and HL60683 from the National Institutes of Health (S.P.K.).