Endothelial cell (EC) derived microparticles (MP) have been identified in the circulation of patients with arterial prothrombotic tendencies such as diabetes, anti-phospholipid syndrome and acute coronary syndrome. CD36 is a Type B scavenger receptor that is expressed on many cells including megakaryocytes and platelets, where it was first identified as glycoprotein IV. It recognizes several classes of ligand including thrombospondins, oxidized LDL, and apoptotic cells, and has been shown to play a role in phagocytosis, angiogenesis and atherosclerosis. The function of CD36 on platelets is incompletely characterized, but our group has recently shown that oxLDL binds to and activates platelets in a CD36-dependent manner. Because MP express phosphatidyl serine (PS) on their surfaces, a potential CD36 ligand, we hypothesize that MP may bind to platelets via a PS-CD36 interaction and transmit an activating signal, thereby promoting a prothrombotic state. To test this hypothesis, we isolated MP from TNFα/cyclohexamide-treated human umbilical vein EC using an established protocol. MP were characterized and quantified by flow cytometry and shown to express CD105 and PS. Washed human platelets (CD105 negative) were incubated with EC-derived MP at a ratio of 1:9 and analyzed by flow cytometry with a fluorescence tagged anti-CD105 antibody. We observed a significant increase in platelet associated CD105 fluorescence in the presence of MP suggesting that CD105 positive MP were physically associated with the platelets. This physical association was confirmed by using two color fluorescence microscopy showing that PKH26 (red) tagged MP formed rosettes around Calcein-Green tagged platelets. With both the flow cytometry and microscopy assays, platelet-MP association was inhibited nearly completely by addition of anti-CD36 antibody (p=0.02) or by using platelets from CD36 null donors. Pre-incubation of MP with annexin V to block exposed PS also significantly inhibited platelet-MP interaction. Furthermore, pretreatment of platelets with other CD36 ligands such as oxLDL inhibited MP-platelet association by more than 50%. We determined the functional impact of the MP-platelet association, observing a significant increase in rate and extent of platelet aggregation to low concentrations (2μM) of ADP and an increase in surface P-selectin expression when platelets were incubated with EC-derived MP prior to addition of agonist. This effect was markedly diminished in platelets from CD36 null donors and also inhibited by pre-incubation with anti CD36 antibody. To study the MP-platelet interaction in vivo, carotid arteries of CD36 wild type and knock out mice were injured by FeCl3 and the thrombosed arteries were sectioned and immunostained with anti-CD105 (red) and anti-CD61 (green) antibodies to detect MP and platelets. We observed robust red staining of the thrombi in wild type mice with substantial co-localization of red with green. Thrombi from CD36 knock out mice showed fewer MP and less co-localization of anti-CD105 with anti-CD61. We also demonstrated that MP isolated from the blood of patients with antiphospholipid syndrome augmented platelet aggregation in a CD36 dependent manner. We propose a model whereby CD36 ligands presented to platelets initiates a signaling cascade that renders them more sensitive to a second “hit” and thus predisposes to formation of pathological thrombosis.

Author notes

Disclosure: No relevant conflicts of interest to declare.