Comprehensive Analysis of Procoagulant Platelets Exhibiting Features of Necrosis, Apoptosis and Platelet Activation

The present protocol provides a comprehensive set of procedures necessary to analyze procoagulant platelets, which exhibit overlapping features of necrosis, apoptosis, and platelet activation.

Platelets circulating in the bloodstream are relatively quiescent but become "activated" upon encountering stimulants or "agonists" at the site of blood vessel injury. Proaggregatory and procoagulant platelets represent two distinct populations of activated platelets. While proaggregatory platelets facilitate the cessation of bleeding, or "hemostasis," by forming a plug of platelets clumped together through fibrinogen bridges, procoagulant platelets dramatically accelerate the coagulation cascade, culminating in fibrin clot formation. An interesting aspect of procoagulant platelets is that their morphology exhibits certain features of "necrosis" and "apoptosis." They may thus represent a form of cell death in platelets, albeit with an important role in thrombosis and hemostasis. This article introduces the concept of procoagulant platelets, their relevance to health and disease, and a comparison of existing methods for their analysis. It then provides comprehensive protocols for analyzing procoagulant platelets, investigating the mechanisms of their formation, and assessing their prothrombotic role in facilitating coagulation. The article concludes with a discussion of key steps, limitations, and troubleshooting principles for the described methods.

There are at least two distinct populations of activated platelets^1,2. Pro-aggregatory platelets are characterized by high integrin activation and low, if any, PS exposure. On the other hand, procoagulant platelets are characterized by low integrin activity and high PS exposure^1,2, providing a surface for the assembly of tenase and prothrombinase complexes³, which is 10⁵-10⁶-fold and 300000-fold more active than individual soluble phase factor IXa and Xa, respectively⁴. Procoagulant platelets thus dramatically accelerate coagulation. An interesting aspect of procoagulant platelets is that their morphology resembles certain features of "necrosis" such as microvesiculation, ballooning of the cell with cytoskeletal disruption, and loss of membrane integrity, as well as those of "apoptosis" such as loss of membrane phospholipid asymmetry with exposure of phosphatidylserine in the outer leaflet^5,6. In other words, procoagulant platelet formation may represent a form of cell death in platelets, albeit with an important physiological function in hemostasis.

Procoagulant platelets are significantly associated with thrombotic disorders. While most healthy individuals do not have circulating procoagulant platelets, ~30% of normal donor platelets adopt a procoagulant phenotype ex vivo following exposure to strong agonists such as thrombin and collagen⁷. Circulating procoagulant platelets have been reported in trauma, where their formation may reflect activation by histone H4^8,9. In most prothrombotic disorders, however, increased levels of procoagulant platelets are only detected following ex vivo stimulation¹⁰. For example, patients with acute stroke in whom >51.1% of their platelets were converted to procoagulant platelets (also known as COATed platelets) by collagen and thrombin had a hazard ratio of 10.72 for recurrent stroke within 30 days compared to patients with lesser procoagulant platelet formation¹¹. Similar results have been reported in patients with transient ischemic attacks and carotid atherosclerosis¹². In contrast, the bleeding disorder Scott syndrome results from a mutation of ANO6, which encodes the phospholipid scramblase TMEM-16F, leading to deficient platelet PS exposure¹³. Idiopathic bleeding disorders and intracranial bleeding may be associated with a decreased ability to generate procoagulant platelets¹⁴.

Hence, assessing procoagulant platelets is part of any analysis of platelet function not only during basic investigations into mechanisms of platelet activation and consequent thrombosis and hemostasis, but also during clinical analysis for risk of thrombosis or bleeding in patients during various pathological states. An International Society on Thrombosis and Haemostasis (ISTH) panel recommended the use of Annexin V binding and P-selectin expression by flow cytometry for distinguishing procoagulants from other platelet subpopulations¹⁵. The article also discusses the various methods that can be used to analyze procoagulant and apoptotic platelets but falls short of describing the processes in detail. These methods include detection of (1) platelet activation by PAC1/JonA or fibrinogen binding (flow cytometry); (2) alpha-granule secretion by P-selectin expression (flow cytometry); (3) PS exposure by Annexin V/lactadherin binding (flow cytometry); (4) loss of membrane integrity by GSAO labeling (flow cytometry); (5) morphological changes like ballooning (microscopy); (6) detection of caspase activation by caspase assay (immunoblotting/luminometry/flow cytometry) or degradation of cytoskeletal substrate gelsolin (immunoblotting); (7) loss of mitochondrial membrane potential by mitochondrial potential-sensitive dyes such as JC-1/Mitotracker (flow cytometry); (8) mitochondrial intrinsic apoptotic markers Bax, Bak and cytochrome c release (immunoblotting); (9) procoagulant function by thrombin generation assay and coagulation factor Xa/Va binding (flow cytometry, microscopy); (10) cytosolic and mitochondrial calcium rise by fluorescent calcium sensitive dyes (flow cytometry, fluorometry, microscopy).

The present study delves into comprehensive protocols for the analysis of procoagulant platelets as well as distinguishing them from proaggregatory and apoptotic platelets. Most procedures described rely on flow cytometry that has the advantages of (1) being readily available and easy to use, (2) requiring low sample volume, and (3) allowing simultaneous detection of multiple subpopulations of platelets (proaggregatory, procoagulant, and apoptotic)¹⁵. These flow cytometry-based protocols are supplemented with functional assays of procoagulant activity based on coagulation factor binding and clot-based thrombin generation assays.

Human participants were recruited in the study for peripheral venous blood sampling after obtaining written informed consent, strictly following the recommendations and approval of the Institutional Review Board of Cleveland Clinic Lerner Research Institute, with all study methodologies conforming to the standards set by the Declaration of Helsinki. Healthy adult participants above 18 years of age were included, while those younger than 18 years, individuals with a recent history of thrombotic events in the past six months, those with a history of alcoholism or drug abuse, and participants who had used anti-platelet or anti-coagulant medications in the past four weeks were excluded. A detailed description of materials and reagents used in the protocols can be found in the Table of Materials.

1. Platelet preparation

1. Draw peripheral venous blood samples into ACD anticoagulant (1:9 v/v) from human participants after obtaining written informed consent, strictly following the recommendations approved by the Institutional Review Board.
2. Centrifuge the blood collected in ACD at 100 x g for 20 min at 22 °C to obtain platelet-rich plasma (PRP) as the supernatant.
3. Centrifuge the PRP at 800 x g for 7 min at 22 °C after adding 3 µM of PGE₁ and 2 mM of EDTA.
4. Discard the supernatant and resuspend the platelet pellet in buffer A (see Table of Materials) by gentle pipetting.
5. Centrifuge the platelets resuspended in buffer A at 800 x g for 7 min at 22 °C.
6. Resuspend the platelet pellet in buffer B (see Table of Materials) by gentle pipetting.
7. Adjust the final platelet count to 1 × 10⁷/mL using an automated cell counter.
   NOTE: Perform all steps under sterile conditions and take precautions to maintain the platelets in a resting state by avoiding exposure to excessive shear during pipetting.

2. Analysis of procoagulant platelets by flow cytometry

1. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL of platelet suspension).
2. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, 100 ng/mL), or their combination for 15 min at room temperature.
3. After stimulation, add 1 µL each of PE-Annexin V, FITC-PAC1, and APC-anti-human CD62P antibody to 100 µL of stimulated platelet suspensions.
4. Incubate the platelets in the dark at room temperature for 30 min.
5. Fix the platelets by adding an equal volume (1:1) of 2% formalin.
6. Analyze the samples by flow cytometry to quantify procoagulant platelets^15,16.
7. Draw an amorphous gate to encompass platelets, separating them from noise and multi-platelet particles.
8. Collect all fluorescence data using four-quadrant logarithmic amplification for 10,000 events in the platelet gate from each sample.
9. Define regions for platelets based on fluorescence positivity or negativity for phosphatidylserine (PS) exposure (Annexin V binding), P-selectin (CD62P) expression, and integrin αIIbβ₃ activation (PAC1 binding).
10. Consider platelets positive for both PS exposure (Annexin V binding) and P-selectin (CD62P) expression but negative for integrin αIIbβ₃ activation (PAC1 binding) as procoagulant platelets (Figure 1) (Table 1).
11. Similarly, consider platelets positive for both integrin αIIbβ₃ activation (PAC1 binding) and P-selectin (CD62P) expression but negative for PS exposure (Annexin V binding) as proaggregatory platelets.
    NOTE: Platelets negative for both integrin αIIbβ₃ activation (PAC1 binding) and P-selectin (CD62P) expression but positive for PS exposure (Annexin V binding) are likely apoptotic platelets^15,16.

3. Analysis of mitochondrial calcium by flow cytometry

1. Dilute washed human platelets to 1 × 10⁶/mL and supplement with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL platelet suspension).
2. Label the platelets with 5 µM Rhod-2 AM (for mitochondrial calcium) for 30 min in the dark.
3. Gate platelets appropriately, as described in step 2.
4. Analyze events in the platelet gate for time-dependent changes in the mean fluorescence of acquired events over 5 min using a density scatter plot of PE fluorescence (for Rhod-2 AM) versus time.
5. Record baseline calcium levels for the first 1 min.
6. Add thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination, and continue acquiring data for another 4 min¹⁷.

4. Analysis of mitochondrial membrane potential by flow cytometry

1. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL of platelet suspension).
2. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination for 15 min at room temperature.
3. After stimulation, add mitochondria labeling dye (see Table of Materials) at a final concentration of 500 nM to the stimulated platelet suspensions.
4. Incubate the platelets for 30 min in the dark.
5. Fix the platelets by adding an equal volume of 2% formalin.
6. Analyze samples by flow cytometry¹⁷.
7. Gate platelets appropriately as described in step 2.
8. Analyze events in the platelet gate for a drop in fluorescence in the PE channel (for the mitochondria labelling dye) of the flow cytometer.

5. Analysis of caspase 3 and caspase 8 activity by flow cytometry

1. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL of platelet suspension).
2. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination for 15 min at room temperature.
3. After stimulation, add either 1:1000 (v/v) apoptosis detection reagent (see Table of Materials) (for both intrinsic and extrinsic apoptosis pathways) or 1:300 (v/v) FITC-IETD-FMK (for caspase 8 of the extrinsic apoptosis pathway) to the stimulated platelet suspensions.
4. Incubate the platelets for 30 min in the dark.
5. Fix the platelets by adding an equal volume of 2% formalin.
6. Analyze samples by flow cytometry¹⁷.
7. Gate platelets appropriately, as described in step 2.
8. Analyze events in the platelet gate for fluorescence in the FITC channel (for either caspase 3/7 or caspase 8) of the flow cytometer.

6. Analysis of prothrombin binding by flow cytometry

1. Conjugate bovine prothrombin with Alexa Fluor 488 dye using a protein labeling kit (see Table of Materials), following the manufacturer's instructions.
2. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL platelet suspension).
3. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination for 15 min at room temperature.
4. After stimulation, add AF488-conjugated bovine prothrombin (100 µg/mL) and 1 µL each of PE-Annexin V and APC anti-human CD62P antibody to 100 µL of stimulated platelet suspensions.
5. Fix the platelets by adding an equal volume of 2% formalin.
6. Analyze samples by flow cytometry¹⁷ for the quantification of prothrombin binding to platelets.
7. Draw an amorphous gate to encompass platelets separate from noise and multi-platelet particles.
8. Collect all fluorescence data using four-quadrant logarithmic amplification for 10,000 events in the platelet gate from each sample.
9. Draw regions for platelets with fluorescence positive or negative for PS exposure (Annexin V binding), P-selectin (CD62P) expression, and AF488-prothrombin (prothrombin binding).
10. Determine the proportion of all platelet events, as well as the proportion of platelets positive for both PS exposure (Annexin V binding) and P-selectin (CD62P) expression (procoagulant platelets), that are positive for prothrombin binding.

7. Analysis of prothrombin binding by confocal microscopy

1. Conjugate bovine prothrombin with Alexa Fluor 488 dye using a protein labeling kit (see Table of Materials), following the manufacturer's instructions.
2. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL platelet suspension).
3. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination for 15 min at room temperature.
4. After stimulation, add AF488-conjugated bovine prothrombin (100 µg/mL) and 1 µL each of PE-Annexin V and APC anti-human CD62P antibody to 100 µL of stimulated platelet suspensions.
5. Fix the platelets by adding an equal volume of 2% formalin.
6. Pellet the fixed platelets onto poly-D-lysine-coated coverslips.
7. Mount coverslips on microscopy slides using an anti-fade solution.
8. Alternatively, coat coverslips with collagen (100 µg/mL) for 1 h in a humid chamber.
9. Block the coated coverslips with 0.5% BSA in PBS for 1 h.
10. Label the platelets with AF488-conjugated bovine prothrombin (100 µg/mL) and 1 µL each of PE-Annexin V and APC anti-human CD62P antibody.
11. Allow labeled platelets to adhere for 20 min on collagen-coated coverslips.
12. Wash coverslips with adherent platelets three times with PBS.
13. Fix coverslips with 2% paraformaldehyde for 1 h.
14. Mount coverslips on microscopy slides using an anti-fade solution.
15. Observe the slides under a confocal microscope at 63× objective magnification.
16. Analyze the images using Fiji software for fluorescence quantification.

8. Clot-based platelet phospholipid-dependent thrombin generation assay

1. Supplement washed human platelets with 2.5 mM calcium (0.5 µL of stock CaCl₂ 0.5 M solution in 100 µL of platelet suspension).
2. Keep one fraction of platelets unstimulated and stimulate the remaining fractions with thrombin (0.1 U/mL, 0.25 U/mL, or 0.5 U/mL), convulxin (20 ng/mL, 50 ng/mL, or 100 ng/mL), or their combination for 15 min at room temperature.
3. Add 100 µL of treated platelets to a mixture of 50 µL of pooled normal plasma and 50 µL of low-turbidity kaolin solution (20 mg/mL), preincubated at 37 °C for 5 min.
4. Add 5 mM of CaCl₂ to the mixture.
5. Monitor the clot formation by turbidimetry on a microplate reader, measuring absorbance at 660 nm every 60 s for 1 h.

A proportion of activated platelets turn "procoagulant" with a characteristic increase in surface expression of both phosphatidylserine (PS) and P-selectin, which distinguishes them from "apoptotic" platelets that are positive only for PS exposure as well as "proaggregatory" platelets that are positive for P-selectin expression. We found that thrombin induces a dose-dependent increase in the proportion of procoagulant platelets positive for both P-selectin and PS expression as detected by binding of FITC-anti-CD62P antibody and PE-Annexin V, respectively (Figure 2). The generation of procoagulant platelets is dependent upon mitochondrial calcium influx^18,19 along the electrochemical gradient across the inner mitochondrial membrane upon agonist-induced cytosolic calcium elevations as well as the consequent cyclophilin D (CypD)-dependent mitochondrial permeability transition pore (mPTP) formation²⁰. In consistence, thrombin-induced increase in procoagulant activity was associated with an increase in mitochondrial calcium (Figure 3) and a drop in mitochondrial membrane potential (Figure 4) as measured using Rhod-2 and Mitotracker Red dyes, respectively. Stimulated platelets exhibit features of apoptosis, including Bax/Bak accumulation^21,22; whether this activates caspase^23,24 and contributes to PS exposure²⁵ remains controversial. Thrombin stimulation of platelets was associated with activation of caspase 8, but not terminal caspases 3 and 7 (Figure 5), as analyzed by FITC-IETD-FMK and apoptosis detection reagent, respectively.

Figure 1: Analysis of procoagulant platelets. Schematic diagram depicting analysis of procoagulant platelets by flow cytometry using FITC-PAC1, APC-anti-CD62P antibodies, and PE-Annexin V as fluorescent probes to detect integrin activation, P-selectin expression, and PS exposure, respectively. Please click here to view a larger version of this figure.

Figure 2: P-selectin expression and phosphatidylserine exposure in platelets upon thrombin stimulation. Representative dot plots showing platelets positive for P-selectin expression and/or PS exposure in (A) unstimulated platelets and upon exposure to thrombin at doses of (B) 0.1 U/mL, (C) 0.25 U/mL and (D) 0.5 U/mL as detected by FITC-anti-CD62P and PE-Annexin V fluorescence using flow cytometry. Figures in parentheses indicate the proportion of total platelets in each quadrant of the plot. Please click here to view a larger version of this figure.

Figure 3: Time-dependent changes in mitochondrial calcium levels in platelets upon thrombin exposure. Representative density plots showing time-dependent changes in fluorescence due to Rhod-2 dye indicating mitochondrial calcium levels in platelets upon exposure to thrombin at doses of (A) 0.1 U/mL, (B) 0.25 U/mL and (C) 0.5 U/mL as detected by flow cytometry. Please click here to view a larger version of this figure.

Figure 4: Mitochondrial membrane potential in platelets upon thrombin stimulation. This figure shows representative histogram overlays depicting fluorescence due to mitochondria labeling dye, indicating mitochondrial membrane potential in platelets upon exposure to different doses of thrombin, as detected by flow cytometry. Please click here to view a larger version of this figure.

Figure 5: Caspase 3/7 and Caspase 8 activities in platelets upon thrombin stimulation. Representative histogram overlays showing fluorescence due to (A) apoptosis detection reagent and (B) FITC-IETD-FMK, respectively, indicating caspase 3/7 and caspase 8 activities in platelets upon exposure to different doses of thrombin as detected by flow cytometry. Please click here to view a larger version of this figure.

Table 1: Flow cytometry markers for identifying platelet populations after stimulation. This table shows flow cytometry markers used to detect distinct populations of platelets that emerge after stimulation of platelets with strong agonists such as thrombin and convulxin.

Procoagulant platelets demonstrate marked and sustained increases in intracellular calcium upon stimulation²⁶, but may be derived through different mechanisms. They are generated upon strong agonist stimulation with collagen and thrombin through distinct mediators, including most prominently mitochondrial calcium influx^18,19 along the electrochemical gradient across the inner mitochondrial membrane upon agonist-induced cytosolic calcium elevations. Once the mitochondrial calcium reaches a certain threshold, it activates mitochondrial permeability transition pore (mPTP) formation in a cyclophilin D (CypD)-dependent manner²⁰. When mitochondria are permeabilized to calcium and can no longer serve as sinks for the cytosolic calcium, supramaximal calcium signaling is observed after mPTP formation²⁶. The sustained high levels of cytosolic calcium, in turn, activate the calcium-dependent phospholipid scramblase TMEM16F activation, leading to PS exposure^27,28. PS exposure is accompanied by the inactivation of integrin α_IIbβ₃ by a dual mechanism involving calpain-dependent cleavage of the integrin β₃ cytoplasmic tail as well as TMEM16F-dependent phospholipid scrambling itself^29,30. Thus, procoagulant platelets, unlike proaggregatory platelets, have an inactive integrin. In addition, PS exposure provides a surface for the binding of coagulation factors, facilitating the assembly of tenase and prothrombinase complexes³. Procoagulant platelets thus dramatically accelerate coagulation⁴.

PS exposure and consequent procoagulant activity can also be induced by caspase-dependent cleavage of scramblase XKR8 in platelets undergoing intrinsic apoptosis, independent of activation^25,31. Although stimulated platelets exhibit features of intrinsic apoptosis, including Bax/Bak accumulation^21,22, whether this activates caspase^23,24 and contributes to PS exposure²⁵ remains controversial. Extrinsic apoptosis by TNFα or FasL has not been reported in platelets despite the presence of most components and some evidence for caspase 8 activation³². Platelets upon prolonged stimulation undergo MLKL-mediated necroptosis¹⁷, which is known to induce PS exposure independent of CypD or caspase³³.

This study provided a method for detecting procoagulant platelets using PS exposure (Annexin V binding), P-selectin (CD62P) expression, and integrin α_IIbβ₃ activation (PAC1 binding) as markers. These markers can help effectively distinguish procoagulant platelets from apoptotic and proaggregatory platelets¹⁵. However, annexin V binding requires the presence of high calcium levels in the medium. Hence, heparin or other thrombin inhibitors must be used when using platelet-rich plasma or whole blood samples. Further, fibrin polymerization inhibitor GPRP is employed when thrombin must be used as one of the agonists with these samples^34,35. Lactadherin or GSAO can be used as alternative markers to Annexin V, but they have their own limitations. While lactadherin is not specific to PS and binds integrins^36,37, GSAO is not commercially available and can only be sourced through research collaboration³⁸.

This article also provided supplemental methods to investigate mechanisms of PS exposure, including mitochondrial calcium influx and mitochondrial permeability transition pore formation (loss of mitochondrial membrane potential). It should be noted here that these methods can be complemented by cytosolic calcium measurements using Fluo-4 and Fluo-5N dyes. As it remains unclear if caspase activation contributes to PS exposure in stimulated platelets, we also enumerate protocols for caspase 8 and caspase 3/7 activation assays. The true "procoagulant" nature of platelets can only be affirmed by their ability to bind coagulation factors and promote thrombin generation. Thus, a thorough study of platelets for PS exposure (Annexin V binding), P-selectin (CD62P) expression and integrin α_IIbβ₃ activation (PAC1 binding), mitochondrial calcium influx, mitochondrial membrane potential, caspase activity along with coagulation factor (prothrombin) binding and platelet phospholipid dependent thrombin generation/coagulation as described in the present article would be necessary and sufficient for a comprehensive analysis of procoagulant platelets.

Authors have no competing interests to disclose.

Paresh P. Kulkarni and Keith R. McCrae, respectively, acknowledge Fellow and Pilot grant awards funded by VeloSano, Cleveland Clinic Foundation.