A Disintegrin And Metalloprotease with ThromboSpondin type 1 repeats (ADAMTS)-13 cleaves newly released ultra large (UL) von Willebrand factor (VWF), thereby inhibiting excessive platelet aggregation and thrombus formation. Inability to cleave ULVWF due to deficiency of plasma ADAMTS13 activity may result in thrombotic thrombocytopenic purpura (TTP), a potentially fatal illness, and other arterial thrombotic diseases (i.e. myocardial and cerebral infarctions). However, little is known about the structural components of ADAMTS13 required for systemic anti-arterial thromboses in vivo. In this study, we determined the biological effect of ADAMTS13 and variants on arterial thromboses in a murine model using two different assays. First, using a ferric chloride-induced carotid arterial occlusion assay, we demonstrated that the times to a complete occlusion (TCO) of carotid artery after topical application of 10% ferric chloride (soaked in a filter paper 1×2 min, for 2 min) in wild-type (C57BL/6) mice and Adamts13-/- mice were 10.0 ± 1.0 min (mean ± SEM) (n=9) and 5.3 ± 0.4 min (n=10), respectively. The difference was statistically highly significant (p<0.0001). An infusion of 10 nM of recombinant human full-length ADATMS13 (FL), a variant truncated after the 8th TSP1 repeat (T8) and after the spacer domain (S) into the Adamts13-/- mice restored the TCO to 12.7 ± 1.7 min (n=12), 8.0 ± 1.8 (n=7), and 22.0 ± 2.1 min (n=12), respectively. These results suggest that the N-terminal fragment of ADAMTS13 up to the spacer domain is sufficient for protection against ferric chloride induced arterial thrombosis. Moreover, an infusion of an ADAMTS13 mutant lacking 6 amino acid residues between Arg659 and Glu664 (d6a) into Adamts13-/-mice did not restore the TCO (5.9 ± 0.6 min, n=11), suggesting the critical role of the spacer domain in anti-arterial thrombosis in vivo. Paradoxically, however, an infusion of a recombinant C-terminal fragment of ADAMTS13 consisting of the TSP1 5–8 repeats and CUB domains (T5C) at the final concentration of 50 nM (10x endogenous murine plasma ADAMTS13 concentration) into wild-type mice significantly shortened the TCO (5.9 ± 1.9 min) (n=11), similar to that in the Adamts13-/- mice receiving injection of PBS alone (p=0.440). These results indicate that the middle and distal C-terminal domains of ADAMTS13 in the context of the whole molecule may also participate in substrate recognition and are required for anti-arterial thrombotic function in vivo. Second, the results from intravital microscopy by visualizing a real-time thrombus formation in the mesenteric arterioles nearly mirrored the data obtained by the carotid arterial occlusion assay. In these experiments, the times to an initial thrombus formation (Ti) (defined as the time to form a thrombus >30 μ m) and the times to a complete occlusion of blood vessel (Tc) were determined. We showed that the Ti and Tc in wild-type C57BL6 mice were 8.8 ± 0.6 min (mean ± SEM) and 13.7 ± 1.0 min (n=12), respectively. The Ti and Tc in the Adamts13-/- mice (same genetic background) were 5.2 ± 0.6 min (n=15) and 9.6 ± 0.8 min (n=15), respectively. The differences in both Ti and Tc between wild-type mice and Adamts13-/- mice were statistically highly significant (p<0.001). An infusion of recombinant human FL, T8, and S, but not d6a at a final concentration of 10 nM significantly prolonged the Ti (FL: 9.7 ± 0.9 min, n=10; T8: 10.1 ± 1.5 min, n=10; S: 11.5 ± 1.5 min, n=10; d6a: 5.3 ± 0.57 min, n=11) and the Tc (FL: 15.3 ± 1.4 min; T8: 21.9 ± 2.2 min; S: 16.2 ± 1.6 min; d6a:10.0 ± 1.1 min). The differences in both Ti and Tc between the control (buffer injected) and experimental groups were all statistically highly significant. We conclude that multiple C-terminal domains of ADAMTS13 are required for systemic anti-arterial/arteriolar thrombosis under (patho) physiological conditions. Our findings may shed more light on pathogenesis of TTP and provide molecular basis for a rational design of novel therapies for TTP and perhaps other arterial thrombotic disorders because of dysfunction of VWF/ADAMTS13 axis.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.