Endothelium but Not Platelet-Derived Protein Disulfide Isomerase Is Required for Fibrin Generation during Thrombus Formation in Vivo.

Abstract

We have previously reported that protein disulfide isomerase is required in wild-type mice for platelet thrombus formation and fibrin generation in an in vivo laser injury model of thrombosis (Cho et al. J. Clin. Invest. , 2008; 118:1123–31). Fibrin deposition after laser injury to the vessel wall in Par4−/− mice, lacking the G protein-coupled platelet thrombin receptor, is independent of platelets or requires minimal platelet activation or accumulation (Vandendries et al. Proc. Natl. Acad. Sci. , 2007; 104:288–92). However, protein disulfide isomerase inhibitors have a dramatic effect on fibrin accumulation in Par4− mice, suggesting that these inhibitors may function by a platelet independent mechanism. Here, we compare the contributions of endothelium and platelet-derived protein disulfide isomerase to fibrin generation in the mouse laser injury model of thrombosis. In vitro studies using cultured human umbilical vein endothelial cells and human aortic endothelial cells show that protein disulfide isomerase can be secreted rapidly into the culture medium from these cells upon thrombin stimulation. Using intravital microscopy, we observe that protein disulfide isomerase is not detectable on the vessel wall prior to laser injury but can be detected on the injured cremaster arteriolar wall and in the developing thrombus very rapidly after laser induced injury in the live mouse. The median integrated fluorescence intensity for protein disulfide isomerase in wild-type mice was compared to wild-type mice injected with 10ug/g mouse of Integrilin, an inhibitor of platelet activation and platelet thrombus formation, and thus, an inhibitor of the contribution of platelet derived protein disulfide isomerase to thrombus formation. Protein disulfide isomerase expression was similar in both treated and untreated animals upto 30 seconds post-laser injury. After 30 seconds, the expression of protein disulfide isomerase in integrilin treated mice was significantly decreased compared to that in untreated mice, indicating that the initial protein disulfide isomerase was derived from the endothelium and later additional protein disulfide isomerase was derived from the platelets following their accumulation in the developing thrombus. Fibrin deposition, a measure of thrombin generation was comparable in wild-type mice that had been treated with Integrilin or treated with a control buffer, suggesting that endothelial-derived protein disulfide isomerase was sufficient for fibrin generation. The rate and amount of fibrin generation was indistinguishable in both groups. Furthermore, inhibition of the protein disulfide isomerase with the function blocking monoclonal antibody RL-90 (3ug/g mouse) eliminated any fibrin deposition in wild-type mice that had been treated with Integrilin. Taken together, these data indicate that endothelium-derived protein disulfide isomerase is necessary to support fibrin deposition in vivo in our laser injury model of thrombus formation. The initial protein disulfide isomerase expressed at the site of injury is derived from endothelial cells but platelets activated at the site of thrombus formation contribute, amplify and sustain protein disulfide isomerase expression.