Protection Against Vascular Leakage in Vivo by a Peptide Mimetic of the Novel Tethered Ligand Generated by Non-Canonical Cleavage of Protease Activated Receptor 1 by Activated Protein C

Abstract 497

Activated protein C (APC) exerts cytoprotective activities on vascular endothelium that require protease-activated receptor 1 (PAR1) whereas thrombin acting via PAR1 causes endothelial disruptive, proinflammatory actions. Last year our laboratory elucidated a unique biochemical mechanism leading to the APC's cytoprotective signaling initiation, revealing that APC can cleave PAR1 at Arg46 and that a synthetic peptide, TR47, comprising PAR1 residues 47–66, stimulates signaling in endothelial cells reflected in Akt phosphorylation and anti-apoptotic activity (see Blood 2011;118:534). Here we report novel in vitro and in vivo insights concerning the downstream effects of APC-specific cleavage at Arg46. First, using the EA.hy926 endothelial cell line, we showed that TR47 induced sustained phosphorylation of glycogen synthase kinase 3 beta (GSK3beta) at Ser9 starting at 30 min. Moreover, the TR47 time-course was similar to Akt phosphorylation. A scrambled control peptide (scrTR47) was unable to induce GSK3beta phosphorylation. TR47-induced phosphorylation of GSK3beta was inhibited by the PAR1 antagonist SCH79797, indicating that TR47-induced signaling required PAR1. Cleavage of PAR1 at Arg41 by thrombin induces phosphorylation of extracellular-regulated kinase (ERK1/2). TRAP peptide (TFLLRNPNDK), the canonical PAR1 agonist, induced strong and immediate phosphorylation of ERK whereas neither TR47 nor scrTR47 induced ERK phosphorylation. In contrast, treatment of EA.hy926 endothelial cells with TRAP did not result in phosphorylation of Akt at Ser473 or GSK3beta at Ser9. In agreement with peptides data, thrombin did not induce Akt or GSK3beta phosphorylation whereas APC did so. Thus, PAR1 cleavage at Arg46 results in phosphorylation of Ser473-Akt and Ser9-GSK3beta, whereas cleavage of PAR1 at Arg41 results in phosphorylation of ERK1/2. Activation of PAR1 by thrombin results in Ras homolog gene family member A (RhoA) activation and disruption of the endothelial barrier. In contrast, activation of PAR1 with APC results in activation of Ras-related C3 botulinum toxin substrate 1 (Rac1) and endothelial barrier protection. Using active Rac1 pulldown assays with p21-activated kinase (PAK-1)-conjugated beads and quantifying the ratio of active Rac1 over total Rac1, we showed that both TR47 and APC, but not scrTR47, activated Rac1. In an endothelium barrier transwell assay using Evans Blue to quantify thrombin-induced leakage, TR47 and APC, but not scrTR47 or TRAP, decreased vascular permeability by 40% (P < 0.05). Thus, cleavage of PAR1 at Arg46 but not at Arg41 results in endothelial barrier protective effects in vitro. To test whether TR47 also reduces vascular leakage in vivo, we setup a novel modification of the modified Miles assay to assess the effect of TR47 on VEGF-induced vascular leakage in the skin. Immunocompetent SKH1 hairless mice were used to avoid the need for hair removal that often can result in artifactual leakage due to inflammation of the skin. Evans Blue was injected intravenously followed 30 min later by 2μg of recombinant mouse APC, 125 μg of TR47 or PBS (i.v.). Recombinant VEGF-165 (75 ng, subcutaneous) or vehicle (BSA) was injected thereafter on the abdomen. After 30 min mice were placed on the Odyssey infrared fluorescence Imager and the total amount of vascular leakage was quantified as the amount of Evans Blue accumulated in the VEGF or BSA injection sites determined by infrared fluorescent at 700 nm. APC decreased leakage by 50%. TR47, but not scrTR47, injected 30 min before VEGF decreased vascular leakage by 45% (P < 0.05, n = 6 mice) compared to PBS control. Neither TR47 nor scrTR47 affected vascular leakage in the absence of VEGF. In summary, the TR47 peptide representing the sequence of the novel N-terminus that is generated by cleavage of PAR1 at Arg46 exerts remarkable biologic activities in vitro and in vivo that reflect the general cytoprotective activity profile of APC but not that of thrombin. Based on these results we propose a novel paradigm for the biochemical mechanisms of APC via PAR1 involving generation of a new N-terminal tethered ligand, which is a biased agonist that initiates APC-like cytoprotective signaling pathways.