Abstract
Anti-platelet integrin GPIIIa49-66 Ab obtained from HIV-ITP patients (or raised in rabbits) induces complement-independent platelet oxidative fragmentation and death by activating platelet 12-lipoxygenase (generation of 12(S)-HETE) and NADPH oxidase (JCI, 113:973, 2004). Platelet oxidative fragmentation is measured by flow cytometry of generated microparticles as well as intracellular DCFH oxidation. We now report that oxidative fragmentation in human platelets is preceded by Ca++ flux and P-selectin activation, n=6. However, the activation mechanism is different from classic platelet activation in that it is not inhibited by PGE1 or dibutryl cyclic AMP and is operative with Gαq−/− mouse platelets, whereas under these conditions, thrombin-induced platelet activation is completely inhibited, n=5–6. We chose to identify putative physiologic ligands that behave similarly to the GPIIIa49-66 Ab, and are therefore capable of regulating platelet reactive oxygen species (ROS) as well as arterial thrombus formation. The GPIIIa49-66 platelet peptide was used as bait to screen a 7-mer peptide phage display library. A peptide was found with 70% homology at the C-terminal position of ADAMTS-18, an ‘orphan’ disintegrin and metalloproteinase with thrombospondin (TSR)-like motifs, with unknown substrate. We have found it present in HUVEC as well as human pulmonary artery endothelial cells, on fixed sections of pathology specimens employing immunohistochemistry with a specific rabbit Ab raised against a C-terminal 18 mer peptide ADAMTS-18 (no staining with preimmune Ab). Recombinant ADAMTS-18 was produced in HEK 293 T cells and shown to induce ROS and oxidative platelet fragmentation in an identical kinetic fashion as anti-GPIIIa49-66 Ab. HUVEC ADAMTS-18 activity could be inhibited by a human scFv Ab raised against its C-terminal 18 mer peptide, as well as the ADAMTS-18 peptide itself, but not by a rabbit Ab against the N-terminal domain or an irrelevant peptide. Endothelial cell secretion and activation of ADAMTS-18 was optimally induced with 0.5 u/ml thrombin at 2 – 4 hrs, n=3–4. The truncated 385 amino acid C-terminal rADAMTS-18 fragment containing the 4 TSR motifs (produced in E.coli) had full activity at (<0.3 uM) whereas the C-terminal 66 amino acid fragment not containing the 18-mer binding site was inactive at 65 fold higher concentration, n=4. The physiologic significance of ADAMTS-18 was supported by demonstrating its secretion into plasma following iv injection of 4–16 u/ml thrombin into mice. Wild type mice have no detectable ADAMTS-18 in their plasma, with a sensitive ELISA assay (1 ng detectability). Thrombin stimulated mice secrete ADAMTS-18 in a concentration dependent manner. Platelet aggregates produced ex vivo with ADP and fibrinogen were destroyed with ADAMTS-18 as documented by LDH release at 1, 2 and 4 hrs of 83, 241 and 260 fold respectively, of PBS buffer control. In vivo tail vein bleeding time was shortened 4.5 fold with 1 hr prior infusion of 25 ug of a polyclonal rabbit IgG against ADAMTS-18, but not with preimmune IgG, n=10. Thus, a new mechanism is proposed for platelet activation, ROS release, death and platelet thrombus regulation, via platelet membrane oxidative fragmentation induced by thrombin-induced secretion and activation of ADAMTS-18.
Circulating angiopoietin-1 could be confounded by ex vivo platelet activation