ADAMTS13's tail tale

In mice, a long form and a short form of the VWF-cleaving protease ADAMTS13 have been identified, the latter lacking the 4 distal carboxyl-terminal domains. While these are not strictly required for regulating normal size distribution of VWF multimers, in this issue of Blood, Banno and colleagues reveal the role of these domains in down-regulating thrombogenesis in vivo.

Since the discovery of ADAMTS13 as a metalloprotease with a multi-domain structure, numerous studies have attempted to shed light on the specific roles of each of the ADAMTS13 domains in digesting large von Willebrand factor (VWF) multimers into smaller, less reactive ones. ADAMTS13 is composed of a signal peptide, propeptide, metalloprotease domain, central TSR (thrombospondin type 1 repeat), Cys-rich region, spacer domain, 7 additional TSRs, and 2 CUB domains. The active site of this enzyme is situated in the metalloprotease domain while the spacer domain plays a crucial role in substrate binding by interacting with a VWF exosite located at the C-terminus of the A2 domain. The exact physiologic significance of the carboxyl-terminal TSRs and the 2 CUB domains still remains unclear, in particular due to the use of different types of in vitro tests, often performed under nonphysiological conditions.

To unravel the in vivo role of the carboxyl-terminal domains of ADAMTS13, Banno and coworkers elegantly take advantage of the presence of 2 kinds of Adamts13 genes in laboratory mouse strains.¹  The 129/Sv strain has the Adamts13 gene encoding full-length ADAMTS13 while several other strains, including C57BL/6, harbor an Adamts13 gene that expresses a truncated form of the enzyme, lacking the 2 C-terminal TSRs and CUB domains due to the insertion of an intracisternal A-particle retrotransposon. By introgressing the C57BL/6-Adamts13 gene onto the 129/Sv genetic background, the authors generate congenic mice that had the distal C-terminally truncated ADAMTS13 on a 129/Sv genetic background (Adamts13^S/S) and use wild-type mice that have full-length ADAMTS13 (Adamts13^L/L) and ADAMTS13^−/− mice on the same 129/Sv genetic background for comparison.

The most obvious role of ADAMTS13 is to regulate VWF multimer size. Indeed, ADAMTS13 digests unusually large VWF multimers into smaller less thrombogenic forms,²  hence preventing the spontaneous intravascular platelet aggregation seen in patients with ADAMTS13 deficiency. Interestingly, Banno et al showed that both Adamts13^L/L and Adamts13^S/S mice do not have ultra large VWF multimers in their plasma, implying that the C-terminal domains are not strictly needed for maintaining normal VWF size. Consequently, the 2 C-terminal TSRS and CUB domains are not essential for the removal of ultralarge VWF multimers from the plasma.

Following VWF size regulation, a fascinating role of ADAMTS13 in attenuating thrombus growth has been described, possibly by cleaving VWF multimers that are peripheral to or incorporated in platelet rich thrombi.³  In this study, Banno et al used the congenic mice to demonstrate that the 2 C-terminal TSRs and CUB domains play a role in the down-regulation of thrombogenesis under high shear conditions. Both in vitro flow chamber experiments at high shear rates and in vivo thrombosis models show that blood from Adamts13^S/S mice is more thrombogenic. This is evidenced by accelerated thrombus formation and decreased time to occlusion respectively when compared with blood from Adamts13^L/L mice. Whether this would translate into an increased risk of thrombosis in patients having comparable truncated forms of ADAMTS13 remains elusive.

In this article, Banno et al provide the first in vivo insights on the physiological significance of the distal carboxyl-terminal domains of ADAMTS13. The exact mechanism of thrombus size attenuation by ADAMTS13 and, in particular, the specific involvement of the carboxyl-terminal domains still remains to be determined. Does ADAMTS13 digest VWF multimers on the surface of the platelet thrombus or is thrombus size attenuation by ADAMTS13 independent of its VWF-cleaving activity? In this context, it is certainly intriguing that the mechanism of VWF size regulation by ADAMTS13 might be different from that of VWF processing during thrombus growth. Clearly, these new findings provide another impetus in the quest to understand the structure-function relationship of ADAMTS13. Obviously, this is not the end of the tale.