147 Platelet cox-1 knockout mouse as a model of the effects of aspirin in the cardiovascular system

Abstract

Introduction Aspirin acts as an anti-thrombotic drug by inhibiting platelet cyclooxygenase-1 (COX-1) and blocking the production of thromboxane A2, but it also has off-platelet actions that may limit its effectiveness. Since mice have very marked differences in aspirin pharmacokinetics they cannot be used to explore the effects of low dose aspirin in humans. To address these questions, we developed platelet COX-1 knockout mice (platelet-COX-1-/-) to recapitulate the effect of low-dose aspirin in humans. Methods and Results Cox-1fl/fl mice were paired with Pf4-Cre mice to delete COX-1 in the MK lineage. The absence of COX-1 in platelets was confirmed by Western blot analysis, confocal immunofluorescence microscopy and by analysing the distribution of the COX-1 signal and mass with an high-throughput confocal imaging system. The production of thromboxane B2 (TxB2) induced by collagen, A23187 and the PAR-4 activating peptide in platelet rich plasma and whole blood was significantly decreased in platelet-COX-1-/- mice. This effect was not further reduced by the addition of aspirin in vitro . Further eicosanomic analysis was performed using LC-MS/MS of samples from untreated and A23187-stimulated whole blood. 11- and 15-hydroxyeicosatetraenoic acid (HETE), alongside TxB2 formation were inhibited after platelet COX-1 deletion. Production of 12-HETE, coming from 12-lipoxygenase (LOX), and 5-HETE coming from leukocyte 5-LOX, was unchanged in platelet-COX-1-/- mice. To better dissect the effects of aspirin in vivo , we used a thrombosis model caused by ferric chloride injury in the mouse carotid artery. Platelet COX-1-/- mice, that recapitulate the effects of low-dose aspirin, formed unstable thrombi that were embolizing before producing a stable vessel occlusion. Interestingly, when a high dose of aspirin (10 mg/kg) was injected into the mice, the thrombus was stable and formed earlier than in the vehicle-treated mice. Conclusions We have produced the first platelet-COX-1-/- mouse and characterised its phenotype and eicosanomic profile. This model is allowing us to mechanistically dissect the platelet and non-platelet effects of aspirin and so better model the effects of clinically important anti-thrombotic therapy.