Ann Hematol (2002) 81:430–435 DOI 10.1007/s00277-002-0502-3

R A P I D C O M M U N I C AT I O N

M. Böhm · T. Vigh · I. Scharrer

Evaluation and clinical application of a new method for measuring activity of von Willebrand factor-cleaving metalloprotease (ADAMTS13)

Received: 26 March 2002 / Accepted: 26 June 2002 / Published online: 14 August 2002 © Springer-Verlag 2002

Abstract Thrombotic thrombocytopenic purpura (TTP) is associated with acquired or congenital deficiency of a plasma von Willebrand factor-cleaving protease (VWFcp). Based on partial amino acid sequence and genomewide linkage analysis of pedigrees with congenital TTP, VWFcp was recently identified as a new member of the ADAMTS family and designated ADAMTS13. We developed a new, rapid, and simple method for measuring VWFcp activity based on the positive correlation between VWF multimeric size and Ristocetin cofactor activity (VWF:RCo). After dilution of plasma with low ionic Tris buffer and activation of the protease with barium chloride, a VWF concentrate is digested in the presence of urea. Subsequently, the residual VWF:RCo of the samples is assessed and used to calculate the VWFcp activity of the samples. The accuracy of the new technique is verified by estimating VWFcp activity for 282 plasma samples with the RCo-based assay and the original immunoblotting assay. The method is reproducible as shown by low intraand interassay coefficients of variation (2.8% and 7.5% for normal samples, respectively, and 8.7% and 12.9% for abnormal samples, respectively). Furthermore, the clinical application of the new method is illustrated by measuring VWFcp of 14 patients with 22 episodes of acute TTP as well as other thrombotic, thrombocytopenic, or hemolytic disorders. Severe VWFcp deficiency was restricted to patients with acute, classic TTP. The majority of patients with low titer inhibitor respond to plasma exchange treatment with increase of VWFcp activity, whereas VWFcp deficiency persists in some patients with high titer inhibitor despite clinical remission. Keywords von Willebrand factor-cleaving protease · Thrombotic thrombocytopenic purpura · Ristocetin cofactor activity · ADAMTS13 M. Böhm · T. Vigh · I. Scharrer (✉) Department of Hemostaseology, Johann Wolfgang Goethe University, Theodor Stern Kai 7, 60590 Frankfurt am Main, Germany e-mail: [email protected] Tel.: +49-69-63015051, Fax: +49-69-63016738

Introduction Thrombotic thrombocytopenic purpura (TTP) was first described by Moschcowitz [26] and is characterized by the classic pentad of thrombocytopenia, microangiopathic hemolytic anemia, neurological signs, renal impairment, and fever. Unusually large molecular forms of von Willebrand factor (VWF) were found in TTP patients and were proposed to have a pathogenic role in the formation of microvascular VWF- and platelet-rich thrombi, which are found in the microcirculation of patients with acute TTP [24, 26]. VWF is released from endothelial cells as large multimers, which are cleaved by a combination of reductase and metalloprotease in normal plasma [10, 29]. Several investigators reported deficiency of a specific metalloprotease, which cleaves VWF at the Tyr842-Met843 peptide bond (VWFcp) in patients with congenital or acquired TTP [1, 3, 11, 12, 15, 32, 33, 35]. The enzyme was recently purified and identified as a new member of the ADAMTS (a disintegrin and metalloproteinase with thrombospondin type I motif) family of metalloproteinases and designated ADAMTS13 [8, 13]. Contemporaneously, Levy et al. [21] performed genome-wide linkage analysis in four pedigrees of humans with congenital TTP and mapped the responsible genetic locus to ADAMTS13 on chromosome 9q34. The VWF-cleaving protease activity of ADAMTS13 is usually measured by incubation of a VWF substrate with diluted plasma samples at low ionic strength in the presence of divalent metal ions and urea or guanidium hydrochloride. Detection of proteolysis is done by multimeric analysis with sodium dodecyl sulfate (SDS) agarose gel electrophoresis or fragment analysis with SDS polyacrylamide gel electrophoresis (PAGE) and subsequent immunoblotting [10, 29]. Degradation of VWF can also be estimated by measuring the collagenbinding activity of the sample [14] or by specific enzyme-linked immunosorbent assay (ELISA) techniques [16, 27]. Recently, a recombinant, truncated, monomeric VWF, which is labeled with green florescent protein on the N-terminus, has been created for determination of

431

proteolysis [28]. Moreover, two methods using the endogenous VWF as a substrate have been published [2, 19]. Recent findings of low VWFcp activities in diseases other than TTP [22, 25] call for accurate quantitative assay methodologies, ideally based on more than one method for measuring the in vitro VWFcp activity of ADAMTS13. We developed a new, rapid, and simple method based on the positive correlation between VWF multimeric size and Ristocetin cofactor activity (VWF:RCo), which is extensively compared to the original immunoblotting method. The new technique was utilized to asses VWFcp activity in 14 TTP patients during acute episodes, therapy, and remission, in 80 healthy controls, in 23 patients with thrombocytopenia and/or hemolysis, and in 14 patients with antiphospholipid syndrome (APS).

Materials and methods Plasma samples Citrated plasma samples were obtained from 14 patients during 22 events of acute TTP before initiation of plasma exchange (PE) therapy with fresh-frozen plasma (FFP). Initial diagnosis was based on clinical symptoms (predominantly neurological disturbances) and laboratory findings such as severe thrombocytopenia, microangiopathic hemolytic anemia as indicated by red cell fragmentocytes in the peripheral blood smear, and significantly elevated lactate dehydrogenase levels. From 11 of 14 patients we also obtained plasma samples during remission. All patients manifested with nonfamiliar TTP; however, genetic analysis had not been performed prior to submitting this paper. We serially collected blood from ten patients during PE therapy. PE combined with corticosteroids and additionally vincristine for patients with severe neurological disorders was efficacious in 13 of 14 patients. One patient with bone marrow transplantation (BMT)-associated TTP died. In addition, plasma samples from 23 patients with acute thrombocytopenia and/or hemolysis, 14 patients with APS, and 80 apparently healthy subjects were analyzed. Platelet-poor plasma was prepared by centrifugation for 40 min at 4°C and 2500 g. The supernatant was subsequently stored at –20°C until tested. All individuals gave informed consent for blood sampling and research use. Assay of VWF-cleaving activity by the Ristocetin cofactor-based method Plasma samples were diluted 1:21 (final volume 210 µl) with 5 mM Tris-HCl buffer, pH 8 containing 12.5 mM BaCl2 and 1 mM PefaBloc SC (AppliChem GmbH, Darmstadt, Germany) followed by 5 min incubation at 37°C for activation of the protease. As a substrate a purified VWFactor (Concentre de Facteur Willebrand Humain Tres Haute Purite) kindly provided by Laboratoire Francais du Fractionnement et des Biotechnologies (Lille, France), was used. The concentrate, which lacks endogenous VWF-cleaving protease activity [1], was reconstituted to 100 U/ml, aliquoted, and stored at –20°C until use. Prior to digestion the substrate was thawed, diluted 1:20 with 5 M urea in 5 mM Tris-HCl, pH 8, and incubated for 10 min at RT. Subsequently, 100 µl of the substrate dilution was added to the diluted plasma and digested overnight at 37°C. Thereafter, the residual Ristocetin cofactor activity (VWF:RCo) was estimated with a commercial test from Dade Behring (Marburg, Germany). The activity of VWFcp in a normal human plasma pool (NHP) of 80 apparently healthy adults (40 F/40 M) in 1:21 dilution was arbitrarily defined as 100%. For calibration, serial dilutions of NHP with imidazole

buffer, pH 7.4 (Technoclon GmbH, Vienna, Austria) were plotted against the residual VWF:RCo. The calibration curve with the equation y=A+(D-A)/(1+e (B*(c-x)) was calculated with Easy fit, version 5.14 software from Tecan (Basel, Switzerland) and used to assess the VWFcp activity of the samples (expressed in % of normal). Assay of VWF-cleaving activity by the immunoblotting method For measurement of VWFcp activity by the immunoblotting method, we used a modification of the method first described by Furlan et al. and Tsai [10, 29]. Plasma samples were diluted 1:5 with 5 mM Tris buffer including 12.5 mM BaCl2 and 1 mM PefaBloc SC. Activation and digestion were done as described for the RCo-based assay. Reaction was stopped by adding disodiumEDTA to a final concentration of 23.5 mM. Multimer analysis was done by SDS electrophoresis on 1.4% agarose [Seakem HGT (P) from Biozym Diagnostics, Hess. Oldenburg, Germany] and immunoblotting using a peroxidase-conjugated anti-VWF antibody (P0226 from Dako, Glostrup, Denmark) as previously described [5]. For quantification, serial dilutions of a normal plasma pool were tested as above. Inhibitor assay For measuring the inhibitory activity against the VWFcp, plasma samples, either neat or diluted with imidazole buffer, were mixed in equal parts with NHP. For reference NHP was diluted 1:1 with imidazole buffer. After 30 min incubation of the mixtures at 37°C, the VWFcp activity was determined with the RCo-based or the immunoblotting method as described. The VWFcp activity of the test sample was divided through the activity of the reference mixture and multiplied by 100 to obtain the residual VWFcp activity. For quantification, samples with residual activity from 25 to 75% were selected to calculate the amount of inhibitor expressed in U/ml as described for coagulation factor VIII inhibitors [17]. A sample causing an inhibition of normal VWFcp activity by 50% was assumed to contain an inhibitor concentration of 1 U/ml.

Results Accuracy and reproducibility of the RCo-based VWFcp activity assay Representative calibration curves for the novel RCobased VWFcp activity assay and the immunoblotting assay are illustrated in Fig. 1. The immunoblotting technique discriminates between samples with VWFcp activity of <3.12 and 3.12–6.25%, whereas the RCo-based assay has a detection limit of 6.25%. The normal range for the RCo-based assay of 80 healthy controls is 52–134% (Fig. 3). The normal cutoff value in the immunoblotting method is 50% (data not shown). To compare the RCo-based assay with the immunoblotting method, 282 plasma samples from TTP patients, patients with thrombocytopenia and/or hemolysis, and healthy subjects were tested in both methods. Results in Fig. 2 demonstrate the high comparability of the methods. The immunoblotting method initially allowed classification of VWFcp activity into four categories: severely deficient (<12.5%), partially deficient (12.5–25%), mildly deficient (25–50%), and normal (>50%). For 71 of 72 severely deficient plasma samples we obtained

432

Fig. 3 VWFcp as measured by the RCo-based assay from 14 patients during 22 acute episodes of TTP, 11 patients during TTP remission, 80 healthy subjects, 23 patients with thrombocytopenia and/or hemolysis without TTP, and 14 patients with APS Table 1 Coefficients of variation (CV) for the RCo-based VWFcp activity assay Abnormal sample Intra assay CV Mean±SD Interassay CV Mean±SD

Fig. 1 A Representative calibration curve for the RCo-based assay. B Representative calibration for the immunoblotting assay with exemplary samples from TTP patients during PE therapy. Lanes 1 and 2 patient with persistent VWFcp deficiency of <3.12% (see also Fig. 5B), lane 3 patient with increase of VWFcp to ~50% during PE (see also Fig. 4)

8.7% (n=20) 46%±4.02 12.9% (n=26) 38%±4.89

Normal sample 2.8% (n=20) 95%±2.68 7.5% (n=6) 97%±7.26

consistent results in both methods (<12.5%). For one sample from a TTP patient during therapy we obtained an activity of 0–12.5% in the immunoblotting method and of 18% in the RCo-based assay. We subsequently increased the sensitivity of our assays to 3.12% in the immunoblotting method and 6.25% in the RCo-based assay (Fig. 1). The 19 partially deficient samples in the immunoblotting method had values in the RCo-based assay from 9 to 39%. For the 64 mildly deficient samples in the immunoblotting method we obtained in the RCobased assay values from 20 to 60%. The 127 normal samples tested were normal in both methods. The reproducibility of the new assay was verified by low intraand interassay coefficients of variation as shown in Table 1. VWFcp activity in TTP patients, patients with thrombocytopenia and/or hemolysis, patients with APS, and healthy subjects

Fig. 2 Comparison of the novel RCo-based assay with the original immunoblotting method. VWFcp activity from 282 plasma samples was assessed with both methods as described. Samples were classified according to results of immunoblotting method: severely deficient (<12.5%, triangles), partially deficient (12.5–25%, diamonds), mildly deficient (25–50%, circles), and normal (>50%, squares). Results from the RCo-based assay are ascribed to the y-axis

Figure 3 displays data of VWFcp activity as measured by the RCo-based assay. Of 22 plasma samples of acute TTP, 20 were severely deficient for VWFcp activity (<12.5% and <6.25%, respectively). Of 22 events of acute TTP, 2 were associated with bone marrow transplantation (BMT-TTP); these patients had VWFcp activity of 39% and 62%, respectively. We obtained data of

433

134%. Twenty-one of 23 patients with thrombocytopenia and/or hemolysis were normal in VWFcp activity. A patient with diarrhea-associated hemolytic uremic syndrome and a patient with graft rejection after kidney transplantation had borderline VWFcp activity of 47 and 51%, respectively. All 14 patients with APS had normal VWFcp activity with a range from 58 to 112%. VWFcp activity during therapy Fig. 4 Exemplary course of platelet count and VWFcp activity during initial episode of TTP from a female patient (67 years) during PE therapy (at least 4 PE sessions/week). The patient was additionally treated with corticosteroids and vincristine

We serially monitored VWFcp activity during PE therapy for nine patients. Six of nine patients demonstrated an increase of VWFcp activity during therapy, which was associated with increase of platelet count. Figure 4 displays exemplary the index episode of TTP from a female patient (67 years). The patient suffered various seizures and was additionally treated with vincristine. An early relapse after 5 days was associated with renewed decrease of VWFcp activity and increase of inhibitory activity (data of inhibitor test not shown). Three of nine patients remained severely deficient for VWFcp activity during therapy. Figure 5A illustrates exemplary a course of the eighth relapse from a female patient where only a transient increase of VWFcp activity from <6.25 to 11% could de detected after first PE session. The patient had an inhibitor titer of 4.3 U/ml at admission, which increased to 29 U/ml after 26 days of PE therapy (Fig. 5 B).

Discussion

Fig. 5 A Exemplary course of platelet count and VWFcp activity during a relapse of TTP from a female patient (70 years, TTP onset at 53 years) during PE therapy (at least 4 PE sessions/week). B Inhibitor titer for the patient at admission and after 25 and 26 days of PE therapy determined as described. Dilution factor of patient plasma is ascribed to the x-axis (zero refers to undiluted plasma). Residual VWFcp is calculated by division through VWFcp activity of reference mixture and ascribed to the y-axis

protease activity during acute phase and remission from 11 of 12 patients with classic TTP. Of 11 patients 3 remained severely deficient with persistent inhibitory activity; 4 of 11 patients demonstrated partial deficiency in remission, whereas in 4 of 11 patients VWFcp activity was completely normalized. To calculate the normal range, the 80 healthy subjects of the NHP were tested individually. They presented VWFcp activity from 52 to

We present a new method for measuring the VWF-cleaving protease activity of ADAMTS13. The comparison between the novel assay and the original immunoblotting method, as well as the clinical data, verify the accuracy of the RCo-based assay. The method is reproducible and does not require special laboratory equipment and expertise. It can be performed overnight and is therefore less time consuming than the immunoblotting technique. The method allows early diagnosis thus rapid initiation of therapy, which is essential for successful cure of the acute episode. The immunoblotting method is more sensitive in the lower range than the RCo-based method; however, VWFcp activity <12.5% and <6.25%, respectively, as measured by the RCo-based method and the immunoblotting method was restricted to patients with TTP. Our data herewith confirm the specific severe VWFcp deficiency in acute TTP found by several investigators [1, 3, 11, 12, 15, 32, 33, 35] and contradict recent findings from others who found VWFcp deficiency to be less specific for acute TTP [23, 25]. The two patients with BMT-TTP demonstrated partially deficient and normal VWFcp activity. This in agreement with others who found VWFcp activity in BMT-TTP partially decreased [35] or normal [34]. The partial VWFcp deficiency in the patient with BMT-TTP might also be a consequence of her severe sepsis as VWFcp activity has been found to be significantly de-

434

creased in sepsis [4]. The difference of BMT-associated TTP from classic TTP concerning etiology, response to treatment, and VWFcp activity points to the necessity for a new classification of thrombotic microangiopathies (TMA) as it has also been suggested by others [20]. Serial monitoring during PE demonstrated an increase of VWFcp activity in six of nine cases, which was associated with an increase of platelet count. Those patients demonstrated normal or partially decreased VWFcp activity in remission, whereas patients with persistent inhibitory activity during therapy remained severely deficient in VWFcp activity despite clinical remission. The therapeutic efficacy of plasma exchange and plasma infusion is considered to be a consequence of reconstitution of VWFcp activity by decreasing the concentration of autoantibodies and substitution of the proteases [7, 9]. The persistency of VWFcp deficiency might be due to the relative insensitivity of the in vitro assay in the lower range, but might also be a hint that PE eliminates other causative, hitherto unidentified triggers. We detected for two patients an increase of inhibitor titer during PE, which was also described by others [11, 18, 30]. The FFP application might lead to antibody production in those patients by anamnestic response. Alternative treatment regimens such as splenectomy [11], immunoadsorption [18], or the monoclonal CD20-antibody rituximab [6] could be viable options for high-titer inhibitor patients who are prone to relapse [Scharrer, Böhm, unpublished] and may be delayed in response to PE [31]. In conclusion, the novel RCo-based method was shown to be a reliable and simple instrument for measuring VWF-cleaving protease activity of ADAMTS13. Thus, it can be used for early diagnoses of TTP, for monitoring of treatment, and for further distinction between TMA entities. Furthermore, it will be a convenient tool for future research on the recently identified protein ADAMTS13, the pathogenesis of TTP, and the postsecretory processing of VWF in general. Acknowledgements We thank Lothar Bergmann, MD and Hans Martin, MD for providing clinical data and plasma samples from two patients. We acknowledge LFB France for supplying the VWF concentrate.

References 1. Allford SL, Harrison P, Lawrie AS, Liesner R, Mackie IJ, Machin SJ (2000) Von Willebrand factor-cleaving protease activity in congenital thrombotic thrombocytopenic purpura. Br J Haematol 111:1215–1222 2. Aronson DL, Krizek DM, Rick ME (2001) A rapid assay for the vWF protease. Thromb Haemost 85:184–185 3. Barbot J, Costa E, Guerra M, Barreirinho MS, Isvarlal P, Robles R, Gerritsen HE, Lämmle B, Furlan M (2001) Ten years of prophylactic treatment with fresh-frozen plasma in a child with chronic relapsing thrombotic thrombocytopenic purpura as a result of a congenital deficiency of von Willebrand factor-cleaving protease. Br J Haematol 113:649–651 4. Bianchi V, Zeerleder S, Kessler P, Robles R, Wuillemin W, Alberio L, Lämmle B (2002) vWF-cleaving protease in severe sepsis and septic shock (abstract 22). Ann Hematol 81 [Suppl]

5. Bukh A, Ingerslev J, Stenbjerg S, Hundahl Møller NP (1986) The multimeric structure of plasma FVIII:Ag studied by electroelution and immunoperoxidase detection. Thromb Res 43:579–584 6. Chemnitz J, Draube A, Schulz A, Scheid C, Fuchs M, Diehl V, Söhngen D (2001) Rituximab in the treatment of severe thrombotic thrombocytopenic purpura (ttp) (abstract 131). Blood 98 7. Chemnitz J, Schulz A, Diehl V, Söhngen D (2001) Thrombotisch-thrombozytopenische purpura (Moschcowitz-Syndrom). Med Klin 96:343–350 8. Fujikawa K, Suzuki H, McMullen B, Chung D (2001) Purification of human von Willebrand factor-cleaving protease and its identification as a new member of the metalloproteinase family. Blood 98:1662–1666 9. Furlan M, Lämmle B (1999) Von Willebrand factor in thrombotic thrombozytopenic purpura. Thromb Haemost 82:592– 600 10. Furlan M, Robles R, Lämmle B (1996) Partial purification and characterisation of a protease from human plasma cleaving von Willebrand factor to fragments produced by in vivo proteolysis. Blood 10:4223–4234 11. Furlan M, Robles R, Solenthaler M, Lämmle B (1998) Acquired deficiency of von Willebrand factor-cleaving protease in a patient with thrombotic thrombocytopenic purpura. Blood 91:2836–2846 12. Furlan M, Robles R, Galbusera M, Remuzzi G, Kyrle PA, Brenner B, Krause M, Scharrer I, Aumann V, Mittler U, Solenthaler M, Lämmle B (1998) Von Willebrand factorcleaving protease in thrombotic thrombocytopenic purpura and the hemolytic-uremic syndrome. N Engl J Med 339:1578– 1584 13. Gerritsen HE, Robles R, Lämmle B, Furlan M (2001) Partial amino acid sequence of purified von Willebrand factor-cleaving protease. Blood 98:1654–1661 14. Gerritsen HE, Turecek PL, Schwarz HP, Lämmle B, Furlan M (1999) Assay of von Willebrand factor (vWF)-cleaving protease based in decreased collagen binding activity of degraded vWF. Thromb Haemost 82:1386–1389 15. Häberle J, Kehrel B, Ritter J, Jürgens H, Lämmle B, Furlan M (1999) New strategies in diagnosis and treatment of thrombotic thrombocytopenic purpura: case report and review. Eur J Pediatr 158:883–887 16. He S, Cao H, Magnusson CGM, Eriksson-Berg M, Mehrkash M, Schenck-Gustafsson K, Blombäck M (2001) Are increased levels of von Willebrand factor in chronic coronary heart disease caused by decrease in von Willebrand factor cleaving protease activity? A study by an immunoassay with antibody against intact bond 842Tyr-843Met of the von Willebrand factor protein. Thromb Res 103:241–248 17. Kasper CK (1991) Laboratory tests for factor VIII inhibitors, their variation, significance and interpretation. Blood Coagul Fibrinolysis 2:7–10 18. Knöbl P, Laczika K, Staudinger T, Haas M, Derfler K, Frass M, Schreiner J, Varadi K Turecek PL (2002) Kinetic of inhibitors of von Willebrand factor cleaving metalloprotease in three patients with severe thrombotic thrombocytopenic purpura (abstract 166). Ann Hematol 81 [Suppl] 19. Krizek DM, Rick ME (2001) Clinical application of a rapid method using agarose gel electrophoresis and Western blotting to evaluate von Willebrand factor protease activity. Electrophoresis 22:946–949 20. Lämmle B, Hunt BJ, Haycock GB, Furlan M (2002) Reply to rebuttal: von Willebrand factor-cleaving protease in childhood diarrhoea-associated haemolytic uraemic syndrome. Thromb Haemost 87:549–550 21. Levy GG, Nichols WC, Lian EC, Foroud T, McClintick JN, McGee BM, Yang AY, Siemieniak DR, Stark KR, Gruppo R, Sarode R, Shurin SB, Chandrasekaran V, Stabler SP, Sabio H, Bouhassira EE, Upshaw JD, Ginsburg D, Tsai HM (2001) Mutations in a member of the ADAMTS gene family cause thrombotic thrombocytopenic purpura. Nature 413:488–494

435 22. Mannucci PM, Canciani MT, Forza I, Lussana F, Lattuada A, Rossi E (2001) Changes in health and disease of the metalloprotease that cleaves von Willebrand factor. Blood 98:2730–2735 23. Mannucci PM, Galbusera M, Noris M, Canciani MT, Remuzzi G, Bianchi Bonomi A (2001) Deficiency of von Willebrand factor (vWF)-cleaving protease activity does not discriminate thrombotic thrombocytopenic purpura (TTP) from the hemolytic uremic syndrome (HUS) (abstract 3140). Blood 98 24. Moake JL, Rudy CK, Troll JH, Weinstein MJ, Colannino NM, Azocar J, Seder RH, Hong SL, Deykin D (1982) Unusually large plasma factor VIII: von Willebrand factor multimers in chronic relapsing thrombotic thrombocytopenic purpura. N Engl J Med 307:1432–1435 25. Moore JC, Hayward CPM, Warkentin TE, Kelton JG (2001) Decreased von Willebrand factor protease activity associated with thrombocytopenic disorders. Blood 98:1842–1846 26. Moschcowitz E (1924) Hyaline thrombosis of the terminal arterioles and capillaries: a hitherto undescribed disease. Proc N Y Pathol Soc 24:21–24 27. Obert B, Tout H, Veyradier A, Fressinaud E, Meyer D, Girma JP (1999) Estimation of the Willebrand factor-cleaving protease in plasma using monoclonal antibodies to vWF. Thromb Haemost 82:1382–1385 28. Raife TJ, Atkinsons B, Christopherson P, Jozwiak M, Montgomery RR (2001) Recombinant, truncated monomeric von Willebrand factor (vWF) for the study of vWF proteolysis (abstract 1667). Thromb Haemost [Suppl July]

29. Tsai HM (1996) Physiologic cleavage of von Willebrand factor by a plasma protease is dependent on its conformation and requires calcium ion. Blood 10:4235–4244 30. Tsai HM (2000) High titers of inhibitors of von Willebrand factor-cleaving metalloproteinase in a fatal case of acute thrombotic thrombocytopenic purpura. Am J Hematol 65:251– 255 31. Tsai HM, Li A, Rock G (2001) Inhibitors of von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura. Clin Lab 47:387–392 32. Tsai HM, Lian EC (1998) Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic Purpura. N Engl J Med 339:1585–1594 33. Tsai HM, Rice L, Sarode R, Chow TW, Moake JL (2000) Antibody inhibitors to von Willebrand factor metalloproteinase and increased binding of von Willebrand Factor to platelets in Ticlopidine-associated thrombotic thrombocytopenic purpura. Ann Intern Med 132:794–799 34. van der Plas RM, Schiphorst ME, Huizinga EG, Hene RJ, Verdonck LF, Sixma JJ, Fijnheer R (1999) Von Willebrand factor proteolysis is deficient in classic, but not in bone marrow transplantation-associated thrombotic thrombocytopenic purpura. Blood 93:3798–3802 35. Veyradier A, Obert B, Houllier A, Meyer D, Girma JP (2001) Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases. Blood 98:1765– 1772