Characterization of human antibodies against ADAMTS13 that develop in patients with acquired TTP

Abstract

Thrombotic thrombocytopenic purpura (TTP) is a thrombotic micro-angiopathy characterized by the absence or dysfunction of the von Willebrand factor cleaving protease ADAMTS13. Functional absence of ADAMTS13 results in impaired cleavage of newly released ultra large von Willebrand factor (UL-VWF) multimers on the surface of endothelial cells. Accumulation of these UL-VWF multimers results in formation of VWF-rich platelet thrombi in the microcirculation resulting in thrombocytopenia, haemolytic anemia, neurological symptoms and/or renal dysfunction. The majority of patients with TTP develop autoantibodies directed towards ADAMTS13. This thesis describes the functional characterization of the autoantibodies directed to ADAMTS13 that develop in TTP patients. Chapter 1 summarizes our current knowledge on the biology and function of ADAMTS13 with emphasis on the properties of antibodies that develop in patients with acquired TTP. Chapter 2 describes work on the isolation of a panel of patient-derived monoclonal antibodies via phage-display. We show that two of the isolated antibodies, designated I-9 and II-1, inhibited the activity of ADAMTS13 in vitro using different VWF substrates. In Chapter 3 we show that the two patient-derived monoclonal anti-ADAMTS13 antibodies target a single epitope on the outer surface of the spacer domain of ADAMTS13 comprising residues Arg660, Tyr661 and Tyr665. We also show that this exposed interactive surface on the spacer domain promotes binding of ADAMTS13 to a complementary exosite comprising residues Glu1660-Arg1668 in the VWF A2 domain. The contribution of Arg660, Tyr661 and Tyr665 to ADAMTS13 activity was determined using different VWF substrates. Residues Arg660, Tyr661 and Tyr665 involved are part of one surface loop on the outer surface of the spacer domain. Inspection of the three dimensional structure of the spacer domain revealed two possible other surface loops in close proximity to residues Arg660, Tyr661 and Tyr665. We assessed whether Arg658 and Phe592 present in these other surface loops contribute to the binding of anti-ADAMTS13 antibodies. The reactivity of polyclonal antibodies present in the plasma of a large cohort of acquired TTP patients was lost after replacement of Arg660, Tyr661, Tyr665, Arg658 and Phe592 by an Ala (chapter 4). These data suggest the presence of a single antigenic surface comprising Arg568, Phe592, Arg660, Tyr661 and Tyr665 is targeted by autoantibodies in patients with acquired TTP. We also measured the class of antibodies in the plasma of this cohort of acquired TTP patients. We concluded that the majority of anti-ADAMTS13 antibodies were composed of IgG. Some patients also have circulating levels of anti-ADAMTS13 IgM and IgA1. The presence of anti-ADAMTS13 IgA1 is intriguing but its significance is not yet known. Binding of ADAMTS13 to VWF is mediated by multiple domains on either protein. We assessed whether chemical modification of surface exposed lysine and tyrosine residues on the surface of ADAMTS13 affected its binding to different VWF substrates (chapter 5). Chemically modified residues were identified by mass spectrometry. Our data show that lysines play a minor role whereas the tyrosine residues on the spacer domain show to have a major role in the interaction of ADAMTS13 with small substrates. These data are consistent with previous findings that Tyr661 and Tyr665 are involved in binding of ADAMTS13 to VWF. Together, these studies increase our knowledge of the characteristics of anti-ADAMTS13 antibodies that develop in patients with acquired TTP. Also novel insight into the structure-function relationship of ADAMTS13 has been obtained.