LETTERS TO THE EDITOR
In vivo von Willebrand factor size heterogeneity in spite of the clinical deficiency of ADAMTS-13
Article first published online: 1 DEC 2011
© 2011 International Society on Thrombosis and Haemostasis
Journal of Thrombosis and Haemostasis
Volume 9, Issue 12, pages 2506–2508, December 2011
How to Cite
DE MEYER, S. F., BUDDE, U., DECKMYN, H. and VANHOORELBEKE, K. (2011), In vivo von Willebrand factor size heterogeneity in spite of the clinical deficiency of ADAMTS-13. Journal of Thrombosis and Haemostasis, 9: 2506–2508. doi: 10.1111/j.1538-7836.2011.04519.x
- Issue published online: 1 DEC 2011
- Article first published online: 1 DEC 2011
- Accepted manuscript online: 27 SEP 2011 11:28AM EST
- Received 20 April 2011, accepted 13 September 2011
von Willebrand factor (VWF) is synthesized by endothelial cells as ultra-large (UL) multimers. The activity of VWF is strongly correlated with multimer size and UL-VWF can spontaneously bind platelet GPIb. As the circulation of UL-VWF can lead to microvascular thrombosis, these hyperactive multimers are digested upon release into smaller, less active fragments by ADAMTS-13. It is generally assumed that VWF proteolysis generates a typical pattern consisting of a complex set of repeating multimers, characterized by a more prominent band surrounded by satellite bands (‘triplets’), visible upon VWF analysis by SDS-agarose gel electrophoresis . At least in vitro, there is clear evidence that ADAMTS-13 is responsible for the generation of these ‘triplets’, also often referred to as VWF size heterogeneity [2,3]. However, whether or not ADAMTS-13 is required for the generation of in vivo VWF size heterogeneity has never been systematically studied. To clarify the association between VWF size heterogeneity and ADAMTS-13 activity in vivo, we used a sensitive VWF multimer assay to analyze multimer patterns of VWF present in plasma of ADAMTS-13-deficient human and murine plasma.
As information on the triplet structure of endothelial cell-derived VWF is limited, we first examined the multimer composition of VWF secreted by human umbilical vein endothelial cells (HUVECs) and blood outgrowth endothelial cells (BOECs) that naturally produce VWF  and from a stably transfected baby hamster kidney (BHK) cell line producing human VWF. Whereas the typical VWF size heterogeneity was clearly visible in normal human plasma (visible as satellite bands on the multimer gel and shoulders flanking the main peaks on the densitogram, Fig. 1A), no VWF size heterogeneity could be observed in VWF that was produced in vitro by BOECs (Fig. 1B), HUVECs or heterologous BHK cells (not shown). The addition of recombinant (r)ADAMTS-13 to the cell culture medium resulted in the disappearance of UL-VWF multimers and the appearance of the typical size heterogeneity (Fig. 1C). These results support the notion that the physiological process leading to VWF size heterogeneity does not occur during in vitro VWF biosynthesis but can be generated by ADAMTS-13 after release.
As endothelial cell-derived VWF is devoid of size heterogeneity in the absence of ADAMTS-13 in vitro, we assessed the size heterogeneity of VWF present in plasma of 7 congenital and 59 acquired thrombotic thrombocytopenic purpura (TTP) patients during the acute phase before patients received any treatment. None of the plasma samples had detectable (> 2%) ADAMTS-13 activity (not shown). ADAMTS-13 inhibitors were absent in patients with congenital TTP, whereas acquired TTP patients were characterized by the presence of ADAMTS-13 inhibitors (not shown). As one would expect from the lack of ADAMTS-13 activity, half of the TTP patients showed no signs of VWF size heterogeneity (representative sample shown in Fig. 1D). Interestingly, however, in 33 out of the 66 samples, we observed distinct VWF size heterogeneity (representative samples shown in Fig. 1E,F). While unfortunately not available for the present TTP samples, it would be interesting to link the exact clinical phenotype of each patient with the corresponding triplet pattern. Although unlikely, we cannot exclude that, in spite of their TTP episodes, TTP patients have residual ADAMTS-13 activity (< 2%) generating VWF size heterogeneity. We therefore analyzed VWF multimer patterns from ADAMTS-13−/− mice (ADAMTS-13B/CN2−/−). Again, as observed in normal murine plasma from wild-type mice (Fig. 1G), size heterogeneity was clearly visible on the VWF multimer patterns from these ADAMTS-13-deficient mice (representative sample shown in Fig. 1H). Recombinant murine VWF produced in vitro only displayed intact VWF multimers, without any signs of size heterogeneity (Fig. 1I). Together, these data indicate that ADAMTS-13 is sufficient but not necessary for VWF size heterogeneity and that other mechanisms besides ADAMTS-13 proteolysis might determine VWF size heterogeneity. Other proteases [6,7] and, especially leukocyte proteases [8,9], have been shown to digest VWF but the in vivo relevance for this proteolysis remains unclear. After depletion of neutrophils in ADAMTS-13−/− mice, we did not observe a change in VWF multimer composition or satellite band appearance (not shown), suggesting that circulating neutrophils do not play a major role in regulating VWF size distribution, at least under normal, non-activating conditions. These results are in accordance with the normal VWF size distribution observed in neutropenic patients . Whether proteases from activated neutrophils could play a physiological role in VWF cleavage remains to be determined. An attempt to identify the origin of the triplet bands in murine (m)VWF of ADAMTS-13-deficient mice failed as a consequence of the relatively low recovery of VWF side bands from plasma, hampering mass spectrometry sequence analysis (not shown).
In conclusion, using a multimer assay sensitive for VWF size heterogeneity, we demonstrated that clinical deficiency of ADAMTS-13 does not necessarily abrogate the presence of VWF satellite bands. These findings point out that the in vivo relationship between VWF size heterogeneity or ‘triplets’ and ADAMTS-13 activity is not as straightforward as generally assumed, which is a critical piece of information for the correct understanding and interpretation of VWF processing in vivo. Our data, therefore, call for prudence when interpreting VWF multimer size heterogeneity and encourage the search for other relevant in vivo post-translational elements that may contribute to the heterogeneous appearance of VWF multimers.
We thank I. Pareyn and N. Vandeputte for their excellent technical assistance and K. Will and A. Pieconka for running multimer gels. S.F. De Meyer is a postdoctoral fellow of the FWO (Fonds voor Wetenschappelijk Onderzoek) Vlaanderen, Belgium. This work was supported by a grant from the FWO (G.0299.06) and from the KU Leuven (GOA/2004/09).
Disclosure of Conflict of Interests
The authors state that they have no conflict of interest.