Approximately 500 μl of murine blood were centrifuged at 200 × g for 15 minutes. The resulting supernatants were transferred to new tubes and used as platelet rich plasma (PRP). The platelet aggregation measurements were done according to the method developed by Born and Cross [26] using an APACT photometer (Labor, Ahrendsburg, FRG) and the APACT software (APACT 1.4). Subsequently samples of the PRP were adjusted with buffer (5 mM Hepes, 150 mM NaCl, pH 7.3) to 250,000 platelets/μl. A constant platelet count was used throughout allowing comparisons between wild type and knock out murine and human samples. 250 μl adjusted platelets were prewarmed to 37°C and used in the assay, and the aggregation process was started by adding 25 μl of a prewarmed 16.5 mg/ml ristocetin solution (DiaMed Diagnostica, Bensheim, FRG; concentration in the assay: 1.5 mg/ml), a reagent which worked reliably in our hands with mouse and human platelets. Each murine assay represents a different animal. Analyses of human blood samples were performed as described in the manufacturer's protocol. Several repeated measurements were done per human individual to control the reproducibility. The acquired aggregation curve data, measured as ascending degree of transmission, were analysed for maximal slope values, while the maximal aggregation amplitude of every single curve was set to 100 %. In general, all platelet procedures were done at 22°C within four hours after collecting the blood sample.