Regulation of thrombus formation by apelin. (A) In vitro thrombus formation performed in a whole-blood perfusion assay of WT or apelin^−/− mice over a fibrillar collagen matrix at a venous (150 seconds⁻¹) or an arterial (1200 seconds⁻¹) shear rate (left). Thrombus formation was quantitated by assessment of the mean percentage of the total area covered by thrombi ± SEM of at least 3 independent experiments (right). Statistical significance was determined by unpaired Student t test (***P < .001). (B) In vivo thrombosis model in mesenteric vessels (venules [V] and arterioles [A]) of WT or apelin^−/− mice after FeCl₃-induced injury. Adhesion and thrombus formation of fluorescently labeled platelets were monitored by intravital video microscopy. Representative images of mesenteric vessels after 0, 20, and 30 minutes are shown (left). Occlusion times of the venules and arterioles in WT and apelin^−/− mice are shown, with means indicated by horizontal lines (right). Statistical significance was determined by unpaired Student t test (**P < .01). (C) In vitro thrombus formation performed in a whole-blood perfusion assay of WT mice over a fibrillar collagen matrix at an arterial shear rate of 1200 seconds⁻¹ in the presence of PBS (control) or apelin-13 (10 μM) (left). Results are expressed as the relative percentage ± SEM of the mean immunofluorescence intensity (IFI) of WT mice (right). Statistical significance was determined by unpaired Student t test (***P < .001). Adhesion and thrombus formation of fluorescently labeled platelets were monitored by intravital video microscopy after injection of PBS or apelin-13 (50 nmol/kg). (D) In vivo thrombosis model in mouse mesenteric vessels after FeCl₃-induced injury. Representative images of mesenteric vessels after 0 and 30 minutes are shown (left). Occlusion times of the venules and arterioles are shown, with means indicated by horizontal lines (right). Statistical significance was determined by 1-way ANOVA with post hoc Dunnett test (***P < .001).