Efficient Barrier Protective Signaling by Activated Protein C Is Mechanistically Linked to Protein C Activation on Endothelial Cells.

Protein C (PC) is activated by thrombomodulin-bound thrombin on the endothelial cell surface and activated protein C (APC) inhibits blood coagulation in a negative feedback loop. Endothelial PC receptor (EPCR) can bind PC/APC and activation of EPCR-bound PC is enhanced. Exogenous APC has barrier protective effects on endothelial cells that depend on EPCR binding and protease activated receptor-1 (PAR1) cleavage and that may contribute to the anti-inflammatory effects of APC. Plasma APC concentrations in vivo are low compared to the substrate PC and in order to induce protective signaling exogenous APC has to compete with PC for EPCR binding. In this study we investigated whether the endogenous PC activation pathway may be linked to efficient protective responses analyzing endothelial barrier permeability in a dual chamber system. When endothelial EA.hy926 cells were incubated for 3 h in the presence of 80 nM purified PC and different concentrations of thrombin a dose-dependent linear increase of APC activity in the cell medium was observed over time. APC generation was detectable upon incubation with 20 pM thrombin or higher and a significant barrier protective response to 20 pM thrombin was found only in the presence of PC. 40 pM thrombin enhanced barrier integrity in the presence and absence of PC, consistent with our previous results. To exclude direct thrombin effects on endothelial permeability and to compare protective effects of exogenous and endogenously generated APC, we used the anticoagulant double mutant thrombin W215A/E217A (WE). WE was about 10 times less active than wildtype thrombin for PC activation in our system. However, PAR1-dependent induction of MAP kinase phosphorylation required more than 1000-fold higher concentrations of the thrombin mutant. Thus, 1–10 nM WE leads to APC generation without directly inducing PAR1-dependent signaling. When cells were incubated with various concentrations of exogenous APC or WE+80 nM PC, barrier protective effects of 5 nM exogenous APC and 2 nM WE+80 nM PC (1.3 nM APC generated after 3 h) were similar. Because APC is generated at a constant rate during the incubation period, the average concentration of generated APC in the cell medium was only about 0.65 nM, suggesting that signaling by endogenously generated APC was significantly more efficient. To conclusively demonstrate that protective effects in response to WE are mediated by APC generation, we used recombinant zymogen wildtype PC and a PC variant with a substitution of the active site serine with alanine (PC S360A). Cells were incubated with control or 80 nM wildtype PC and PC S360A, in the presence or absence of WE (4 nM) and exogenous APC (3.3 nM). WE induced protective signaling only in the presence of wildtype PC but not PC S360A. Barrier protective effects of exogenous APC were blocked by both wildtype PC and PC S360A, consistent with their expected role as competitive inhibitors for APC binding to EPCR. These data demonstrate that efficient barrier enhancement by APC is indeed mechanistically coupled to the PC activation pathway. Signaling by endogenously generated APC may play an important role in the regulation of inflammation.