Identification of novel cryptic epitopes uncovers a distinct and consistent open ADAMTS13 conformation in iTTP

Introduction: Plasma ADAMTS13 (pADAMTS13) circulates in a closed conformation. Besides the well-described Spacer (S)-CUB domain interaction, recent AlphaFOLD2 predictions suggested additional interdomain interactions in the closed ADAMTS13 conformation. In immune-mediated thrombotic thrombocytopenic purpura (iTTP), patient IgG autoantibodies open ADAMTS13 to expose a cryptic S domain epitope recognized by our monoclonal antibody (mAb) 1C4. This open ADAMTS13 conformation has gained clinical relevance as a biomarker for diagnosing iTTP and monitoring an increased relapse risk in remission patients. Still, it remained unclear whether iTTP patients display one distinct or multiple different open ADAMTS13 conformations. Therefore, we aimed to identify additional mAbs that recognize cryptic epitopes in open pADAMTS13 to further characterize the conformation of open pADAMTS13 in iTTP patients.

Method: In ELISA, our large mAb library was screened to identify mAbs that recognize cryptic epitopes. Epitopes were considered cryptic when mAbs could not bind to closed pADAMTS13, but could bind to pADAMTS13 upon conformational opening with our anti-CUB1 mAb 17G2. ADAMTS13 antigen and conformation were evaluated for 25 healthy donors (HDs) and 25 iTTP patients. The conformation of HD pADAMTS13 was screened upon preincubation with IgGs purified from iTTP patients.

Results: Besides our anti-S (1C4) and anti-metalloprotease (6A6) mAbs, four mAbs were found to recognize previously unidentified cryptic epitopes in the disintegrin-like (1D5), seventh (9C12) and eight (19H4) thrombospondin type-1 repeat and CUB1 (10D2) domains. HD screening using each mAb defined a cut-off for distinguishing the open and closed pADAMTS13 conformations. Nearly all HDs showed a closed conformation with all cryptic epitopes concealed. Upon 17G2 preincubation, pADAMTS13 opened, exposing all cryptic epitopes. Three HDs exhibited a partially open conformation as both 6A6 and 9C12 epitopes were exposed, while the 1C4, 1D5, 19H4 and 10D2 epitopes remained concealed. Consistently, all iTTP patients exhibited a distinct open conformation exposing all cryptic epitopes, regardless of 17G2 presence. Incubation of closed plasma ADAMTS13 from HDs with iTTP patient-purified IgGs induced the same distinct open conformation with all cryptic epitopes exposed.

Conclusion: Our study provides new insights into the open ADAMTS13 conformation in two ways. Firstly, we demonstrate that the open conformation exposes multiple cryptic epitopes across various domains, extending beyond those defined by the S-CUB interaction. This suggests that the closed form of ADAMTS13 likely adopts a compact, folded structure stabilized by more interdomain interactions than previously recognized. Secondly, we reveal that once opened, ADAMTS13 adopts a distinct open conformation that is consistently observed in all iTTP patients.

Publication History

Article published online:
13 February 2025