Novel R41Q-PAR1-Modified Mice Enable Proof-of-Concept Studies for in Vivo Mechanisms of Action for Thrombin (IIa) and Activated Protein C (APC)

Introduction: Thrombin (IIa) and activated protein C (APC) are serine proteases involved in coagulation and inflammatory responses that affect many cell types in the body. IIa employs the GPCR, protease activated receptor 1 (PAR1), to promote endothelial barrier disruption, vascular leakage, and inflammation. In contrast, APC requires PAR1 for its opposing actions to stabilize endothelial barriers and to provide anti-inflammatory and anti-apoptotic activities. Studies of murine in vivo injury models using PAR1 knockout mice show that APC requires PAR1 to reduce sepsis-induced mortality and to provide robust neuroprotection following ischemic stroke. Extensive in vitro studies support the hypothesis that IIa’s cleavage at R41 in PAR1 initiates its signaling. Although APC was long thought, paradoxically, to act also via cleavage at R41, we recently proposed that APC’s cleavage at R46 initiates its endothelial barrier-protective and cytoprotective signaling via biased signaling. Since PAR1 knock-out mice cannot provide mechanistic data for testing these hypotheses in vivo, mice carrying the point mutation of R41 to Q in PAR1 were generated to enable mechanistic studies to test whether or not IIa and APC require Arg41 for their PAR1-dependent effects.

Methods: Using C57BL/6-derived embryonic stem cells and standard gene targeting methods, we prepared C57BL/6 mice carrying the PAR1 mutation of R41 to Q. IIa-induced and APC-induced signaling, detected as phosphorylation of ERK1/2 or Akt in Brain Microvascular Endothelial Cells (BECs), was quantified using immunoblotting. BECs were obtained from homozygous 41QQ-PAR1 mice and wild type 41RR-littermates. Endothelial barrier disruption of cultured BECs was assayed using Trans-Endothelial Resistance (TER) assays (iCelligence, Acea, San Diego). Mortality of wild type and 41QQ-PAR1 mutant mice that was caused by live E. coli-induced pneumonia and to endotoxin was determined using standard methods. The ability of a cytoprotective-selective murine APC mutant (5A-APC) to reduce mortality of E.coli-challenged wild type and homozygous mutant mice was determined.

Results: Upon breeding of R41Q-PAR1 heterozygous mice, the progeny did not fit a Mendelian pattern and yielded only 14% rather than 25% homozygous 41QQ mice. This reduced yield of homozygous mutant mice was similar to the previously reported low yield of homozygous PAR1 knockout mice. Homozygous 41QQ-PAR1 mice showed normal protein expression in BECs for PAR1 and endothelial cell protein C receptor (EPCR) antigens. When BECs from homozygous mutant mice were compared to those from wild type littermates, the IIa-induced vascular disruption in TER assays was greatly reduced by the mutation. Intracellular Ca²⁺ release, a hallmark of IIa-induced signaling, was greatly impaired (>90%) in BECs from homozygous mutant mice compared to wild type controls. IIa-induced phosphorylation of ERK1/2 in BECs was also significantly reduced by the mutation whereas APC-induced phosphorylation of Akt was not significantly affected. In murine sepsis-induced mortality studies, homozygosity for the R41Q-PAR1 mutation conveyed considerable resistance to death induced by either E. coli pneumonia or endotoxin in female mice but not in male mice. Tests to determine whether 5A-APC rescued male mice from sepsis-induced lethality showed that homozygous 41QQ-PAR1 mice were entirely responsive to 5A-APC therapy because 5A-APC treatment reduced mortality from 50 % to 0 % (see Figure). Wild type control mice also showed a beneficial response with reduced mortality in response to 5A-APC therapy, as previously described.

Conclusions: These studies show that mutation of Arg41 to Gln in murine PAR1 diminishes or eliminates signaling induced by IIa but not by APC. Moreover, the ability of cytoprotective-selective 5A-APC to reduce bacteria-induced septic mortality in 41QQ-PAR1 mutant mice provides strong in vivo proof-of-concept data for PAR1 activation caused by non-canonical cleavage by APC. In summary, the 41QQ-PAR1 mutant mouse provides a unique and powerful tool to define in vivo requirements for cleavage sites that enable PAR1 signaling activities induced by IIa, APC or other proteases.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal