Abstract 3194

In the blood, platelets are normally prevented from activation by endothelial inhibitors (i.e. prostacycline, ectonucleotidase). Dysfunctional endothelial cells loose their protective properties and favor platelet adhesion to matrix proteins, platelet aggregation and thrombus growth. Collagen fibers are highly thrombogenic and the platelet Glycoprotein (GP)VI predominantly mediates collagen-induced platelet responses. GPVI is a platelet specific receptor of the immunoglobulin (Ig) superfamily containing two extracellular Ig domains, a single transmembrane domain and a short cytoplasmic tail. GPVI signals through the immunoreceptor tyrosine-based activation motifs (ITAM) of the non-covalently associated immune receptor adaptor FcRg dimer. There is growing evidence that optimal binding of GPVI to collagen depends on the formation of GPVI dimers at the platelet surface: only dimeric GPVI binds to collagen and inhibits collagen-induced platelet aggregation and not monomeric GPVI. Moreover, crystallographic data showed dimerization of GPVI ectodomains. However, the valence of GPVI on resting and activated platelets is still debated.

We have obtained an anti-human GPVI monoclonal antibody (9E18), that binds to dimeric GPVI with a 200 fold higher affinity than to monomeric GPVI. In flow cytometry on whole blood, while the 3J24 antibody labels >95% platelets, 9E18 hardly binds to resting platelets with less than 3% positive platelets. The level of 9E18-positive platelets moderately increased (10-15%) after platelet isolation suggesting it could reflect platelet activation. Binding of 9E18 was indeed significantly increased on ADP- or TRAP-activated washed platelets (25±1.9 % and 36±7% positive platelets respectively). Additionally, increased binding of 9E18 was triggered by the GPVI agonists, collagen, convulxin or the activating 9O12 IgG. At sites of vascular lesion, platelet adhesion is initiated by the shear-dependent interaction of GPIb with vWF, assumed to favor GPVI-collagen interaction. When a platelet rich plasma was submitted to a shear of 4000 s-1 for 5 min, 9E18-positive platelets increased from 3.6±1.6% to 7±2% in the whole platelet population and to 26±7.7% on small aggregates (p<0.05).When a2b1 and aIIbb3 were blocked, the relation between the 9E18 binding to stimulated platelets and platelet binding to collagen was linear (r2 = 0.847, p=0.0012, n=8). Interestingly, the cAMP elevating agent PGE1 further lowered the level of 9E18-binding to resting platelets and dropped it to basal values on ADP- or TRAP-treated platelets. Apyrase reduced by 50% TRAP-induced binding of 9E18 whereas indomethacin had no effect. PMA triggered binding of 9E18 on platelets (p<0.001) while the Tyr-phosphatase inhibitor PAO, strongly inhibited PMA-induced 9E18 binding to platelets (p<0.0019) and GPVI-dependent platelet adhesion to collagen.

Altogether, these data indicate that 9E18 permit to quantify GPVI dimers on platelets. They show that (i) GPVI is mainly monomeric on resting platelets, (ii) dimerisation is an active process triggered by shear, soluble agonists and matrix proteins, (iii) the level of GPVI dimers is related to the capacity of platelets to adhere to collagen, (iv) GPVI dimerisation is completely prevented in the presence of agents increasing cAMP or by PAO. These data suggested that the formation of GPVI dimer is strictly controlled on resting platelets and that GPVI dimers could thus represent a new marker of platelet activation and susceptibility to collagen. Indeed, in a population of hospitalized patient, a positive correlation was observed between 9E18 binding and P-selectin exposure on platelets.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution