Alterations in platelet secretion differentially affect thrombosis and hemostasis

We genetically manipulated the major platelet vesicle-associated membrane proteins (VAMP2, VAMP3, and VAMP8) to create mice with varying degrees of disrupted platelet secretion. As previously shown, loss of VAMP8 reduced granule secretion, and this defect was exacerbated by further deletion of VAMP2 and VAMP3. VAMP2(Delta)3(Delta)8(-/-) platelets also had reduced VAMP7. Loss of VAMP2 and VAMP3 (VAMP2(Delta)3(Delta)) had a minimal impact on secretion when VAMP7 and VAMP8 were present. Integrin alpha IIb beta 3 activation and aggregation were not affected, although spreading was reduced in VAMP2(Delta)3(Delta)8(-/-) platelets. Using these mice as tools, we asked how much secretion is needed for proper thrombosis and hemostasis in vivo. VAMP2(Delta)3(Delta) mice showed no deficiency, whereas VAMP8(-/-) mice had attenuated formation of occlusive thrombi upon FeCl3-induced arterial injury but no excessive bleeding upon tail transection. VAMP2(Delta)3(Delta)8(-/-) mice bled profusely and failed to form occlusive thrombi. Plasma-coagulation factors were normal in all of the strains, but phosphatidylserine exposure was reduced in VAMP2(Delta)3(Delta) and VAMP2(Delta)3(Delta)8(-/-) platelets. From our data, an similar to 40% to 50% reduction in platelet secretion in vitro (dense and a granule) correlated with reduced occlusive thrombosis but no compromise in hemostasis. At a >50% reduction, thrombosis and hemostasis were defective in vivo. Our studies are the first systematic manipulation of platelet exocytic machinery to demonstrate a quantitative linkage between in vitro platelet secretion and hemostasis and thrombosis in vivo. The animals described will be invaluable tools for future investigations into how platelet secretion affects other vascular processes.