Characterization of the patient's GPVI molecular defect. (A) Flow cytometry on whole blood using the monoclonal antibody 3J24.2 showed decreased GPVI expression on the patient's platelets, compared with control. (B) Serial quantities of proteins in sodium dodecyl sulfate–platelet lysate from the patient or a control were immunoblotted with a human polyclonal anti-GPVI (hpAb). GPVI of the control migrated as a single 58-kDa band, whereas the patient's GPVI migrated as a smear and presented a partial deficiency (note that the 10-μg control GPVI band is saturated; the 35-kDa band is nonspecific). When 10 μg control and patient platelet proteins were immunoblotted with monoclonal antibody 3J24.2 or 9O12.2, the patient's GPVI was hardly detectable. (C) DNA sequence showing the single nucleotide substitution in exon 3 (C172T), resulting in the Arg38 substitution in Cys. (D) Purified recombinant wild-type (WT) and R38C GPVI-Fc were analyzed on sodium dodecyl sulfate–8% acrylamide gels and Coomassie blue staining. WT GPVI-Fc migrated as an approximately 220-kDa band and R38C GPVI-Fc as a smear in nonreducing conditions (NR), and both migrated as a major 95-kDa band after reduction with 5% β-mercaptoethanol (R). (E) Binding of WT and R38C GPVI-Fc to immobilized collagen. Results represent mean ± SD of 2 experiments performed in triplicate. The K_d values are 1.67 ± 0.26 μg/mL and 1.4 ± 0.17 μg/mL for R38C and WT GPVI-Fc, respectively. (F) Three-dimensional structure of the extracellular domain of GPVI with localization of R38. R38 and other important residues for collagen binding, such as K59, K41, and R166,⁶  are localized on the extracellular domain of GPVI as crystallized by Horii et al.⁸  The 2 disulfide bridges are represented and encircled in black.