Evidence that GPVI is expressed as a mixture of monomers and dimers, and that the D2 domain is not essential for GPVI activation: GPVI dimerisation is not critical for ligand binding

Collagen has been proposed to bind to a unique epitope in dimeric GPVI and the number of GPVI dimers has been reported to increase upon platelet activation. However, in contrast, the crystal structure of GPVI in complex with collagen-related peptide (CRP) showed binding distinct from the site of dimerisation. Further fibrinogen has been reported to bind to monomeric but not dimeric GPVI. In the present study we have used the advanced fluorescence microscopy techniques of single molecule microscopy, fluorescence correlation spectroscopy (FCS) and bioluminescence resonance energy transfer (BRET), and mutagenesis studies in a transfected cell line model to show that GPVI is expressed as a mixture of monomers and dimers, and that dimerisation through the D2 domain is not critical for activation. As many of these techniques cannot be applied to platelets to resolve this issue, due to the high density of GPVI and its anucleate nature, we used Förster resonance energy transfer (FRET) to show that endogenous GPVI is at least partially expressed as a dimer on resting and activated platelet membranes. We propose that GPVI may be expressed as a monomer on the cell surface and forms dimers in the membrane through diffusion giving rise to a mixture of monomers and dimers. We speculate that the formation of dimers facilitates ligand binding through avidity.