PB0385 - GPIbα shedding is restricted to the inner platelet membrane

Background: GPIbα is an abundant transmembrane receptor on the platelet cytoplasmic membrane. Following strong and sustained platelet activation, ADAM17 catalyzes GPIbα proteolysis thereby releasing the soluble glycocalicin ectodomain.

Aims: We aimed to understand the enzyme-substrate relationship in space and time.

Methods: ADAM17 and GPIbα were detected using labeled monoclonal antibodies in human resting platelets or after activation via GPVI and PARs, using immunofluorescence and western blot analysis.

Results: To chase glycocalicin, resting platelets were preincubated with anti-GPIbα monoclonal antibody before activation. After activation, the majority of antibody was detected in the platelet pellet with only 11.2±4.0% released in supernatant despite 82.1±4.9% of GPIbα effectively shed. Furthermore, preincubation of resting platelets with anti-GPIbα antibody 5G6, targeting the ADAM17 cleavage site, did not inhibit glycocalicin release (93±6% relative to control without 5G6). When 5G6 was present during activation, glycocalicin release was partially inhibited (59±11% relative to control, p=0.02; n=5). Finally, pretreatment of platelets with the membrane impermeable O-sialoglycoprotein endopeptidase removed a 45kDa N-terminal fragment of surface-exposed GPIbα but had no effect on the molecular weight or amount of glycocalicin released in platelet supernatant following activation-dependent shedding. Statistically, >80% of GPIbα was an ADAM17 substrate, while < 2% of released GPIbα was surface-exposed. This agrees with our transmission electron microscopy findings using anti-ADAM17 immunogold labeling, showing ADAM17 resides almost exclusively (96±2% of label; n≥25 platelets) within the inner membranes of both resting and activated platelets. This intracellular detection was confirmed by immunofluorescence microscopy and flow cytometry where successful ADAM17-antigen detection required permeabilization with saponin. In addition, only membrane-permeable sheddase inhibitors marimastat, GM6001 and KP-457 significantly inhibited GPIbα shedding (88±5%, 88±4% and 95±1%, respectively). In contrast, membrane-impermeable ADAM17 inhibitors D1A12 (11±11%) and recombinant prodomain (7±10%; ≥6) could not inhibit metalloproteolysis of GPIbα.

Conclusion(s): The majority of GPIbα is a substrate for ADAM17, but surface-exposed GPIbα is not.