JNK1^−/− mice display impaired thrombus formation in vitro and in vivo. The role of JNK1 in thrombus formation was investigated in vitro (A), in a whole-blood perfusion assay carried out over a fibrillar collagen matrix at a shear rate of 1500 seconds⁻¹ and in vivo (B), in a model of arteriolar and venular thrombosis induced by photochemical damage of the cecum. (A) Whole blood from WT and JNK1^−/− mice, collected in PPACK (80μM), was fluorescently labeled by incubation with rhodamine 6G (10 μg/mL) for 10 minutes, then perfused through fibrillar collagen-coated glass microcapillaries at a shear rate of 1500 seconds⁻¹ for 90 seconds. After perfusion, the formation of thrombi was observed under an epifluorescence microscope (original magnification ×20). The mean percentage of the total area covered by platelets plus or minus SEM was calculated in 3 independent experiments, and the extent of thrombus formation was evaluated in all 3 dimensions, by calculating relative thrombus volume, measured as the IFI per μm² of each thrombus plus or minus SEM. The statistical significance of differences between WT and JNK1^−/− mice was determined in unpaired Student t tests (***P < .001). (B) Photographs show the progression of thrombus formation induced by photochemical injury to cecum arterioles (a) and venules (v) in WT and JNK1^−/− mice. Calcein-AM fluorescently labeled platelets (10⁸ platelets) were injected into the mice, with an intravenous bolus of the photoactivated dye Rose Bengal (20 mg/kg body weight). Injury to cecum arterioles and venules was induced by exposing the vessels to a wavelength of 540 nm for 30 seconds. Thrombus formation was visualized in real time using an inverted epifluorescent microscope (Nikon Eclipse TE 2000-U) using a lens at 20×/0.5 numeric aperture (Nikon). Images were acquired using a CCD CoolSNAP HQ2 camera (Photometrics/Roper Scientific) and processed with Metamorph 7.0r1 software (Universal Imaging Corporation). Vessel occlusion was defined as the stopping of blood flow for at least 1 minute. The arterioles and venules irrigating the cecum selected for analysis had diameters of 60 plus or minus 10 μm and 107 plus or minus 15 μm, respectively. The dot plot shows occlusion times for arterioles and venules as a result of photochemically induced thrombosis in WT (n = 7) and JNK1^−/− (n = 6) mice. Means are indicated by horizontal lines. Statistical differences between WT and JNK1^−/− mice were evaluated in 2-tailed Mann-Whitney tests (**P < .01; ***P < .001).