Abstract
Plasma metalloprotease ADAMTS13 modulates the activity of von Willebrand factor (VWF) by cleaving the adhesive glycoprotein substrate when it is conformationally unfolded by shear stress or chaotropic agents such as urea. Without ADAMTS13, VWF may become fully unfolded by shear stress in the circulation, causing VWF-platelet binding and arteriolar/capillary platelet thrombosis characteristic of thrombotic thrombocytopenic purpura (TTP). However, patients with ADAMTS13 deficiency show considerable variability in thrombotic severity, suggesting the VWF-platelet interaction is modified by environmental and possibly genetic factors. In mouse models of targeted ADAMTS13 inactivation (KO), a thrombotic phenotype occurs in a mouse strain (CASA/Rk, thrombosis susceptible [TS] strain), either spontaneously or following shiga toxin challenge, but not in other strains (C57BL/6, 129x1/Sv & FVB/N, thrombosis resistant [TR] strains). Paradoxically SDS agarose gel electrophoresis reveals that the size of VWF multimers is larger in the plasma of both groups of mice than in normal human plasma (NHP). Furthermore, mouse VWF size is not affected by the difference in mouse ADAMTS13 activity levels, although it is slightly smaller in the TS strain than in the TR groups. ADAMTS13 KO shifts the VWF multimers to slightly larger size in the TS strain but has no discernible effect in the TR strains. We hypothesize that VWF is conformationally less flexible in the TR mice. Consequently, its size is not modulated by ADAMTS13, nor does it cause platelet thrombosis in ADAMTS13 KO mice. To determine if VWF is less flexible in the TR mice, we incubate plasma samples, which contain VWF and ADAMTS13, with varying concentrations of urea. Increased VWF fragment levels, accompanied by decreased VWF multimer sizes, are detected at 1.0 – 1.25 M urea for the NHP, at 1.75 – 2.0 M urea for the TS strain plasma, and at 2.0 – 2.25 M urea for the TR plasma. The difference is independent of the plasma ADAMTS13 activity levels. A difference in VWF cleavage is also observed when the plasma samples are exposed to shear stress in a long capillary-tubing or when VWF isolated by plasma cryoprecipitation from the various mouse strains is incubated with full length recombinant mouse (rm) ADAMTS13 in the presence of urea. In contrast, no strain difference is observed when rmVWF proteins are pretreated with urea before incubation with rmADAMTS13. To determine if mouse plasma contains activity that decreases the responsiveness of VWF cleavage to urea, we incubate VWF purified from NHP with plasma samples of the various mouse strains before urea is added to the mixtures and find that cleavage of human VWF is most responsive to urea in NHP, less responsive in the TS strain plasma, and least responsive in the TR strain plasma. The suppressive activity in mouse plasma is transmitted as a genetic trait, since intercross of the TS mice with a TR strain yields F1 mice that contain intermediate suppressive activity. Further SDS PAGE analysis reveals that the TS strains contain a 400-kD VWF fragment in the plasma instead of a 375-kD band in the TR strain plasma and a 350-kD band in NHP. Cleavage of urea-pretreated rmVWF by rmADAMTS13 generates the 375-kD band and to a lesser extent the 400-kD band. In contrast, cleavage of urea-pretreated mouse plasma VWF by rmADAMTS13 generates the band expected of the plasma donor strain, suggesting that mice contain varying levels of a modifier affecting the size of VWF fragments. Genome-wide linkage analysis of the offspring generated by intercross between the TS strain’s close relative (CAST/EiJ) and a TR strain localizes the VWF modifying trait to a 4.3-mb region on mouse chromosome 6. In summary, mouse plasma contains genetically determined activities that may decrease the conformational responsiveness of VWF cleavage to shear stress and affect the size of its cleavage fragments. Further investigation is needed to identify the genes affecting the cleavage of VWF by ADAMTS13 and how the genetic traits may modulate the thrombosis in patients with ADAMTS13 deficiency.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author