Expression of Functional Human Proteinase Activated Receptor (PAR)-1 on Mouse Platelets

Thrombin activates platelets by specific cleavage of PARs leading to platelet aggregation. Human (h) platelets express PAR-1 and PAR-4, each with a different sensitivity to thrombin and specific kinetics of activation/deactivation influencing thrombogenesis and thrombus stability. PAR-1 and PAR-4 are candidates for the generation of anti-thrombotic drugs; indeed, a PAR-1 inhibitor (Vorapaxar) is already in clinical use. However, side effects such as bleeding warrant a continuous effort towards the generation of novel alternative inhibitors. In this regard, human PAR-4 may be a more selective antithrombotic target with a lesser impact on hemostasis. Mice are commonly used to test anti-thrombotic drugs and their mechanisms of action, but cannot be used in this instance because mouse (m) platelets express different PAR subtypes than human, i.e. PAR-3 and PAR-4. Moreover, mPAR-3 cannot be cleaved by thrombin and only acts as a co-receptor enhancing mPAR-4 response. In human platelets, PAR-1 and PAR-4 can act independently of one another. Yet, a different platelet membrane receptor, unrelated to PARs, which binds thrombin with high affinity - glycoprotein (GP) Ibα in the GPIb-IX-V complex - enhances PAR-1 response to low-dose thrombin through a mechanism only partially elucidated. Humanizing mouse platelets with respect to PAR expression would greatly facilitate explaining the functional interplay of GPIb with PAR-1 and PAR-4 and, notably, the role of these different PARs in hemostasis and thrombosis. Attempts to introduce hPAR-1 in mouse platelets have been so far unsuccessful (French et al. 2016, Arachiche et al. 2014) and claims have been made that obstacles to achieve this goal may be unsurpassable. Using a different approach, we introduced a floxed hPAR-1 transgene in the ROSA26 locus of C57BL6/J mice under the strong CAG promoter. The presence of an IRES sequence and a second transgene for eGFP allowed us to screen efficiently for platelets expressing hPAR1. Mice positive for the floxed-transgene were then bred with either a germ-line Cre recombinase-expressing strain (EIIa-Cre) or one in which Cre expression was linked to a platelet specific receptor (PF4-Cre). The percentage of platelets isolated from EIIa-Cre/PAR1^+/- mice expressing eGFP was between 35 to 40% and breeding to homozygosity did not increase transgene expression. In contrast, eGFP expression was positive in nearly all PF4-Cre/PAR1^+/-mice, suggesting that suppression of hPAR1 expression at the germ-line level could be advantageous for megakaryocyte development. As a functional test for transgene-expressed hPAR-1 we measured aggregation and intracellular calcium increase in washed platelets from PF4-Cre/PAR1^+/+ mice stimulated by 5 μmol/liter of a specific PAR-1 activation peptide (TFLLRN, P1-AP). In both assays, the response of human or hPAR-1 expressing mouse platelets was indistinguishable; littermate control platelets lacking transgene expression (floxed hPAR1) did not respond to a 10 times higher concentration of peptide. Interestingly, when stimulated with a low dose of thrombin (0.25 nM) aggregation and intracellular calcium increase of PF4-Cre/PAR1^+/+ platelets was similar to littermate controls. We reasoned that this could be a consequence of the presence of mPAR3, owning to its high affinity for thrombin. Thus we proceeded to cross PF4-Cre/PAR1^+/+ with mPAR3^-/- mice, so to obtain mice that express only mPAR4 and hPAR1. Aggregation induced with P1-AP (5 μM) was similar with platelets expressing PF4-Cre/PAR1^+/+ and mPAR3^-/-/PF4-Cre/PAR1^+/+. As expected from the cofactor role of mPAR3, the minimal amount of thrombin necessary to aggregate fully platelets from mPAR3^-/-/PF4-Cre/PAR1^-/- was higher than in wild-type mice (2 nM vs 0.25 nM); however, 2 nM was also the minimum thrombin concentration necessary to activate mPAR3^-/-/PF4-Cre/PAR1^+/+ platelets. Therefore, the presence of a functional hPAR1 responsive to P1-AP does not influence the response of mouse platelets to low thrombin concentrations mediated by mPAR4, indicating that hPAR-1 and mPAR-4, like the human homologue, respond to thrombin independently of one another. In conclusion, to date we have shown that expression of a functional hPAR1 in mouse platelets is possible and should provide a versatile animal model to address several open questions on the role of platelet PAR-1 and GPIbα in hemostasis and thrombosis.