N-glycan-mediated shielding of ADAMTS13 prevents binding of pathogenic autoantibodies in immune-mediated TTP.

Immune-mediated thrombotic thrombocytopenic purpura (iTTP) is an autoimmune disorder caused by the development of autoantibodies targeting different domains of ADAMTS13. Profiling studies have shown that residues R568, F592, R660, Y661, and Y665 within exosite-3 of the spacer domain provide an immunodominant region of ADAMTS13 for pathogenic autoantibodies that develop in patients with iTTP. Modification of these 5 core residues with the goal of reducing autoantibody binding revealed a significant tradeoff between autoantibody resistance and proteolytic activity. Here, we employed structural bioinformatics to identify a larger epitope landscape on the ADAMTS13 spacer domain. Models of spacer-antibody complexes predicted that residues R568, L591, F592, K608, M609, R636, L637, R639, R660, Y661, Y665, and L668 contribute to an expanded epitope within the spacer domain. Based on bioinformatics-guided predictions, we designed a panel of N-glycan insertions in this expanded epitope to reduce the binding of spacer domain autoantibodies. One N-glycan variant (NGLY3-ADAMTS13, containing a K608N substitution) showed strongly reduced reactivity with TTP patient sera (28%) as compared with WT-ADAMTS13 (100%). Insertion of an N-glycan at amino acid position 608 did not interfere with processing of von Willebrand factor, positioning the resulting NGLY3-ADAMTS13 variant as a potential novel therapeutic option for treatment of iTTP.