Grand Rounds: “A New Mechanism of Platelet Thrombus Regulation Induced by ADAMTS-18: Clinical Implications”

November 6, 2008

Bellevue Amphitheater

Commentary by Yelena Kopyltsova MD, PGY-2

Please also see the clinical vignette presented before last week’s grand rounds.

Last week’s grand rounds speaker was Simon Karpatkin MD, NYU Professor of Medicine & Director of the NYU Division of Hematology. In the 80s, multiple cases of autoimmune thrombocytopenic purpura were observed in patients with AIDS.  It was noted that these patients had elevated anti-platelet IgG levels (extracted from immune complexes), which correlated strongly with low platelets.  Dr. Karpatkin and his colleagues became interested in elucidating the mechanism of this autoimmune process, as it was suggestive of new immune complex mediated reaction.


They found that antibody to the platelet β3 Integrin peptide GPIIIa49-66 might be implicated, as the antibody titers correlated inversely with platelet counts.  In animal models, severe ITP was induced in mice exposed to this antibody.  It also appeared to be a complement independent process, as complement deficient mice still became thrombocytopenic upon exposure to antibody against GPIIIa49-66.

The next cornerstone was realization that oxidation was involved in the process.  Through a series of immunoblots and flow cytometry experiments, the NADPH oxidase pathway was implicated.  It is known that leukotrienes, produced by a 5-lipoxygenase in granulocytes, can act as an oxidizing agent.  Platelets are unique in that they have a 12-lipooxygenase (12-LO) pathway.  This was shown to be necessary to activate the platelet NADPH oxidase.  For example, in mice deficient in proteins involved in the NADPH oxidase pathway, including gp91, Rac-2, 12-LO, platelet destruction was not observed, but it was induced in wild type mice exposed to their Ab.

Experiments involving phage libraries showed that antibody to GPIII49-66 bears homology to ADAMTS 18 (AD-18), which belongs to a family of ADAMTS, proteins involved in sculpting the subendothelial endothelial matrix as well as other functions.  Immunohistochemistry revealed that AD-18 is found in human umbilical vein endothelial cells (HUVEC).  Thrombin cleaves the C-terminal end of AD-18 releasing a platelet reactive C-terminal 45KD fragment.  It binds to platelets in a saturation-dependent manner. In fact, both the AD-18 fragment and HIV-ITP GPIII antibody bind to the same GPIIIa49-66 epitope on the platelets to initiate oxidative destruction.  When HUVEC cells were exposed to thrombin, as expected, there was increased platelet fragmentation capacity.  However, cells pretreated with hirudin + thrombin, remained intact.

Next, using a human library to generate high affinity monoclonal antibodies, A11 was created, which is essentially a monoclonal antibody against GPIII49-66.  So how is this clinically relevant?  In the next series of experiments, thrombosis of the carotid artery was induced in mice.  Then either ADAMTS 18 or A11 were added to prevent thrombosis.  In a carotid artery in which blood flow has been occluded for two hours, addition of A11 opened up the artery in a few minutes.  Other studies were performed following post-ischemic cerebral infarction in mice.  This showed ~50% improvement in cerebral infarction. There was no difference in the size of post ischemic infarction, whether A11 was added 2 hours before or after arterial occlusion.  In fact, at 8 days after the stroke, mice exposed to A11 showed protection from cerebral damage with both functional and postural reflexes.

In summary, endothelial injury leads to platelet adhesion and accretion.  Platelets also activate thrombin generation.  In turn, thrombin activates endothelium, and release a C-terminal ADAMTS 18 fragment.  Next, activated C-terminal ADAMTS-18 binds/clusters platelet GPIIIa49-66 which activates the platelet NADPH oxidation pathway, leading to platelet/thrombus oxidative fragmentation and clearance.

These findings are important, as now we have potential drug targets useful in dissolution of arterial thrombi.

Leave a Reply

Your email address will not be published. Required fields are marked *