The allosteric activation of ADAMTS13

Abstract

ADAMTS13 regulates the multimeric size of Von Willebrand Factor (VWF) in plasma by specific, shear-dependent proteolysis. Despite circulating in a seemingly proteolytically competent form, and with a very long active plasma half-life, ADAMTS13 is specific for VWF, and is resistant to plasma inhibitors. This unprecedented specificity has been attributed to exosite interactions between the ADAMTS13 Spacer, Cysteine-rich, Disintegrin-like and Metalloprotease domains and the unraveled VWF A2 domain, which serve to direct proteolysis. I hypothesized that one or more exosite interactions allosterically activate ADAMTS13 by modulating the conformation of its active site, specifically shaping it to accommodate the VWF scissile bond. My first aim was to characterise the independent contribution of each ADAMTS13 exosite to VWF binding and proteolysis. I generated a new VWF A2 domain fragment (VWF96) as a substrate for ADAMTS13, with/without mutations that ablate each exosite interaction separately. I developed a novel ELISA that specifically detects VWF96, and optimized protocols for monitoring the kinetics of proteolysis. My data show that disruption of the Spacer and Cysteine-rich domain exosite interactions reduces proteolysis by ~15-to-20-fold. These deficits were due to reductions in VWF binding. Intriguingly, a remarkable role in proteolysis was revealed for the Disintegrin-like domain exosite interaction; when ablated, this exosite caused a ~780-fold reduction in proteolysis, due to a ~15-fold reduction in binding (Km), and a ~56-fold reduction in substrate turnover (kcat), consistent with the existence of an allosteric link between the Disintegrin-like domain exosite and the active site in the Metalloprotease domain. ADAMTS13 was previously shown to adopt a folded conformation, maintained by inter-domain interactions. My second aim was to characterise the mechanism by which anti-Spacer/anti-CUB domain monoclonal antibodies unfold ADAMTS13, enhancing proteolysis. I showed that the antibody-induced unfolding primarily enhances the substrate turnover (kcat), rather than substrate binding (Km), which would be consistent with an allosteric link between the non-catalytic domains and the active site. Taken together, these findings provide insights into the molecular mechanisms by which the non-catalytic domains of ADAMTS13 influence the active site and confer specificity to VWF proteolysis, and highlight the importance of conformational changes in ADAMTS13 to its enzymatic function.