Procoagulant Role Of Necrotic Platelets Demonstrated Using Novel Platelet Necrosis Marker

Strong agonist stimulation generates a platelet subpopulation characterized by phosphatidylserine (PS) exposure, loss of mitochondrial membrane potential and high fibrinogen retention. This population is proposed to be procoagulant, dependent on formation of the mitochondrial permeability transition pore (mPTP) with a distinct role from activated aggregratory platelets. These platelets have features of necrosis. The functional relevance of necrotic platelets in vivo is unknown due to lack of a suitable marker for these platelets. We show that a novel small molecule cellular necrosis marker, GSAO¹, labels a procoagulant platelet subpopulation with features of necrosis and use it to explore the functional role of these platelets.

We demonstrated using flow cytometry analysis of washed human platelets that fluorescently tagged GSAO labels a subpopulation of P-selectin positive platelets after thrombin and collagen stimulation with features of necrosis: high annexin V binding, calcein loss and dependence on exogenous calcium. This population is not dependent on the intrinsic apoptosis pathway as there was no change with pancaspase inhibition using ZVADFMK prior to dual agonist stimulation (p=0.567, n=5). In contrast, inhibition of mPTP formation through cyclophilin-D inhibition with cyclosporine A significantly inhibited GSAO+ve platelet generation (p<0.001, n=5), confirming dependence on the mitochondrial necrosis pathway. Mass spectrometry analysis of biotin-GSAO labelled proteins from platelets after streptavidin pull down identified thromboxane A synthase (TBXAS-1) as the major binding ligand after dual stimulation. Binding to TBXAS-1 was abrogated by dithiol alkylation, showing the mechanism of retention of GSAO in necrotic platelets is via covalent cross linking of closely-spaced cysteine thiols in the ligand. This allows persistent signal from the probe within the necrotic platelet with no evidence of washout.

GSAO+ve platelets correlated with procoagulant potential as measured by peak and endogenous thrombin potential in the calibrated automated thrombogram (CAT) assay. Linear regression analysis showed a significant relationship between % change in GSAO+ve platelets and % change in peak thrombin after treatment with cyclosporine A or in absence of exogenous calcium (R²=0.648, p<0.01), indicating that GSAO identifies a procoagulant subpopulation. In contrast, no relationship was seen between P-selectin and peak thrombin values (R²=0.002). Inhibition of platelet activation by aspirin had no effect on the generation of GSAO+ve platelets indicating a potential uncoupling between platelet activation and necrosis pathways.

After establishing that the imaging compound does not affect platelet function and coagulation in vitro, or thrombus formation in vivo, we went on to investigate the presence of GSAO+ve necrotic platelets in thrombus formation in a collagen dependent (ferric chloride) and collagen independent (laser injury) murine model of thrombosis. Confocal intravital imaging of the cremaster arterioles with fluorescent GSAO and tagged-CD42b demonstrated GSAO+ve platelets in the occlusive platelet aggregate after initiation with 10% ferric chloride. The GSAO+ve aggregating platelets specifically colabeled with calcium sensing dye rhodamine 2 indicating high sustained intracellular calcium, consistent with a necrotic phenotype. There was no signal with active site replaced control GSCA. In contrast, the laser injury model showed minimal staining with GSAO three minutes post laser injury.

Using a novel platelet necrosis marker, we are able to demonstrate that necrotic platelets are procoagulant and present in the occlusive ferric chloride model and not in the non-occlusive laser injury model of thrombosis. This suggests excess platelet necrosis may be a key driving factor underlying pathological occlusive thrombi. GSAO is a promising tool for understanding factors that potentiate platelet necrosis which may offer attractive anti-thrombotic targets.

1. Park D, Don AS, Massamiri T, et al. Noninvasive imaging of cell death using an hsp90 ligand. J Am Chem Soc. 2011;133(9):2832-2835.