Species-Dependent Variability of ADAMTS-13 Mediated Proteolysis of Human Recombinant von Willebrand Factor.

Abstract

ADAMTS-13 is a plasma metalloprotease that regulates platelet adhesion and aggregation through cleavage of von Willebrand factor (VWF) multimers within the A2 domain between residues tyrosine 1605 and methionine 1606. Genetic (Upshaw-Schulman syndrome) and acquired severe functional defects of ADAMTS-13 are associated with the presence of non-proteolysed ultra-large VWF multimers. These multimers are the most thrombogenic form of this molecule and spontaneously bind to platelets causing thrombotic thrombocytopenic purpura (TTP), a syndrome characterized by microvascular thrombosis and platelet consumption. We investigated proteolysis of human rVWF by ADAMTS-13 present in plasma of different species typically used as preclinical animal models. Our substrate was a recombinant, Chinese hamster ovary (CHO) cell line-derived rVWF, processed to mature rVWF by exposure to rfurin and highly purified by a series of chromatography steps. Multimer analysis of rVWF shows absence of satellite bands consistent with intact subunits not cleaved by ADAMTS-13. rVWF was incubated with BaCl2-activated plasma samples from different species, 4 hours at 37°C with urea. ADAMTS-13-mediated degradation of VWF was measured by collagen binding (CB) assay and proteolysis visualized by high-resolution multimer analysis stained with a polyclonal anti-human VWF antibody and by non-reducing SDS-PAGE, followed by immunoblotting with a polyclonal anti-human VWF antibody. Incubation of rVWF with normal human plasma resulted in the loss of collagen binding activity and disappearance of high molecular weight multimers with simultaneous formation of the specific satellite bands leading to a multimeric pattern similar to pdVWF. In contrast, using the same amount of animal plasma (50 mU/ml human-equivalent concentration), differences were found in the ADAMTS-13 activity levels in terms of the degree of proteolytic degradation of rVWF multimers as follows as measured by CBA: Mouse plasma samples from wild type or from mice lacking FVIII and VWF, all with a C57BL/6J background, showed no ADAMTS-13 activity towards human rVWF. No activities were observed in rat and guinea pig plasma. Porcine plasma showed only minimal activity. Samples from monkey strains (baboon, rhesus and cynomolgus) and from beagle dogs showed medium activities with a residual CB of 60%. Rabbit plasma was as effective as human plasma in cleaving and degrading VWF resulting in a residual CB of 20%. The quantitative results were confirmed by multimer visualization. We conclude that animal plasma differs markedly in plasma ADAMTS-13 activity and capacity to cleave human VWF multimers. Our study confirms previous in vivo studies with human rVWF where no proteolysis was seen in mice, and little proteolysis was observed in dogs (Turecek et al., Blood 1997) and pigs (Roussi J et al. Blood.Coag.Fibrinol. 1998). The present study also suggests caution is needed in interpreting the (patho)physiology and metabolism of human VWF from animal models.