We studied how the intrinsic and extrinsic coagulation pathways contribute to thrombus formation. Ex vivo experiments were performed with citrated blood re-calcified before perfusion over different thrombogenic substrates. This allowed evaluation of fibrin deposition along with platelet adhesion/aggregation in real time and under relevant shear rates. Under these conditions, human melanoma cells expressing human tissue factor (TF) supported platelet adhesion/aggregation and fibrin formation, but only under relatively low shear rate (200–800 s-1). Anti-TF antibodies markedly inhibited both platelet and fibrin deposition, and no reactivity was observed using the parental cell line not transfected with TF. Blood perfusion over fibrillar collagen type I resulted in platelet adhesion/aggregation and subsequent fibrin deposition even under relatively high shear rate (1500 s-1). Corn trypsin inhibitor (CTI), a specific inhibitor of coagulation factor XIIa, had no effect on platelet adhesion/aggregation but inhibited fibrin deposition, while anti-TF antibodies had no significant effect. Preincubation of blood with prostaglandin E1 or a monoclonal antibody against integrin αIIbβ3 blocked not only thrombus growth on the surface but also fibrin formation. Such a finding indicates that thrombin generation leading to fibrin deposition occurs after platelet activation. Blood perfusion over dermal fibroblast extracellular matrix (F-ECM) from mice expressing human but lacking murine TF resulted in the rapid formation of platelet- and fibrin-rich thrombi at all shear rates tested (up to 1500 s-1). Using this thrombogenic substrate containing both collagen and TF, we found that blockade of the intrinsic coagulation pathway with CTI had a minimal effect on platelet adhesion/aggregation and fibrin formation, while inhibition of the extrinsic coagulation pathway with anti-human TF antibodies reduced thrombus growth and stabilization and abolished fibrin formation. The results of these ex vivo studies demonstrate that the relative contribution of the intrinsic and extrinsic coagulation pathways to fibrin and platelets deposition depends on the composition of the thrombogenic surface exposed to blood flow, and indicate also that CTI and anti-TF antibodies can be used as specific inhibitors of thrombin generation through the intrinsic and extrinsic coagulation pathways, respectively. We then established a model of carotid artery injury induced by ferric chloride to evaluate whether the information obtained under controlled experimental conditions ex vivo can help interpret the mechanisms underlying arterial thrombus formation in vivo. In this model, all control C57BL6 mice exhibited a stable carotid artery occlusion within a predictable time frame. To obtain a quantitative parameter of the progression of thrombosis we calculated a flow index. This represents the ratio between the volume of blood that flew through the injured artery from the induction of the lesion to occlusion (or to the end of the predetermined 30 min observation period) and the volume of blood that during the same period of time should have flown through the artery if the initial (pre-injury) flow rate had been maintained. We found that blocking the intrinsic coagulation pathway with CTI injected through the jugular vein significantly delayed, and in most cases prevented, the formation of a stable occlusion, resulting in a significantly altered flow index. The injection of antibodies against human TF, on the other hand, resulted in a trend towards thrombus destabilization, but in most cases the artery still occluded and the flow index was less significantly altered that with CTI. Of note, CTI at a dose that inhibited carotid artery occlusion had no effect on the tail bleeding time. Co-administration of CTI and anti-TF antibodies showed a cooperative effect across the tested concentration range. In conclusion, extrinsic and intrinsic coagulation pathways have complementary roles in thrombus formation and stabilization, and the specific contribution of each depends upon the nature of the thrombogenic surface, i.e. of the causative lesion. The marked effect of factor XIIa inhibition in preventing carotid artery occlusion suggests that a functional link between contact phase activation and tissue factor pathway leading to thrombin generation may be operative under defined conditions in vivo.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author