Signaling by Activated Protein C in Endothelial Cells Is PAR1 Dependent but Distinct from Thrombin Signaling.

Thrombin initiates fibrin formation and platelet activation, and activates protein C, generating activated protein C (APC) that inhibits blood coagulation by a negative feedback loop. Thrombin has also proinflammatory effects through activation of cellular protease activated receptor-1 (PAR1). Endothelial cell protein C receptor (EPCR) can bind both protein C and APC and activation of EPCR-bound protein C is enhanced. Results from animal models and clinical trials indicate that APC has potent protective effects in systemic inflammation that are independent from its well established anticoagulant function and recombinant APC was recently approved to treat patients with severe sepsis. The molecular basis for APC’s anti-inflammatory effects is incompletely understood. We have identified PAR1 and EPCR as part of a novel APC signaling pathway in quiescent endothelial cells, raising the question how the same receptor PAR1 can mediate both pro- and anti-inflammatory effects. In an overexpression system in PAR-deficient fibroblasts, wildtype PAR2 but not a PAR2 variant with an Arg36 to Ala substitution at the P1 position was activated by APC, indicating that APC can activate PAR2 in addition to PAR1 through a canonical cleavage mechanism. Therefore, we tested whether endothelial cell PAR2 can be activated by APC under conditions where endogenous PAR2 expression is upregulated. Even when PAR2 expression was highly upregulated in inflammatory cytokine-stimulated human umbilical endothelial cells (HUVECs), signaling by APC was strictly dependent on PAR1 cleavage and signaling. Consistent with these results in HUVECs, intravenous injection of APC in wildtype, PAR1−/−, and PAR2−/− mice demonstrated that PAR1 is the major murine receptor that mediates induction of the transcript for monocyte chemoattractant protein-1 in the lung in response to APC. This indicates that indeed the same receptor PAR1 mediates signaling by APC and thrombin both in vitro and in vivo. To test the possibility that APC diminishes proinflammatory thrombin-PAR1 signaling by downregulating cellular levels of functional PAR1, we tested whether preincubation with APC can desensitize Erk1/2 phosphorylation by thrombin. Phospho-Erk1/2 was induced by APC dependent upon PAR1 cleavage, but APC-pretreated cells still responded to PAR1-dependent thrombin signaling, suggesting that only a fraction of the cellular PAR1 is subject to cleavage by APC. These results indicate that APC does not block thrombin signaling by desensitation at the receptor level. Large-scale gene expression profiling demonstrated that APC and thrombin had specific effects on gene expression in tumor necrosis factor α (TNFα )-perturbed endothelial cells that were not detected in quiescent cells. Transcripts for several proapoptotic genes including p53 and thrombospondin-1 were downregulated by APC but not by thrombin or PAR1 agonist peptides in TNFα-stimulated HUVECs. Western blotting confirmed that in TNFα-perturbed HUVECs pretreatment with APC significantly reduced the increase in cellular p53 protein levels in response to the cytotoxic doxorubicin. This APC effect was dependent on EPCR binding and PAR1 cleavage. Thrombospondin-1 protein levels were similarly downregulated by APC but upregulated by thrombin. Both down- and upregulation by APC and thrombin, respectively, were PAR1 dependent. These findings demonstrate that the same receptor on the same cell type can mediate opposite biological effects and they suggest that EPCR cosignaling may modify PAR1-dependent APC signaling in endothelial cells.