Thrombosis during atherosclerotic plaque erosion or rupture is a major cause of death and disability in individuals with cardiovascular disease. Circulating oxidized lipids are risk factors for this condition and are carried in oxidized low-density lipoprotein particles (oxLDL). CD36 is a pattern recognition receptor that recognizes oxLDL and consequently promotes a prothrombotic platelet phenotype. We previously reported that recognition of oxLDL by CD36 promotes activation of redox sensitive signaling pathways. This phenotype promotes phosphatidylserine (PSer) externalization. However, signaling events downstream of CD36 are unclear. We hypothesize that CD36 signaling through the MAP kinase redox sensor ERK5 promotes a procoagulant phenotype that supports arterial thrombosis in dyslipidemia. Platelets isolated and washed from healthy human donors were stimulated with oxLDL and were stained with fluorophore-conjugated Annexin V to detect exposure of procoagulant PSer. OxLDL-stimulated platelets showed rapid PSer externalization with maximal PSer between 5-30 min (11.0±0.35%). To mimic physiological conditions, platelets were sensitized with oxLDL before activating the collagen receptor GPVI with the snake venom convulxin (CVX). OxLDL sensitization showed marked enhancement of PSer externalization up to 78±4.6% compared to oxLDL alone (21.9±2.8%) or CVX alone (23±1.6%). To determine the mechanisms of oxLDL sensitized PSer exposure by CVX, platelets were pre-treated with the CD36 blocking antibody FA6, ERK5 inhibitor BIX02188, and inhibitors of "classic" inducers of PSer exposure, cyclosporin A to cyclophilin D and Z-VAD-FMK to caspases. Pre-treatment with FA6, BIX02188, and Z-VAD-FMK, but not cyclosporin A, prevented oxLDL-induced enhanced PSer exposure by CVX. These studies were complemented by using platelets from mice with genetic deletion of CD36, ERK5, cyclophilin D, and Bak/Bax, all of which showed a 45% reduction in PSer externalization by oxLDL compared to wild type controls. Caspase activation was determined mechanistically by immunoblotting for cleaved caspase products. OxLDL stimulated platelets showed enhanced cleaved caspase formation that was prevented to unstimulated levels by inhibiting CD36 and ERK5. PSer exposure links platelets to tissue factor-mediated fibrin formation. Kinetic analysis of fibrin formation ex vivo showed oxLDL decreased the onset time for fibrin to form by CVX from 5.6±0.1 min to 3.4±0.13 min, which was prevented by inhibiting CD36 with FA6, or ERK5 and Bak/Bax with inhibitors. CD36-mediated fibrin formation in dyslipidemia in vivo was studied using a laser-injury thrombosis model on high fat diet-fed (HFD) atherogenic apoE null mice. These mice showed rapid (within 30 sec of injury) enhancement of fibrin and platelet accumulation that was sustained over time. In contrast, HFD-fed apoE:CD36 double null mice did not show accelerated fibrin and platelet accumulation; levels were similar to that observed in control diet conditions. Furthermore, we employed a second thrombosis model by transplantation of the epigastric artery, which has a surface rich with collagen to mediate thrombosis, into the carotid artery lumen. ApoE null mice were transplanted with bone marrows from ERK5 expressing mice or platelet-specific ERK5 null mice. Compared to control diet-fed apoE null mice, fibrin accumulation was enhanced in HFD-fed apoE null mice expressing platelet ERK5. ApoE null mice lacking platelet ERK5 did not show enhanced fibrin formation; HFD was the same as control diet. In conclusion, oxLDL promotes procoagulant PSer and fibrin formation ex vivo through a CD36, ERK5, and caspase-dependent mechanism. In vivo fibrin formation in dyslipidemia were rescued by targeted genetic deletion CD36 or platelet ERK5 in two models of diet-induced arterial thrombosis. These data suggest that the CD36-ERK5-caspase signaling axis could be a potential target to prevent arterial thrombosis in dyslipidemia.


Jobe:CSL: Consultancy; Shire: Consultancy; Octapharma: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.