Background/Purpose:

Human-murine chimeric variants have been used to study specific epitope recognition by anti-neutrophil cytoplasmic antibodies (ANCAs) targeting proteinase 3 (PR3) in patients with ANCA-associated vasculitis.  In prior work, a human-murine chimeric PR3 variant (Hm5) was generated by substituting the three human hydrophobic amino acids in Epitope 5 of PR3 by their murine hydrophilic counter parts.¹ Rather than loss of binding of PR3-ANCA to Hm5 ¹ we observed more avid binding of Hm5 by PR3-ANCAs compared with the mature conformation of PR3. Further studies were completed to better characterize this phenomenon.

Methods:

Recombinant target antigens carrying C-terminal poly-HIS tags (PR3, Hm5, myeloperoxidase [MPO], human neutrophil elastase [HNE]) were anchored to nickel-coated plates and used for PR3- and MPO-ANCA detection.  Serum samples were obtained from participants in the Wegener’s Granulomatosis Etanercept (WGET) and in Rituximab versus Cyclophosphamide for ANCA-Associated Vasculitis (RAVE) trials. DNA barcode-enabled sequencing of the antibody repertoire of plasmablasts from 5 participants in the RAVE trial was performed, and 25 human monoclonal antibodies (moAb) were selected for recombinant expression and tested for reactivity with PR3, Hm5, MPO and HNE. Antigen-binding fragments (Fab) of anti-PR3 moAbs were used for inhibition.

Results:

All PR3-ANCA positive samples reacted equally with or had increased recognition of Hm5 compared to PR3 (Figure 1A and 1B). In 2 of 52 serum samples from MPO-AAV patients included in RAVE, ANCA binding was detected when using Hm5 as antigen, but not when using PR3; all other samples were unreactive with either PR3 variant.  All 25 human moAb derived from patient plasmablasts tested negative for reactivity with PR3, MPO and HNE, but one of these (moAb518) had strong selective reactivity with Hm5. The binding of moAb518 could be inhibited by Fabs from moAbs specific for Epitope 3, but not those specific for Epitope 5 where the 3 mutated residues of Hm5 are located (Figure 1C).  The preferred binding of PR3-ANCA from patients to Hm5 could be reduced by inhibition with Epitope-3 specific Fabs in the same way as inhibition with moAb518 (Figure 1D).

Conclusion:

(1) Hm5 can serve as an antigen for high sensitivity PR3-ANCA immunoassays. (2) The plasmablast-derived human moAb518 binds selectively to Epitope 3 on Hm5, indicating an unexpected mutational effect on Epitope 3 that is located on the opposite side of Epitope 5. (3) New studies are needed to explain how mutational effects of distal residues can affect conformational epitopes of an antigen.

Reference

1.            Silva F, Hummel AM, Jenne DE, Specks U. Discrimination and variable impact of ANCA binding to different surface epitopes on proteinase 3, the Wegener’s autoantigen. Journal of autoimmunity. 2010;35(4):299-308.