Anucleate platelets perform two fundamental processes, activation and apoptosis. We elaborated an approach for selective and concurrent stimulation of platelet apoptosis and/or activation, processes important in haemostasis and platelet clearance. Human platelets were treated with BH3 mimetic ABT-737, thrombin, calcium ionophore A23187 and matched diluents. Apoptosis was determined as mitochondrial inner membrane potential (ΔΨm) depolarization and activation as P-selectin exposure. At optimal treatment conditions (90–180 min, 37°C), ABT-737 predominantly induced apoptosis, when 77–81% platelets undergo only ΔΨm depolarization. The ABT-737 impact on ΔΨm depolarization is strongly time- and temperature-dependent, and much higher at 37°C than at room temperature. In contrast, when platelets were treated with thrombin for 15–90 min at either temperature, activation-only was predominantly (79–85%) induced, whereas A23187 triggers both apoptosis and activation (73–81%) when platelets were treated for 15–60 min at 37°C or 15–90 min at room temperature. These data demonstrate that, depending on the triggering stimulus, platelets predominantly undergo ΔΨm depolarization-only, P-selectin exposure-only, or both responses, indicating that platelet apoptosis and activation are different phenomena driven by different mechanisms. The described model provides a basis for studying differential pharmacological manipulation of platelet apoptosis and activation and their role in haemostasis, thrombosis and platelet clearance.
Platelet activation is an essential reaction contributing to multiple platelet functions (Smyth et al, 2009; Semple et al, 2011). Over the past fifteen years, it has been also recognized that, as with nucleated cells (Kroemer & Reed, 2000; Ashkenazi, 2002; Danial & Korsmeyer, 2004; Kroemer et al, 2007; Hotchkiss et al, 2009), anucleate platelets are able to undergo programmed cell death, apoptosis, in response to diverse stimuli (Vanags et al, 1997; Brown et al, 2000; Li et al, 2000; Leytin & Freedman, 2003; Gyulkhandanyan et al, 2012; Leytin, 2012). These fundamental platelet responses, activation and apoptosis, may play a key role in the major platelet-dependent processes, haemostasis and thrombosis, and in regulation of the platelet lifespan. However, the absence of a valid model for selective stimulation of platelet apoptosis and activation, or both responses, is a serious limitation for studying the role of apoptosis and activation in platelet clearance and haemostatic function.
To develop an approach to examining platelet activation versus apoptosis, we treated human platelets with three agents, a pro-apoptotic BH3-only mimetic, the anti-cancer drug ABT-737 (Oltersdorf et al, 2005; Mason et al, 2007; Zhang et al, 2007; Dasgupta et al, 2010; Mutlu et al, 2012), the potent platelet agonist thrombin (Coughlin, 2005; Lundblad & White, 2005; Leytin et al, 2006a, 2007; Bahou, 2007; Lopez et al, 2008) and calcium ionophore A23187 (Leytin et al, 2004, 2006a, 2009; Rand et al, 2004; Mutlu et al, 2012), known inducers of apoptosis and/or activation in nucleated cells and platelets. We concurrently determined an apoptosis marker, depolarization of mitochondrial inner transmembrane potential (ΔΨm) (Kroemer & Reed, 2000; Leytin et al, 2004, 2006a, 2007, 2009; Rand et al, 2004; Gyulkhandanyan et al, 2012), and a marker of platelet activation, exposure of P-selectin (CD62) on the platelet surface (Stenberg et al, 1985; Michelson et al, 1991; Leytin et al, 2000a, 2007). As the result of this study, we elaborated a model for selective and simultaneous triggering of platelet apoptotic and/or activation responses. This model can serve as an important experimental tool for elucidating the role of platelet apoptosis and activation in platelet clearance and haemostatic function.
Materials and methods
Reagents and solutions
Bovine serum albumin (BSA), dimethylsulfoxide (DMSO) and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), thrombin, and Gly-Pro-Arg-Pro peptide (GPRP) were purchased from Sigma (St Louis, MO, USA). ABT-737 was purchased from Selleck (Houston, TX, USA). A23187 was purchased from Calbiochem (San Diego, CA, USA). Phycoerythrin (PE)-conjugated anti-P-selectin antibody (anti-CD62, clone AC1·2) was purchased from BD Biosciences (San Jose, CA, USA), and green-fluorescent cationic dye 3,3′-dihexyloxacarbocyanine iodide DiOC6(3) was purchased from Invitrogen (Carlsbad, CA, USA).
Buffer A was composed of phosphate-buffered saline (PBS, Invitrogen) supplemented with 1 mmol/l MgCl2, 5·6 mmol/l glucose, 0·1% BSA and 10 mmol/l HEPES, pH 7·4 and used for preparation of thrombin and GPRP solutions, and as the buffer and control diluent for the thrombin-treated platelets. Buffer B was composed of buffer A supplemented with 0·11% dimethyl sulfoxide (DMSO) and used as the buffer and control diluent for the ABT-737- and A23187-treated platelets. Stock solutions of thrombin and GPRP were dissolved in buffer A and stock solutions of ABT-737 and A23187 were dissolved in 100% DMSO, and stored at −80°C.
Preparation and treatment of platelets
Venous blood from healthy volunteers was anticoagulated with 0·32% sodium citrate. Platelet-rich plasma (PRP) was obtained by centrifugation at 180 g for 15 min at room temperature (RT) and diluted 1:10 with DMSO-free buffer A or buffer B containing 0·11% DMSO.
Platelet apoptotic and/or activation events were determined in five platelet groups treated with: (i) diluent buffer A, (ii) diluent buffer B, (iii) 1 U/ml thrombin (Leytin et al, 2000a, 2006a), (iv) 30 μmol/l ABT-737 (Zhang et al, 2007) and (v) 10 μmol/l A23187 (Leytin et al, 2009). For this, 40 μl of diluted PRP aliquots were incubated for 15, 30, 60, 90 and 180 min at 37°C or RT with 10 μl of either diluent A without DMSO, diluent B with 0·11% DMSO, 5 U/ml thrombin in diluent A containing 12·5 mmol/l GPRP, 150 μmol/l ABT-737 in diluent B and 50 μmol/l A23187 in diluent B, respectively.
Concurrent determination of ΔΨm depolarization and CD62 exposure
For concurrent quantification of platelet apoptosis and activation, dual-staining flow cytometric assay was performed. Following 15–180 min platelet treatments with diluents A and B, thrombin, ABT-737 and A23187, 50 μl samples were incubated for 20 min at RT in the dark with 10 μl of 600 nmol/l DiOC6(3) for determination of ΔΨm depolarization and 2·5 μl of anti-CD62 antibody for determination of CD62 exposure, resulting in a final DiOC6(3) concentration of 96 nmol/l. Samples were then diluted to 500 μl with buffer A and analysed by a FACSCalibur flow cytometer using CellQuest and FCS Express software (BD Biosciences, San Jose, CA) (Leytin et al, 2006a, 2007).
Twenty thousand events were acquired and platelets were detected in the platelet-specific gate (Leytin et al, 2009) followed by dual-fluorescence dot-plot analysis. Platelet apoptosis (ΔΨm depolarization) was determined as downregulation of DiOC6 (3)-stained platelets and platelet activation (CD62 exposure) as upregulation of anti-CD62 binding to the platelet surface.
Data were analysed using GraphPad Prism 5 software (GraphPad Software, San Diego, CA, USA) and presented as means ± standard errors of means (SEM) for 4–9 independent experiments. The statistical significance of the differences inside and between platelet groups was determined by Student's t-test and one-way anova with Dunnett's multiple comparison test as appropriate. Differences were considered significant when P <0·05.
Flow cytometric analysis of platelet apoptosis and/or activation
Platelet apoptosis and activation have been determined by flow cytometry as ΔΨm depolarization (Kroemer & Reed, 2000; Leytin et al, 2006a, 2009; Gyulkhandanyan et al, 2012) and CD62 exposure (Stenberg et al, 1985; Michelson et al, 1991; Leytin et al, 2000a,b, 2007) respectively, well-known markers of these platelet responses. As shown in Fig 1, platelet apoptosis is characterized by downregulation of ΔΨm marker, and platelet activation by upregulation of CD62 marker. Flow cytometric dual-colour dot-plot analysis allows quantification of the percentage of cell distribution between four platelet subpopulations: (i) non-apoptotic and non-activated ‘double-negative’ cells (Apo−Act−), characterized by a high value of ΔΨm marker and low value of CD62 marker (Fig 1A, E), (ii) apoptotic-only but non-activated ‘single-positive’ cells (Apo+Act−), characterized by low ΔΨm and low CD62 (Fig 1B, E), (iii) activated-only but non-apoptotic ‘single-positive’ cells (Apo−Act+), characterized by high ΔΨm and high CD62 (Fig 1C, E), and (iv) both apoptotic and activated ‘double-positive’ cells (Apo+Act+), characterized by low ΔΨm and high CD62 (Fig 1D, E). Flow cytometric dot plots illustrating distribution of platelets between these four subpopulations are shown for platelets treated with diluent buffer B containing 0·11% DMSO (Fig 1A), BH3 mimetic ABT-737 (Fig 1B), the natural platelet agonist thrombin (Fig 1C), and calcium ionophore A23187 (Fig 1D).
Platelets treated with diluent buffers for 15–180 min at 37°C or RT are predominantly non-apoptotic and non-activated
We analysed the effects of platelet treatments with DMSO-free diluent A, the control buffer for thrombin treatment, and diluent B containing 0·11% DMSO, the control buffer for ABT-737 and A23187 treatments, on ΔΨm depolarization and CD62 exposure in human platelets (Figure S1 and Table S1, Online Supplement). Given that it has been shown that exposure of platelets to 37°C more strongly induces platelet apoptosis and reduces platelet lifetime than incubation at RT (Holme & Heaton, 1995; Bertino et al, 2003), we analysed the temperature- and time-dependent effects of control diluent buffers and platelet agonists on ΔΨm depolarization compared to CD62 exposure.
Treatment of platelets with diluent buffer A without DMSO for 15–90 min at 37°C did not induce platelet apoptosis or activation (Figure S1A). Following 15 min treatment with this diluent, 95·7 ± 0·7% platelets did not undergo ΔΨm depolarization or CD62 exposure. Although during longer treatment a trend for decrease of the size of the double-negative Apo−Act− platelet subpopulation was observed (93·5 ± 0·9% at 30 min, 91·8 ± 1·5% at 60 min and 90·2 ± 2·0% at 90 min), this decrease did not reach the level of significance (P >0·05), and only at 180 min did the percentage of cells in Apo−Act− subpopulation decrease significantly to 83·1 ± 4·0% (Figure S1A and Table S1, P <0·001). The presence of 0·11% DMSO in diluent buffer B also did not significantly enhance the effect of diluent on platelet apoptosis and activation at 37°C, i.e. the percentage of cells in Apo−Act− subpopulation progressively but insignificantly decreased, from 94·8 ± 0·7% at 15 min to 88·2 ± 2·3% at 90 min (P >0·05), but after 180 min treatment this decrease became significant (Figure S1A and Table S1: 80·6 ± 4·6%, P <0·001). As shown in Figure S1B and Table S1, similar time-dependence profiles were observed when platelets were treated with diluent buffers A and B at RT.
Thus, during treatment of platelets with diluent buffers for 15–90 min, the platelet population was mostly (88–97%) non-apoptotic and non-activated. During the next 90 min of treatment, the percentage of Apo−Act− cells was slightly but significantly decreased to 81–84%. However, these non-apoptotic and non-activated cells still remained predominant in the total platelet population. For the whole treatment period of 15–180 min with diluent buffers, the percentage of cells in the double-negative Apo−Act− platelet subpopulation did not depend on whether the platelet treatment was performed at 37°C or RT or in the presence or absence of 0·11% DMSO (Figure S1 and Table S1).
The apoptosis-only single-positive Apo+Act− subpopulation was the most prominent of three non-predominant moderate or minor subpopulations (Apo+Act−, Apo−Act+ and Apo+Act+) and showed an insignificant (P >0·05) trend to increase during 15–90 min treatments at 37°C and RT but was significantly (P <0·01/0·001) increased to 12·4 ± 5·1–16·6 ± 4·7% at 180 min treatment (Table S1).
ABT-737 predominantly induces platelet apoptosis during treatment for 90–180 min at 37°C
Figure 2A demonstrates the time-dependent effect of BH3 mimetic ABT-737 on apoptosis-only (Apo+Act−) platelet response, manifested as the ΔΨm depolarization in the absence of CD62 exposure, when platelets were treated with ABT-737 for 15–180 min at 37°C and RT. Following 30–180 min treatment at 37°C, the percentage of cells in the Apo+Act− subpopulation was progressively increased (Fig 2A) and was significantly higher than at 15 min treatment (Table 1: P <0·01 at 30 min and P <0·001 at 60–180 min). Furthermore, for all treatment times at 37°C, the percentage of cells in the Apo+Act− subpopulation was significantly higher than in the activation-only (Apo−Act+) subpopulation (Table 1: P <0·05 for 15 min, P <0·0001 for 30–180 min). At 90 min of treatment, the Apo+Act− subpopulation became predominant and the percentage of cells in this subpopulation reached the maximum of 76·7 ± 2·9%, which was not significantly different from the value of 81·4 ± 2·7% at 180 min (Fig 2A, P =0·24).
Table 1. Distribution of cells between four platelet subpopulations during treatment of platelets with BH3 mimetic ABT-737, thrombin and calcium ionophore A23187: Time- and temperature-dependence
Cells in four platelet subpopulations (% of total)
Treatment at 37°C
Treatment at RT
n = 8–9
n = 6–7
n = 8
n = 4–9
n = 4–7
n = 4–8
Platelets were treated with 30 μmol/l ABT-737, 1 U/ml thrombin or 10 μmol/l A23187 for 15–180 min at 37°C or RT and analyzed by flow cytometry. The percentage of cells in Apo−Act−, Apo+Act−, Apo−Act+ and Apo+Act+ subpopulations were analyzed as indicated in Figure 1. Data are presented as means ± SEM; data for predominant at optimal treatment conditions platelet subpopulations are presented in bold. The differences between 30–180 min versus 15 min treatments were calculated by one-way anova with Dunnett's multiple comparison test: ns (P >0·05), not significant differences, *P <0·05, **P <0·01, ***P <0·001.
Treatment of platelets with ABT-737 at RT was characterized by a significantly lower and slower stimulation of ΔΨm depolarization than at 37°C. For all treatment times of 15–180 min, the percentage of cells in the Apo+Act− subpopulation at RT was less than time-matched values at 37°C (Fig 2A). Thus, at 90 min of RT treatment, 19·6 ± 6·7% platelets underwent apoptosis versus 76·7 ± 2·9% at 37°C (Table 1 and Fig 2A, P <0·0001), and at 180 min of treatment these values were 59·6 ± 13·4% and 81·4 ± 2·7% at RT and 37°C, respectively (Table 1 and Fig 2A, P <0·05).
Over the entire treatment period at 37°C, the percentage of cells in the Apo+Act− subpopulation in ABT-737-treated platelets was significantly higher than in platelets treated with time-matched diluent B (Table S2A, Online Supplement: P <0·01 at 15 min and P <0·0001 at 30–180 min). In contrast, at RT, during the first 90 min the differences between these treatments were not significant (P >0·05 at 15–90 min) and the difference became significant (P =0·016) only at 180 min treatment (Table S2A).
Thrombin predominantly induces platelet activation during treatment for 15–180 min at 37°C or RT
In contrast to ABT-737, treatment of platelets with the physiological agonist thrombin predominantly induces activation-only (Apo−Act+) platelet response, manifested as the CD62 exposure in the absence of ΔΨm depolarization. Even during the first 15 min of treatment at 37°C or RT, the percentage of cells in this subpopulation reached the maximal level of 84–85% of the total platelet population (Fig 2B and Table 1). For both temperatures, this high level of platelet activation persisted up to 90 min; the differences between 30–90 min versus 15 min treatments were not significant (Table 1, P >0·05). However, during prolonged treatment for 180 min, the size of Apo−Act+ subpopulation was slightly but significantly decreased, to 69·8 ± 2·3% at 37°C (Table 1, P <0·001) and 80·2 ± 0·9% at RT (Table 1, P <0·05). This decrease was associated with the enhanced effect of thrombin on platelet apoptosis during prolonged 180 min treatment resulting in the increase of the percentage of cells in the Apo+Act− and Apo+Act+ subpopulations at 37°C (Table 1: 11·2 ± 2·4%, P <0·05 and 9·8 ± 1·9%, P <0·001, respectively) and in the Apo+Act+ subpopulation at RT (Table 1: 8·4 ± 1·1%, P <0·001).
The temperature-dependence of the effect of thrombin on platelet activation and apoptosis was compared at 37°C and RT. For 15–60 min treatments, no differences were found in the effect of thrombin on the percentage of cells in the Apo−Act+ subpopulation at 37°C versus RT (Fig 2B, P >0·05). At 90 min and 180 min treatments, however, the decrease of the percentage of cells in the Apo−Act+ subpopulation was prominent at 37°C in comparison with RT (Fig 2B: P < 0·05 at 90 min and P <0·01 at 180 min).
Throughout the entire treatment period of 15–180 min, the percentage of cells in the Apo−Act+ subpopulation was much higher in thrombin-treated platelets than in diluent A-treated platelets both at 37°C and RT (Table S2B: P <0·0001).
A23187 predominantly induces both platelet apoptosis and activation during treatment for 15–60 min at 37°C or 15–90 min at RT
With calcium ionophore A23187 treatment, the platelets became both apoptotic and activated (Apo+Act+), undergoing concomitantly ΔΨm depolarization and CD62 exposure (Fig 2C). Maximal levels of apoptotic and activation responses were reached during the first 15–30 min of treatment at both temperatures, when the majority (75–81%) of the cells were in the Apo+Act+ subpopulation (Table 1). Then, at 37°C, the percentage of double-positive Apo+Act+ cells progressively decreased to 72·7 ± 2·0% at 60 min (P >0·05), 63·6 ± 2·5% at 90 min (P <0·01) and 33·1 ± 0·9% at 180 min (P <0·001) with a concomitant increase of the percentage of single-positive apoptotic but non-activated (Apo+Act−) cells from the mean values of 11·7–11·8% at 15–30 min to 21·1 ± 2·6% (P <0·01) and 31·6 ± 3·0% (P <0·001) at 60 and 90 min, respectively, and 62·8 ± 1·5% (P <0·001) at 180 min (Table 1).
This significant decrease of the size of the double-positive Apo+Act+ subpopulation with a simultaneous increase of single-positive Apo+Act− subpopulation during prolonged 90–180 min treatment of platelets with A23187 at 37°C (Table 1) probably reflects shedding of P-selectin (CD62) from the platelet surface to the extracellular medium with the formation of soluble CD62 (Michelson et al, 1996; Berger et al, 1998). It appears that this CD62-shedding was slower at RT than at 37°C. Thus, at 60 and 90 min treatments, the percentage of cells in the Apo+Act+ subpopulation was significantly higher at RT than at 37°C (Fig 2C: P <0·05 and P <0·01, respectively), but after 180 min treatment the percentage of cells in this subpopulation decreases equally at both temperatures (Fig 2C, P >0·05). Between 90 and 180 min treatments at 37°C and RT, approximately 50% of Apo+Act+ cells lost exposed CD62: during 90 min treatment at 37°C on average 63·6% cells were present in the Apo+Act+ subpopulation versus 33·1% at 180 min and at RT these values were 76·7% versus 36·9%, respectively (Table 1).
Comparison of A23187 effects on the two single-positive platelet populations, apoptosis-only (Apo+Act−) and activation-only (Apo−Act+), shows that starting from treatment for 30 min at 37°C and 60 min at RT, the calcium ionophore induced significantly stronger formation of platelets with depolarized ΔΨm rather than platelets with CD62 exposed on their surface (Fig 3, Table 1).
For all treatment times at 37°C and RT, the percentage of cells in the Apo+Act+ subpopulation of A23187-treated platelets was much higher than in the Apo+Act+ subpopulation of diluent B-treated platelets (Table S2C: P <0·0001).
Despite the lack of a nucleus, platelets are capable of performing multiple vital functions of nucleated cells and are involved in diverse physiological and pathological processes beyond haemostasis and thrombosis (Smyth et al, 2009; Semple et al, 2011). Furthermore, platelets have the essential machinery for executing programmed cell death, i.e., apoptosis (Vanags et al, 1997; Li et al, 2000; Leytin & Freedman, 2003; Mason et al, 2007; Kile, 2009; Leytin, 2012).
The role of platelet apoptosis, based on the determination of different apoptosis markers, has been studied in human diseases and animal models of diseases (reviewed by Gyulkhandanyan et al, 2012) including immune thrombocytopenia (Piguet & Vesin, 2002; Catani et al, 2006; Leytin et al, 2006b; Winkler et al, 2012), Bernard–Soulier syndrome (Rand et al, 2010), chronic uraemia (Bonomini et al, 2004), bacterial infection (Yeh et al, 2010), malaria (Piguet et al., 2002) and type 2 diabetes (Cohen et al, 2002). Interestingly, murine (Piguet & Vesin, 2002; Leytin et al, 2006b) and human immune thrombocytopenia (Winkler et al, 2012) and thrombocytopenia in a murine model of malaria (Piguet et al., 2002) were ameliorated by treatment with anti-apoptotic agents concurrent with inhibition of apoptosis in platelets. Platelet apoptosis was also studied in stored platelet concentrates (Li et al, 2000; Bertino et al, 2003; Leytin & Freedman, 2003; Perrotta et al, 2003; Leytin et al, 2008).
Animal models have been further used for tracking apoptotic platelets during ageing of platelets in the normal rabbit circulation (Rand et al, 2004) and in a canine model of suppressed thrombopoiesis (Pereira et al, 2002). Apoptotic status of circulating platelets in these studies was detected by ΔΨm depolarization and phosphatidylserine exposure on the cell surface (Pereira et al, 2002; Rand et al, 2004). Another approach is genetic mutations of Bcl-2 family proteins in mice: deletion of the anti-apoptotic Bcl-XL protein reduces the half-life of murine platelets and causes thrombocytopenia, whereas in pro-apoptotic Bak deficient mice platelet life-span is longer than normal (Mason et al, 2007). Furthermore, injection of the pro-apoptotic BH3 mimetic ABT-737 induces platelet clearance from the circulation in mice (Mason et al, 2007) and dogs (Zhang et al, 2007). Infusion of rabbit platelets treated in vitro with calcium ionophore A23187 into recipient rabbits resulted in platelet clearance (Rand et al, 2004); in contrast, in vitro treatment with thrombin did not induce clearance of rabbit platelets (Rand et al, 2004).
Although the role of platelet activation in haemostasis and thrombosis is well-documented, the role of platelet apoptosis in these vital processes and of platelet apoptosis versus activation in platelet clearance are still to be elucidated. To provide a methodological basis for investigation of these unresolved questions, the current study established a model for selective targeted stimulation in vitro of platelet apoptosis without affecting activation, platelet activation without affecting apoptosis, as well as both responses. Furthermore, we determined the conditions when neither apoptosis nor activation was affected.
In contrast to previous reports (Leytin et al, 2004, 2006a, 2007, 2009; Mason et al, 2007; Zhang et al, 2007; Dasgupta et al, 2010; Vogler et al, 2011; Mutlu et al, 2012), this study establishes a model that permits dissociation of platelet apoptosis and activation. For this, we elaborated a simple and effective method of preparation of three platelet populations that are highly enriched for apoptotic-only, activated-only, or both apoptotic and activated platelets. This novel methodology involves: (i) concurrent use of three platelet agonists in one study: BH3 mimetic ABT-737 for selective induction of platelet apoptosis, thrombin for selective induction of platelet activation and calcium ionophore A23187 for selective induction of apoptosis plus activation, as well as matched control diluent buffers for determining conditions when neither apoptosis nor activation are induced, (ii) concurrent flow cytometric analysis of the crucial manifestation of intrinsic pathway of platelet apoptosis, ΔΨm depolarization, together with the platelet activation event, P-selectin (CD62) exposure on the cell surface, (iii) detailed statistical analysis of platelet distribution between four platelet subpopulations (Apo+Act−, Apo−Act+, Apo+Act+ and Apo−Act−), and (iv) direct comparison of the sensitivity of platelet apoptotic and activation responses to time- and temperature-titration of platelets with ABT-737, thrombin and calcium ionophore.
This study design permitted us to define the optimal conditions for selective stimulation of the fundamental platelet processes of apoptosis and activation, separately or together. Maximal stimulation of apoptosis-only platelet response at the level of approximately 80% was achieved by ABT-737 treatment for 90 min at 37°C (Fig 2A, Table 2); shorter treatment at 37°C or treatment at RT for 15–180 min did not cause maximal stimulation (Table 1). However, if required by a particular study design, longer treatment with ABT-737 for 180 min at 37°C could be used, also yielding an Apo+Act− subpopulation at the level of about 80% (Fig 2A, Table 1). In contrast, predominant stimulation of an activation-only platelet response (by thrombin) and stimulation of both apoptosis and activation in the same cells (by A23187) was achieved by a shorter treatment for 15–30 min at 37°C or RT (Fig 2B, C); if required, a prolonged treatment up to 90 min at 37°C or RT could be used for thrombin, or up to 60 min at 37°C and up to 90 min at RT for A23187 (Tables 1 and 2). For obtaining the maximal level (88–97%) of non-apoptotic non-activated cells, platelets can be treated for 15–90 min at 37°C or RT with DMSO-free diluent buffer A or diluent buffer B containing DMSO (Table 2).
Table 2. Optimal conditions for the maximal stimulation of predominant apoptotic and/or activation platelet responses triggered by treatment of platelets with BH3 mimetic ABT-737, thrombin and calcium ionophore A23187 or the absence of apoptotic and activation responses in platelets treated with diluents A and B
Mean ± SEM and ranges of means are based on the data presented in Table 1 for Apo+Act−, Apo−Act+ and Apo+Act+ subpopulations, and in Table S1 for Apo−Act− subpopulation.
90 min, 37°C
76·7 ± 2·9
15–90 min, 37°C
15–90 min, RT
15–60 min, 37°C
15–90 min, RT
15–90 min, 37°C/RT
Time- and temperature-dependency experiments indicate that the platelet activation response to thrombin is strong (84–85% CD62-positive cells) and fast (15 min at 37°C or RT) (Table 1). In contrast, the apoptotic response to thrombin required a much longer time (180 min at 37°C or RT), when not more than 20% cells undergo ΔΨm depolarization (Table 1). A dose-titration study of CD62- and ΔΨm-responses also demonstrated much higher sensitivity of platelet activation to thrombin compared to platelet apoptosis (Leytin et al, 2007).
Characterizing effects of an RT environment on platelet apoptosis and activation is important for the clinically relevant setting of platelet transfusion therapy (Leytin et al, 2008), because conventional blood banking storage of platelets for transfusion is at 22°C (Perrotta & Snyder, 2007). Furthermore, platelets traverse central and peripheral circulations, undergoing repeated exposure to lower temperatures at the body surface (Hoffmeister et al, 2003) and this periodic exposure of platelets to environmental temperatures may impact platelet apoptosis, activation and clearance.
Presented data demonstrate that BH3 mimetic ABT-737 much more efficiently triggers the mitochondrial apoptotic event ΔΨm depolarization in platelets at 37°C than at RT. In contrast, the platelet agonist thrombin almost equally induces CD62 exposure from α-granules to the platelet surface at both temperatures, indicating different temperature dependencies of ΔΨm and CD62 markers of platelet apoptosis and activation respectively. The stimulatory effects of elevated (37°C and 43°C) temperatures on intrinsic apoptosis pathway have been also shown in mitochondria isolated from nucleated cells (Pagliari et al, 2005; Lazarou et al, 2010). It was reported that prolonged (30–60 min) but not early (5–15 min) incubation of Bak with isolated mitochondria of HeLa cells at 37°C results in Bak activation, mitochondrial outer membrane permeabilization (MOMP) and cytochrome c release (Lazarou et al, 2010). Activation of Bak and Bax by elevated temperatures is downregulated with anti-apoptotic Bcl-2 and Bcl-XL proteins and upregulated with pro-apoptotic BH3-only proteins (Pagliari et al, 2005; Lazarou et al, 2010).
Our study also shows that, depending on the triggering stimulus, a high level of platelet apoptosis is not necessarily associated with platelet activation, a high level of platelet activation is not necessarily associated with platelet apoptosis, and that the apoptotic and activation responses may be concomitant. The different effects of ABT-737, A23187 and thrombin as inducers of apoptosis and/or activation may reflect the different actions of these agonists on mitochondrial membrane permeabilization and ΔΨm depolarization.
The biomedical significance of the current study includes several aspects. Firstly, the seminal methodological aspect consists of elaboration of a simple and effective method of preparation of three platelet populations which are highly enriched for apoptotic-only, activated-only, or both apoptotic and activated platelets. These specific populations, which otherwise are not easily isolated by flow cytometric sorting, can be readily obtained by treatment of platelets with appropriate inducers (ABT-737, thrombin and A23187). Secondly, this study, taken together with previously reported data (Leytin et al, 2007, 2008; Schoenwaelder et al, 2009), demonstrates that platelet apoptosis and activation are different phenomena, and opens possibilities for targeted differential pharmacological manipulation of platelet apoptosis and activation. Third, selective stimulation of apoptosis or activation enables the study of the mechanisms of these fundamental cellular responses in anucleate platelets. Fourth, the current study provides a basis for future investigations on the roles of platelet apoptosis and activation in platelet clearance, haemostasis, thrombosis and other platelet-associated diseases.
AVG and VL designed the research study. AVG performed the research and analysed the data. AVG, AM, JF and VL analysed the data and wrote the paper.
Conflicts of interest
The authors have no competing interests.
This work was supported by a grant from the Heart and Stroke Foundation of Ontario, Canada.