Abstract
Individuals whose platelets lack dense core or alpha-granules suffer varying degrees of abnormal bleeding, implying that granule cargo contributes to hemostasis. Despite these clinical observations, little is known regarding the effects of impaired platelet granule secretion on thrombus formation in vivo. The release of cargo from platelet granules requires a group of membrane proteins called SNAREs (Soluble NSF Attachment Protein Receptors) that mediate fusion of granule membranes to the plasma membrane and open canalicular system. Endobrevin/VAMP-8 is the primary vesicular-SNARE (v-SNARE) responsible for efficient release of dense core and a-granule contents. To evaluate the importance of VAMP-8-mediated secretion on the kinetics of thrombus formation in vivo, we measured platelet accumulation following laser-induced vascular injury in VAMP-8−/− mice. Three different phases of thrombus formation - initiation, maximal accumulation, and stabilized platelet accumulation - were tested. Analysis of initial thrombus formation from wild-type and VAMP-8−/− mice showed that average platelet accumulation in VAMP- 8−/− mice was 23% of accumulation in wild-type mice (P=0.009) at 30 sec following injury. There was a trend towards smaller maximal thrombus size in VAMP-8−/− mice, but the difference was not statistically significant (P=0.1). Average stabilized platelet accumulation at 180 sec in VAMP-8−/− mice was 40% of wild-type mice (P=0.05). Thus, thrombus formation is delayed and decreased in VAMP-8−/− mice, but not absent. Dense granule release occurs more rapidly than alpha-granule release, which does not occur for 2–3 min following laser-induced vascular injury. Agonist-induced dense granule release from VAMP-8−/− platelets is defective. To directly evaluate the role of dense granule release on the kinetics of thrombus formation, we assessed thrombus formation in the mouse model of Hermansky-Pudlak syndrome, ruby-eye, which lack dense granules. Thrombus formation following laser-induced vascular injury was nearly abolished in ruby-eye mice such that maximal platelet accumulation was 15% that of wild-type mice. In vitro, the thrombin doses required to induce irreversible aggregation in wild-type, VAMP-8−/−, and ruby-eye platelets were 25 mU, 50 mU, and 150 mU, respectively. Incubation with apyrase had little effect on thrombin-induced aggregation of VAMP-8−/− or ruby-eye platelets. In contrast, incubation of wild-type platelets with apyrase reduced their thrombin sensitivity compared to that of ruby-eye platelets. Supplementation with a substimulatory ADP concentration reversed the thrombin-induced aggregation defect in VAMP-8−/− and ruby-eye mice. Thus, defective ADP release is the primary abnormality leading to impaired aggregation in VAMP-8−/− and ruby-eye mice. Tail bleeding times were assessed in VAMP- 8−/− mice to evaluate the role of VAMP-8 in hemostasis. In contrast to ruby-eye mice, which have a markedly prolonged bleeding time, tail bleeding times in VAMP-8−/− mice were not significantly prolonged compared to those in wild-type mice. These results demonstrate the importance of VAMP-8 and dense granule release in the initial phases of thrombus formation and validate the distal platelet secretory machinery as a potential target for anti-platelet therapies.
Endobrevin/VAMP-8–dependent dense granule release mediates thrombus formation in vivo
