P2Y12 protects platelets from apoptosis via PI3k-dependent Bak/Bax inactivation

Authors

  • S. ZHANG,

    1. Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai
    Search for more papers by this author
    • These two authors contributed equally to this work.

  • J. YE,

    1. Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai
    Search for more papers by this author
    • These two authors contributed equally to this work.

  • Y. ZHANG,

    1. Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai
    Search for more papers by this author
  • X. XU,

    1. Department of Clinical Laboratory, Fudan University Shanghai Cancer Center, Shanghai
    Search for more papers by this author
  • J. LIU,

    1. Department of Biochemistry and Molecular & Cell Biology, Key Laboratory of Cell Differentiation and Apoptosis of Chinese Ministry of Education, Institute of Medical Science, Shanghai Jiao Tong University School of Medicine, Shanghai, China
    Search for more papers by this author
  • S. H. ZHANG,

    1. Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai
    Search for more papers by this author
  • S. P. KUNAPULI,

    1. Department of Physiology and the Sol Sherry Thrombosis Research Center, Temple University School of Medicine, Philadelphia, PA, USA
    Search for more papers by this author
  • Z. DING

    1. Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai
    Search for more papers by this author

Zhongren Ding, Key Laboratory of Molecular Medicine, Ministry of Education, and Department of Biochemistry and Molecular Biology, Fudan University Shanghai Medical College, Shanghai 200032, China.
Tel.: +86 21 5423 7896; fax: +86 21 6403 3738.
E-mail: dingzr@fudan.edu.cn

Abstract

Summary.  Background: Platelet ADP receptor P2Y12 is well studied and recognized as a key player in platelet activation, hemostasis and thrombosis. However, the role of P2Y12 in platelet apoptosis remains unknown. Objectives: To evaluate the role of the P2Y12 receptor in platelet apoptosis. Methods: We used flow cytometry and Western blotting to assess apoptotic events in platelets treated with ABT-737 or ABT-263, and stored at 37 °C, combined with P2Y12 receptor antagonists or P2Y12-deficient mice. Results: P2Y12 activation attenuated apoptosis induced by ABT-737 in human and mouse platelets in vitro, evidenced by reduced phosphatidylserine (PS) exposure, diminished depolarization of mitochondrial inner transmembrane potential (ΔΨm) and decreased caspase-3 activation. Through increasing the phosphorylation level of Akt and Bad, and changing the interaction between different Bcl-2 family proteins, P2Y12 activation inactivated Bak/Bax. This antiapoptotic effect could be abolished by P2Y12 antagonism or PI3K inhibition. We also observed the antiapoptotic effect of P2Y12 activation in platelets stored at 37 °C. P2Y12 activation improved the impaired activation responses of apoptotic platelets stressed by ABT-737. In platelets from mice dosed with ABT-263 in vivo, clopidogrel or deficiency of P2Y12 receptor enhanced apoptosis along with increased Bak/Bax activation. Conclusions: This study demonstrates that P2Y12 activation protects platelets from apoptosis via PI3k-dependent Bak/Bax inactivation, which may be physiologically important to counter the proapoptotic challenge. Our findings that P2Y12 blockade exaggerates platelet apoptosis induced by ABT-263 (Navitoclax) also imply a novel drug interaction of ABT-263 and P2Y12 antagonists.

Introduction

Platelets are small, anucleated fragments shed by megakaryocytes and circulate in the blood with an average lifespan of 10 days. Triggered by physiological and pathological stimuli, platelets are activated and undergo adherence, shape change, aggregation, granule release and TXA2 synthesis, which play critical roles in hemostasis and thrombosis. Apoptosis is genetically programmed cell death morphologically characterized by nuclear condensation and cytoplasmic shrinkage. Recent studies have demonstrated that in spite of being anucleate, depending on the stimulus levels, platelets also experience apoptosis, a series of responses different from activation [1] including PS exposure on the platelet surface, depolarization of mitochondrial inner transmembrane potential, activation of caspase-3, Bax translocation and cytochrome C release [2–6]. Platelet apoptosis plays an important role in platelet clearance from the circulation, thus determines the life span of platelets. Enhanced platelet apoptosis has been implicated in a number of pathological processes, such as type 2 diabetes [7], immune thrombocytopenia [8], Bernard–Soulier syndrome [9] and thrombocytopenic purpura induced by Helicobacter pylori [10] or malaria [11].

Considerable efforts have been made to explore the mechanism underlying platelet apoptosis. Among the platelet surface receptors intensively studied in platelet activation, PAR1, GPIb and GPIIb/IIIa have been reported to participate in platelet apoptosis [4,12,13]. As a target of successful antiplatelet drugs, the P2Y12 receptor plays a pivotal role in platelet activation [14] and has been thoroughly studied in platelet activation; however, the role of the P2Y12 receptor in platelet apoptosis has not been reported. In this study, we sought to explore the role of the P2Y12 receptor in platelet apoptosis and the underlying mechanism. ABT-263 (Navitoclax), the orally available derivative of ABT-737, a Bcl-xL-inhibitory BH3 mimetic, is now under clinical evaluation for the treatment of leukemia and other malignancies [15]. As a new class of proapoptotic agents, both ABT-737 and ABT-263 induce rapid platelet apoptosis and death [16]. Using platelets treated with BH3 mimetic ABT-737 or ABT-263, we found that P2Y12 activation protects human and mouse platelets against apoptosis via PI3k-dependent Bak/Bax inactivation in vitro. Importantly, we found that P2Y12 blockade exaggerates platelet apoptosis induced by ABT-263 in vivo, implying a novel drug interaction of ABT-263 and P2Y12 antagonists. The similar antiapoptotic effects of P2Y12 activation was observed in platelets stored at 37 °C.

Materials and methods

Materials

MRS2179, wortmannin, LY294002, Q-VD-OPh, serotonin and anti-G-actin antibody were purchased from Sigma (St Louis, MO, USA). ADP, thrombin and collagen were purchased from Chrono-Log (Havertown, PA, USA). AR-C69931MX was a gift from AstraZeneca (Loughborough, United Kingdom). Clopidogrel was from Sanofi-Aventis (Hangzhou, China). ABT-737, ABT-263, anti-Akt, anti-phospho-Akt (Ser-473), anti-Bad, anti-phospho-Bad (Ser-112 and Ser-136), anti-pan 14-3-3, anti Bcl-xL, anti-Bak, anti-Bax, HRP-labeled goat anti-rabbit and goat anti-mouse secondary antibody were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against G-actin and COX IV were from Invitrogen (Gaithersburg, MD, USA). PE P-selectin and fluorescein isothiocyanate (FITC) PAC-1 antibodies were purchased from BD Biosciences (Heidelberg, Germany). Annexin V-FITC and 5,5′,6,6′-tetrachloro-1,1′,3,3′- tetraethylbenzimidazolylcarbocyanine chloride (JC-1) were purchased from BioVision (Mountain View, CA, USA). Pro-light HRP Enhanced Chemiluminescence Detection Reagent was from Tiangen Biotech (Beijing, China).

Platelet preparation and cell culture

All experiments using human subjects were performed in accordance with the Declaration of Helsinki and approved by the Institutional Review Board Fudan University. Only healthy volunteers who had not taken aspirin or other non-steroidal anti-inflammatory drugs for at least 14 days were recruited and informed consent was obtained before blood collection. Platelets were separated as described previously [17]. In briefly, fresh blood was anticoagulated with a 1/7 volume of acid-citrate-dextrose (2.5% trisodium citrate, 2.0% glucose and 1.5% citric acid). After centrifugation, isolated platelets were washed twice with CGS buffer (0.123 m NaCl, 0.033 m glucose and 0.013 m trisodium citrate, pH 6.5), then centrifuged at 600 ×g for 5 min, resuspended in Tyrode’s buffer (138 mm NaCl, 2.7 mm KCl, 2 mm MgCl2, 0.42 mm NaH2PO4, 5 mm glucose, 10 mm HEPES, 0.2% bovine serum albumin and 0.02 U mL−1 apyrase, pH 7.4) to a final concentration of 5 × 107 mL−1, supplemented with 1 mm CaCl2, and then incubated at room temperature for 1 h to recover to a resting state. HeLa cells from ATCC were cultured in Dubelcco’s Modified Eagle Medium supplemented with 10% fetal bovine serum.

Platelet treatment

For P2Y12 activation, washed platelets (5 × 107 mL−1) were incubated with MRS2179 100 μm for 1 h, then treated with ADP 5 μm for 3 min. For ABT-737-induced platelet apoptosis, washed platelets (5 × 107 mL−1) were treated with different concentrations of ABT-737 for indicated times. For PI3K inhibition, washed platelets were treated with wortmannin (100 or 1 μm) or LY294002 (25 μm) for 10 min. For P2Y12 antagonism, washed platelets were treated with AR-C69931MX 100 nm for 10 min. For caspase inhibition, washed platelets were treated with Q-VO-OPh (50 μm) for 15 min. Non-stirring conditions were employed to avoid platelet aggregation in the above work [2].

Animal studies

All animal experiments were performed according to the Guide for the Care and Use of Medical Laboratory Animals (Ministry of Health, China, 1998) and with the ethical approval of the Shanghai Medical Laboratory Animal Care and Use Committee as well as the Ethical Committee of Fudan University. Three groups of wild-type C57Bl6/J mice and one group of P2Y12 knockout mice were used (n = 6 in each group). Except for the wild-type control group, all the other three groups were administered with ABT-263 (25 mg kg−1) via oral gavage 4 h before blood collection. For the clopidogrel treated wild-type group, mice were dosed with clopidogrel (30 mg kg−1) once daily for 3 days before ABT-263 treatment. Mice were anesthetized with pentobarbital sodium and blood was collected from the abdominal aorta into syringes containing 3.8% citrate (9:1). Blood was pooled and washed platelets were prepared and resuspended in Tyrode’s buffer at a final concentration of approximately 106 μL−1 as described previously [17]. To study ABT-263-induced thrombocytopenia, blood was collected into tubes containing potassium EDTA and platelet numbers were then determined on a Sysmex XS-800i automated cell counter (Sysmex Corporation, Kobe, Japan).

Measurement of platelet aggregation

Platelet aggregation was analyzed using a lumi-aggregometer (Model 700; Chrono-Log, Haverston, PA, USA) under stirring conditions (900 rpm) at 37 °C. Agonists were added to initiate aggregation for 3 min. The baseline was set using Tyrode’s buffer as a blank.

Measurement of P-selectin expression and PAC-1 binding by flow cytometry

Washed human platelets were treated with thrombin and collagen in the presence of excessive amounts of FITC-conjugated monoclonal antibody (PAC-1) and PE-conjugated monoclonal antibody (P-selectin) for 20 min at room temperature. The samples were then fixed at 4 °C with 1% paraformaldehyde, and then analyzed by FACSCalibur flow cytometer (BD Biosciences, San Jose, USA) using dual-color analysis. The levels of P-selectin expression and PAC-1 binding were expressed as the percentages of positive cells. The negative cut-off for each antibody was set using resting platelets that gave < 1% of cells positive for binding of P-selectin or PAC-1.

Measurement of phosphatidylserine exposure

Washed platelets (5 × 107 mL−1) were mixed with binding buffer (10 mm HEPES, 10 mm NaOH, 140 mm NaCl, 2.5 mm CaCl2, pH 7.4), PE-conjugated anti-CD61 (clone VI-PL2; eBioscience, San Diego, CA, USA) and annexin V-FITC at a ratio of 10:50:1:1 [3]. Samples were gently mixed by rocking, incubated at room temperature for 20 min in the dark and then analyzed by FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA) using dual-color analysis to acquire platelet (CD61)-positive events.

Assay of the dissipation of mitochondrial membrane potential

Mitochondrial depolarization was determined with JC-1. In brief, washed platelets were combined with JC-1 (1 mg mL−1, dissolved in dimethylsulfoxide) for 15 min. The final volume was brought to 500 μL for immediate analysis using a FACSCalibur flow cytometer.

Caspase-3 activity assay

Caspase-3 activity assay was performed in 96-well microtiter plates by incubating 10 μL platelet lysate per sample in 80 μL reaction buffer and 10 μL caspase-3 substrate (Ac-DEVD-pNA) (2 mm). Samples were incubated at 37 °C for 4 h and were measured with an ELISA reader at an absorbance of 405 nm. The specific caspase-3 activity, normalized to total proteins of platelet lysates, was expressed as a fold of the baseline caspase-3 activity of the vehicle-treated sample.

Subcellular fractionation

The mitochondrial fraction of platelets was isolated using a mitochondria isolation kit (Pierce, Rockford, IL, USA) according to the manufacturer’s instructions. Briefly, washed platelets were suspended in mitochondrion isolation buffer A for 3 min, buffer B for 5 min and buffer C for another 5 min. The supernatant was then centrifuged at 2000 × g for 10 min at 4 °C. The supernatant was subjected to further centrifugation at 12 000 × g for 15 min to yield the mitochondrial pellet and the mitochondria-free cytosolic fraction (supernatant).

Western blotting and immunoprecipitation

Washed platelets were lysed in RIPA buffer. Cell lysates (60 or 120 μg) were subjected to SDS–PAGE, electroblotted onto poly(vinylidene difluoride) membranes, incubated with specific antibodies and detected with electrochemiluminescence. To determine Bak dimerization, total proteins were run under non-reducing conditions. For immunoprecipitation, crosslinked antibodies were incubated with platelet lysate for 2 h at 4 °C. Beads were washed with lysis buffer containing 1% CHAPS before elution in SDS-loading dye and Western blotting.

Platelet viability assay

Viable platelets were measured by a colorimetric assay, based on measuring the mitochondrial reduction of MTT to formazan. MTT dissolved in Tyrode’s buffer at 5 mg mL−1 (10 μL) was added to the platelet suspension (100 μL). The samples were incubated for 2 h. The supplements were then removed and the formazan crystals were solubilized by adding formazan solubilization solution (100 μL). Metabolic activity was quantified by measuring light absorbance at 570 nm.

Statistical analysis

All data are expressed as mean ± standard error of the mean (SEM). Differences between the groups were analyzed by anova followed by a Newman–Keuls test using GraphPad Prism version 4.0 (Graphpad Inc, San Diego, CA, USA). P-values < 0.05 were considered statistically significant.

Results

P2Y12 receptor activation ameliorates ABT-737-induced apoptotic responses in human platelets

Among the typical apoptosis events, PS exposure is a key feature and commonly used as a platelet apoptotic marker [2,3], whereas mitochondrial membrane potential depolarization is an early apoptosis indicator reflecting mitochondrial disruption [2], and caspase-3 is the main executioner of apoptosis [18,19]; therefore, we used these three typical apoptosis events to evaluate the effects of P2Y12 activation in platelet apoptosis.

We first confirmed the proapoptotic effects of BH3 mimetic ABT-737 and explored the role of P2Y12 activation in ABT-737-induced platelet apoptosis. Apoptosis induced by ABT-737 occurs much more slowly than that induced by agonists, which occurs within minutes of agonist exposure. After 1 h of exposure to ABT-737, only low levels of apoptosis were detected, as assessed by PS externalization, ΔΨm depolarization and caspase-3 activation; 3 h of exposure to ABT-737 induced extensive platelet apoptosis concentration and time dependently (Fig. 1A,B). At almost all the concentration and time points treated with ABT-737, P2Y12 receptor activation (stimulated with ADP in the presence of P2Y1 receptor antagonist MRS2179) significantly inhibited apoptosis events and caspase inhibition (treated with Q-VO-PDh) almost abolished apoptosis events in human platelets. In contrast, stimulation with thrombin or collagen did not inhibit ABT-737-induced apoptosis events in human platelets (Fig. S1).

Figure 1.

 Protective effects of P2Y12 activation on apoptosis induced by ABT-737 in human platelets. (A) Human washed platelets were treated with different concentrations of ABT-737 for 3 h in the absence or presence of P2Y12 activation or caspase inhibition. (B) Human washed platelets were treated by 10 μM ABT-737 for different times in the absence or presence of P2Y12 activation or caspase inhibition. The typical apoptotic events in platelets were determined. (C–E) Human washed platelets were pretreated with or without PIK3 inhibitor wortmannin (100 nM), LY294002 (25 uM) or P2Y12 agonist AR-C69931MX (ARC, 100 nM) for 10 min, then incubated with ABT-737 (10 μM) for 3 h in the absence or presence of P2Y12 activation. The typical apoptotic events of (C) PS exposure, (D) mitochondrial inner transmembrane potential (ΔΨm) depolarization, and (E) caspase-3 activity in platelets were determined as described in Materials and methods. Data are represented as Means ± standard error of the mean (SEM) from three to four independent experiments. *< 0.05 and **< 0.01 compared with the control group; #< 0.05 and ##< 0.01 compared with the ABT-737+ MRS2179+ ADP treated group. Representative flow cytometric histogram of PS exposure and dot plots of ΔΨm depolarization were also shown in panel C and D, respectively.

The most prominent proapoptotic effects of ABT-737 and the antiapoptotic effects of P2Y12 activation happened at 3 h after exposure to ABT-737 10 μm (Fig. 1A,B), we therefore treated platelets with ABT-737 at 10 μm for 3 h for further studies. Further analysis of these apoptotic markers showed that pretreatment with a P2Y12 antagonist ARC-69931MX abolished P2Y12 receptor-mediated protective effects (Fig. 1C–E), suggesting that the protective role is mediated by P2Y12; meanwhile, PI3K inhibitor wortmannin or LY294002 treatment achieved comparable efficacy as AR-C69931MX (Fig. 1C–E), suggesting that the protective effects of P2Y12 activation in platelets is PI3K dependent.

P2Y12 protects human platelets from apoptosis via PI3K-Akt pathway

The antiapoptotic role of the PI3K/Akt pathway has been reported in several cell types including platelets [20]. P2Y12-mediated Akt activation through phosphorylation of Ser-473 is well documented in platelets [21,22]. Therefore, we assessed Akt Ser-473 phosphorylation to characterize the underlying mechanism of P2Y12 activation in platelets. ABT-737 time-dependently decreased Akt phosphorylation [23], which was prevented by P2Y12 activation (Fig. 2A). Both the effects of P2Y12 activation on Akt phosphorylation and apoptosis are counteracted by PI3K inhibitor wortmannin or P2Y12 antagonist ARC-69931MX (Fig. 2B), suggesting the involvement of the PI3K/Akt pathway in P2Y12-mediated antiapoptotic effects in platelets.

Figure 2.

 P2Y12 activation inhibits ABT-737-induced Akt and Bad dephosphorylation in human platelets. (A) Human washed platelets were treated with ABT-737 (10 μM) for different times in the absence or presence of P2Y12 activation (MRS2179 plus ADP stimulation). Whole cell lysates were harvested and analyzed by Western blots with indicated antibodies. (B) Human washed platelets were pretreated with or without PIK3 inhibitor wortmannin (1 μM) or P2Y12 agonist AR-C69931MX (ARC, 100 nM) for 10 min, then incubated with ABT-737 (10 μM) for 3 h in the absence or presence of P2Y12 activation. Whole cell lysates were analyzed by Western blots with indicated antibodies. Data are presented from three independent experiments. Means ± standard error of the mean (SEM) of the densitometric measurement of Western blots from three independent experiments are shown. *< 0.05 and **< 0.01 compared with the control group; #< 0.05 and ##< 0.01 compared with the ABT-737+ MRS2179+ ADP-treated group.

One of the Akt downstream targets associated with cell survival is Bad, a proapoptotic Bcl-2 family protein, which is inactivated by binding to 14-3-3 proteins upon phosphorylation [24]. To further dissect the mechanism underlying the protective effects of P2Y12 activation on platelet apoptosis, we analyzed Bad phosphorylation in platelets. Bad phosphorylation at Ser-112 and Ser-136 was significantly decreased in ABT-737-treated human platelets in a time-dependent manner (Fig. 2A), whereas P2Y12 activation restored Bad phosphorylation at both sites. AR-C69931MX or wortmannin treatment blocked P2Y12-mediated restoration of Akt and Bad phosphorylation (Fig. 2B).

P2Y12 protects human platelets from apoptosis via PI3k-dependent Bak/bax inactivation

Phosphorylation of Bad provokes Bad binding to 14-3-3 proteins, which sequestrates Bad in cytoplasm, resultantly prevents Bad from entering the mitochondrion and heterodimerizing with Bcl-xL in nucleated cells [24]. To test whether such event occurs in anucleate platelets, we determined the effect of P2Y12 activation on Bad/14-3-3 binding and Bad/Bcl-xL complex levels. Coimmunoprecipitation studies revealed human platelets treated with ABT-737 had decreased the Bad/14-3-3 protein interaction (Fig. 3A) and increased Bad/Bcl-xL complex levels (Fig. 3B). Conversely, P2Y12 activation restored the Bad/14-3-3 association and caused dissociation of the Bad/Bcl-xL complex, which was abolished by AR-C69931MX or wortmannin pretreatment as expected (Fig. 3A,B).

Figure 3.

 P2Y12 activation leads to PI3k-dependent change of Bcl-2 family proteins in human platelets treated with ABT-737. Human washed platelets were treated under conditions identical to those used in Fig. 2B. Whole cell lysates were analyzed by immunoprecipitation and Western blots using indicated antibodies. The effect of ABT-737 on (A) the association between 14-3-3 and Bad, (B) the association between Bad and Bcl-xL, (C) the association between Bcl-xL and Bak, (D) the association between Bcl-xL and Bax, the association between 14-3-3 and Bax, (E) the dimerization of Bak and (F) the distribution of Bax in mitochondrial and cytosolic fractions were assessed as described in Methods. Data are presented from three independent experiments. Means ± standard error of the mean (SEM) of the densitometric measurement of Western blots from three independent experiments are shown. *< 0.05 and **< 0.01 compared with the control group; #< 0.05 and ##< 0.01 compared with the ABT-737 +  MRS2179 +  ADP-treated group.

As an important antiapoptotic protein, Bcl-xL inhibits apoptosis by competing with Bak or Bax, two proapoptotic proteins belonging to the Bcl-2 family, and neutralizes their proapoptotic activity in nucleate cells [25]. Therefore, we evaluated whether Bcl-xL coimmunoprecipitated with Bak or Bax in platelets. As shown in Fig. 3C, Bcl-xL associated with Bak in platelets. ABT-737 treatment significantly diminished the Bcl-xL/Bak binding, which was abrogated by P2Y12 activation and restored by AR-C69931MX or wortmannin treatment. Notably, no interactions of Bcl-xL/Bax and 14-3-3/Bax were detected in human platelets; in contrast, abundant Bcl-xL and 14-3-3 binding to Bax were seen in nucleated HeLa cells (Fig. 3D).

Bcl-xL is the major homeostatic regulator of platelet survival [26]. Recently, Kodama et al. [27] found either Bak or Bax was sufficient to cause platelet apoptosis in the absence of Bcl-xL, indicating both Bak and Bax are regulatory targets of Bcl-xL. Therefore, we sought to explore whether P2Y12 activation influences Bak dimerization, an active form of Bak [28], and Bax translocation from the cytosol to the mitochondrial [29]. As shown in Fig. 3E,F, Bak dimerization and Bax translocation occurred when platelet apoptosis was induced by ABT-737, whereas P2Y12 activation inhibited Bak dimmerization and Bax translocation. Furthermore, treatment with AR-C69931MX or wortmannin restored Bak dimerization and Bax translocation, further confirming that the antiapoptotic effect of P2Y12 activation is PI3K dependent.

P2Y12 activation improves impaired platelet activation responses of apoptotic platelets stressed by ABT-737

It has been reported that apoptotic platelets exposed to ABT-737 exhibited an impaired activation response to stimuli [10,16]. To determine the effect of P2Y12 activation on platelet activation function in apoptotic platelets, we measured platelet activation responses after exposure to ABT-737. Consistent with a previous report, ABT-737-treated human platelets showed significantly impaired activation responses to thrombin and collagen in comparison with control platelets, including aggregation, α-granule release and integrin αIIbβ3 activation, determined by the maximal aggregation rate, P-selectin expression and PAC-1 binding, respectively (Table 1). Nevertheless, prestimulation of the P2Y12 receptor improved the impaired activation responses stressed by ABT-737. Taken together, these data indicated that P2Y12 activation could improve activation responses of apoptotic platelets stressed by ABT-737 through its antiapoptotic or prothrombotic effect.

Table 1.   P2Y12 activation improves platelet activation responses of apoptotic platelets treated with ABT-737
  Thrombin (1 U mL−1) as agonistCollagen (2 μg mL−1) as agonist
Maximal aggregation (%)P-selectin positive (%)αIIbβ3 activation (%)Maximal aggregation (%)P-selectin positive (%)αIIbβ3 activation (%)
  1. Human washed platelets were incubated with ABT-737 (10 μm) for 3 h to induce apoptosis in the absence or presence of P2Y12 activation. The activation responses were determined. Before platelet function analysis, platelets were treated with AR-C69931MX and MRS2179 (no additional treatment if platelets have been treated with MRS2179 in the previous steps) to rule out the effect of ADP receptors on platelet function. Data presented are representative of three to four independent experiments. Means ± standard error of the mean (SEM) from three to four independent experiments are shown. *< 0.05 compared with the ABT-737 group.

Control81 ± 687 ± 677 ± 566 ± 783 ± 569 ± 6
ABT-73710 ± 317 ± 811 ± 515 ± 213 ± 417 ± 4
ABT-737 +  MRS + ADP29 ± 6*44 ± 9*45 ± 10*34 ± 4*37 ± 11*35 ± 5*

P2Y12 receptor activation inhibited spontaneous apoptotic cell death in stored human platelets

In 37 °C stored human platelets, PS exposure increases with caspase-3 activation, which leads to spontaneous apoptosis [30]. The intrinsic mitochondrial pathway mediated by Bak and Bax plays a central role in the spontaneous apoptosis induced by 37 °C storage [31]. Spontaneous platelet apoptosis shows quite similar characteristics to those triggered by ABT chemicals [27], we hence examined the effect of P2Y12 activation on platelet apoptosis induced by 37 °C storage. Similar to ABT-737-induced apoptosis, P2Y12 activation attenuated PS exposure (Fig. 4A), ΔΨm depolarization (Fig. 4B) and caspase-3 activation (Fig. 4C), and improved impaired platelet viability (Fig. 4D) induced by 37 °C storage, which was reversed by pretreatment with wortmannin, LY294002 or ARC-69931MX. P2Y12 activation also inhibited dephosphorylation of Akt and Bad (Fig. 4E,F) induced by 37 °C storage. Moreover, P2Y12 activation inhibited Bak dimerization (Fig. 4G) and Bax translocation from cytosol to mitochondria (Fig. 4H) triggered by 37 °C storage, which was restored by pretreatment with wortmannin or ARC-69931MX. Overall, these findings clearly indicated that P2Y12 activation could rescue human platelets from apoptosis provoked by 37 °C storage via PI3K-dependent Bak and Bax inactivation.

Figure 4.

 Protective effects of P2Y12 activation on spontaneous apoptosis in human platelets induced by 37 °C storage. Human washed platelets were pretreated with or without PIK3 inhibitor wortmannin (100 nM), LY294002 (25 uM) or P2Y12 agonist AR-C69931MX (ARC, 100 nM) for 10 min. Platelets were further stored in 37 °C for 3 days in the absence or presence of P2Y12 activation. The effects of 37 °C storage on (A) PS exposure, (B) ΔΨm depolarization, (C) caspase-3 activation and (D) platelet viability were assessed. (E) Human washed platelets were stored in 37 °C for different times in the absence or presence of P2Y12 activation. Phosphorylation of Akt (Ser-473) was assessed. (F–H) Human washed platelets were pretreated with or without PIK3 inhibitor wortmannin (1 μM) or P2Y12 agonist AR-C69931MX (ARC, 100 nM) for 10 min. Platelets were further stored in 37 °C for 3 days in the absence or presence of P2Y12 activation. The effects of 37 °C storage on (F) phosphorylation of Bad (Ser-112 and Ser-136), (G) Bak dimerization and (H) the distribution of Bax in mitochondrial and cytosolic fractions were assessed as described in Methods. Data presented are representative of three to four independent experiments. Means ± standard error of the mean (SEM) from three to four independent experiments are shown. *< 0.05 and **< 0.01 compared with the control group; #< 0.05 and ##< 0.01 compared with the 37 °C storage + MRS2179+ ADP-treated group.

P2Y12 blockade enhanced ABT-263 induced platelets apoptosis in vivo

ABT-737 and ABT-263 (Navitoclax) have been reported to induce mouse platelet apoptosis in vitro and in vivo, respectively [19]. ABT-263, a second-generation analog of the BH3-mimetic ABT-737 designed to overcome the poor oral bioavailability of ABT-737, is now in clinical trials to treat leukemia and other malignancies [32]. Both ABT-737 and ABT-263 have been regarded as interchangeable because they both bind Bcl-2 and Bcl-xL with high affinity without inhibiting Mcl1 or Bcl2A1 [19]. We further investigated the antiapoptotic effects of P2Y12 activation in mouse platelets. P2Y12 receptor activation inhibited the typical apoptosis events in mouse platelets treated with ABT-737 in vitro, which was abolished by treatment with wortmannin, LY294002 or ARC-69931MX (Fig. S2). To investigate whether P2Y12 receptor pathway affects platelet apoptosis in vivo, we administered ABT-263 orally to the mice. Platelets from C57Bl6/J mice dosed with ABT-263 exhibited typical apoptotic changes such as PS exposure (Fig. 5A), ΔΨm depolarization (Fig. 5B) and caspase-3 activation (Fig. 5C). These apoptotic changes were enhanced in P2Y12 knockout mice and clopidogrel-dosed mice. Moreover, ABT-263 dosing increased Akt dephosphorylation (Fig. 5D), Bak dimerization (Fig. 5E) and Bax translocation to mitochondria (Fig. 5F), which were further enhanced in P2Y12 knockout and clopidogrel-dosed mice. Interestingly, ABT-263-induced thrombocytopenia was also enhanced in P2Y12 knockout and clopidogrel-dosed mice (Fig. 6). These results extended our in vitro findings to in vivo settings and imply that P2Y12 receptor is an endogenous protective factor against platelet apoptosis.

Figure 5.

 P2Y12 blocking exaggerates platelet apoptosis in mice dosed with ABT-263. C57Bl6/J wild-type and P2Y12 knockout mice were treated as described in Animal studies. The effect of ABT-263 administration on (A) phosphatidylserine (PS) exposure, (B) mitochondrial depolarization, (C) caspase-3 activation, (D) phosphorylation of Akt (Ser-473), (E) Bak dimerization, and (F) the distribution of Bax in mitochondrial and cytosolic fractions were assessed as described in the Methods. Data presented are representative of three to four individual experiments. Means ± standard error of the mean (SEM) of flow cytometry or densitometric measurements of Western blot data are also shown at the bottom of the typical flow cytometry or Western blot results. *< 0.05 and **< 0.01 compared with the control group; #< 0.05 and ##< 0.01 compared with the ABT-263-treated group.

Figure 6.

 P2Y12 deficiency exaggerates ABT-263-induced thrombocytopenia. C57Bl6/J wild-type mice and P2Y12 knockout mice were treated as described in Animal studies. Blood was collected before (A) and 6 h after ABT-263 administration (B), and platelet numbers were determined. Data are expressed as means ± standard error of the mean (SEM) from five mice. *< 0.05 compared with the ABT-263 (P2Y12+/+) group.

Discussion

P2Y12 receptor plays a central role in platelet activation including aggregation, granule release and TXA2 production. In this study, we showed that P2Y12 protects platelets from apoptosis as evidenced by reduced PS exposure, diminished ΔΨm depolarization and decreased caspase-3 activation in human and mouse platelets in vitro and in vivo, consistent with that P2Y12 activation protects astrocytes from apoptosis [33].

In comparison with the protective effects of P2Y12 activation on platelet apoptosis here, Tonon et al. [34] reported apoptosis-like events in platelets stimulated with ADP in the absence of phosphatidylserine exposure. ADP activates both proapoptotic Gq-coupled P2Y1 and antiapoptotic Gi-coupled P2Y12, as demonstrated in nucleate cells [33]. We observed P2Y1 activation alone led to weak PS exposure in platelets, which was antagonized by P2Y12 activation (Fig. S3). Similarly, serotonin which only activates Gq-coupled receptor 5HT2A in platelets, also induced weak PS exposure in platelets, agreeing with a previous report [35].

The PI3K-Akt signal axis has been shown to be a dominant survival pathway in different nucleate cell types. The dephosphorylation and phosphorylation of PI3K-Akt in response of diverse stimuli regulate the balance between apoptosis and survival [36]. In our in vitro study, the protective effect of P2Y12 is dependent on the signaling mediated by PI3K-Akt pathway, as demonstrated by the inhibitory effects of specific PI3K inhibitors. Using Western blot, we were able to show that Akt and Bad were dephosphorylated in both ABT-737-stressed and 37 °C stored platelets, whereas P2Y12 activation increased phosphorylation levels of Akt and Bad. In immunoprecipitation assays, we found P2Y12 activation potentiated Bad binding to 14-3-3 as a consequence of Akt and Bad phosphorylation, hence displaced Bcl-xL from Bad and permitted free Bcl-xL to exert its prosurvival role.

Akt becomes activated upon platelet stimulation with diverse agonists, such as collagen, ristocetin/von Willebrand factor, thrombin, arachidonic acid and platelet-activating factor [37,38]. In addition to Akt activation, these agonists simultaneously trigger calcium mobilization, which leads to PS exposure and other apoptotic-like events. Unlike these agonists, P2Y12 receptor activation induces Akt activation but does not initiate calcium mobilization, which implies a specific role of P2Y12 in protection against platelet apoptosis. In agreement with our observations of the PI3K-dependent antiapoptotic effect of P2Y12 activation, Maria et al. [20] reported that type-1 cannabinoid receptor activation inhibits platelet apoptosis by activating PI3K/Akt signaling. Taken together, these data suggested an important role of the PI3K-Akt signal in regulating survival and death in anucleate platelets.

Genetic studies have revealed that the major downstream effectors mediating platelet apoptosis are proapoptotic Bak and Bax [26,27,39]. In this study, we found that P2Y12 activation could exert an antiapoptotic effect through stabilization of the interaction of Bcl-xL and Bak, resultantly reducing Bak dimerization, the active status of Bak and the active status of Bak. Surprisingly, in spite of dramatically reduced Bax translocation to mitochondria upon P2Y12 activation, we were unable to detect an interaction between Bcl-xL and Bax in platelets, indicating that Bcl-xL does not inactivate Bax directly. A downstream protein of Bcl-xl or a converted form of Bcl-xl may interact with Bax and inactivate Bax.

In addition to release from platelet dense granules, ADP can be produced by hydrolysed ATP from damaged tissues and dying cells. Intact cells can also release ATP, which is metabolized to ADP under normal physiological conditions. Endogenous ADP regulates a myriad of cell functional responses via purinergic P2Y receptors including P2Y1 and P2Y12in vivo [40]. To better understand the specific role of the P2Y12 receptor on platelet apoptosis in vivo, we checked platelet apoptosis in ABT-263-dosed mice deficient in P2Y12 or pretreated with clopidogrel. Our in vivo studies demonstrated that P2Y12 blockade exaggerated ABT-263-induced platelet apoptosis, Bak dimerization and Bax translocation to mitochondria. These results extend our in vitro findings, strongly suggesting that the P2Y12 receptor is an important physiological antiapoptotic effector in platelets.

ABT-263 is under clinical trials to treat solid tumors and hematologic malignancies. Both clinical and animal studies have shown that the major toxicity of ABT-263 is transient thrombocytopenia owing to apoptosis of platelets [41]. Given the risk of cardiovascular events in tumor patients, anticancer therapy is often combined with antiplatelet treatment. In this context, the enhanced ABT-263-induced platelet apoptosis in P2Y12 knockout or clopidogrel-dosed mice raised a question whether antiplatelet treatment with the P2Y12 receptor antagonist could exacerbate thrombocytopenia in tumor patients taking ABT-263, which requires clarification in further clinical trials.

In conclusion, we have demonstrated that P2Y12 activation is able to prevent platelets from apoptosis. We further conclude that the PI3K-dependent Akt-Bad-14-3-3-Bcl-xL-Bak/Bax axis is responsive for the antiapoptotic role of P2Y12 activation (Outlined in Fig. 7). Based on the antiapoptotic effect of P2Y12 on platelet and astrocytes, specific P2Y12 receptor agonists could play a role in the prevention and treatment of thrombocytopenia and neurodegenerative diseases characteristic of platelets and glioma cell apoptosis. However, an increased thrombotic tendency owing to P2Y12 activation should be considered in such situations.

Figure 7.

 Possible regulation of platelet apoptosis by P2Y12 activation. ADP binding to P2Y12 receptor inhibits apoptosis by activating PI3K/Akt pathway: P2Y12 activation induced Akt and Bad phosphorylation. Upon phosphorylation by Akt, Bad is sequestered in the cytoplasm by the adapter protein 14-3-3, which prevents a Bad association with Bcl-xL. As a consequence, free Bcl-xL heterodimerizes with Bak proteins, prevents Bak dimerization in mitochondria, thus antagonizing its proapoptotic activity. In an indirect way, Bcl-xL inactivates Bax by inhibiting its translocation into mitochondria.

Acknowledgement

This work was partially supported by the National Natural Science Foundation of China (No. 81100344, 81000968, 30973529), the Scientific Research Foundations for the Returned Overseas Chinese Scholars, Education Ministry and State Personnel Ministry of China, and Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning, and a grant from National Institutes of Health National Heart, Lung and Blood Institute (grant HL60683).

Disclosure of Conflict of Interests

The authors state that they have no conflict of interest.

Ancillary