Article Text

Download PDFPDF

OP0177 N-glycosylation sites in the variable domain of b cell receptors specific for citrullinated antigens
  1. LM Slot1,
  2. RD Vergroesen1,
  3. L Hafkenscheid1,
  4. MT Koning2,
  5. T Rispens3,
  6. H Veelken2,
  7. RE Toes1,
  8. HU Scherer1
  1. 1Department of Rheumatology
  2. 2Department of Hematology, Leiden University Medical Center, Leiden
  3. 3Department of Immunopathology, Sanquin Research, Amsterdam, Netherlands


Background Recent structural analysis of anti citrullinated protein antibodies (ACPA) in serum and synovial fluid of patients with rheumatoid arthritis (RA) revealed that the vast majority (>90%) of secreted ACPA IgG molecules carry N-glycans in the variable (Fab) domain. This remarkable degree of Fab-glycosylation is absent from ACPA-depleted control IgG and from autoantibodies in other diseases. So far, it is unclear why ACPA carry this feature and which biological effects these glycans mediate in the context of RA. Of note, however, N-glycosylation requires a specific amino acid consensus sequence in the protein backbone, which is very rare in germline-encoded variable region genes.

Objectives To study the B cell receptor (BCR) repertoire of ACPA-expressing B cells to determine the frequency, origin and localisation of N-glycosylation sites in ACPA Fab domains.

Methods Citrullinated antigen-specific and non citrulline-reactive control B cells were identified in peripheral blood of ACPA-positive RA patients by antigen-specific tetramer staining and isolated by fluorescence activated cell sorting. Cells were either sorted in pools and directly lysed, or sorted as single cells and cultured for two weeks followed by the detection of ACPA-positive culture wells by ELISA. Full-length immunoglobulin (Ig) rearrangements were identified by anchoring reverse transcription of Ig sequences, amplification by nested PCR and either next generation sequencing (NGS, PacBio platform) or, for single cell transcripts, Sanger sequencing (scSeq). Sequence reads were analysed using IMGT V-QUEST tools.

Results The mean number of nucleotide mutations in heavy chains (HC) of IgG BCRs derived from ACPA-expressing B cells was high (33 in NGS, 48 in scSeq samples; similarity to germline: 88% in NGS, 84% in scSeq). NGS identified 12 unique IgG clones derived from 4 donors, of which 10 (83%) had at least one N-glycosylation site in the HC or light chain (LC). scSeq identified 86 unique IgG clones derived from 6 donors, of which 68 (79%) had N-glycosylation sites. For 57/86 IgG clones, we could determine the combination of HC and LC sequences. In these, only 7 (12%) clones had no sites, while 19 (33%) had one, 23 (41%) had two, 5 (9%) had three and 3 (4%) clones had four sites. 57 sites were found in the HC and 34 in the LC. All N-glycosylation sites were created by somatic mutations and not encoded in the germline sequence. Several sites were located in antigen-engaging regions. No correlation was found between the number of N-glycosylation sites and the number of somatic mutations.

Conclusions We demonstrate that B cell surface-expressed ACPA-IgG molecules carry a remarkably high frequency of N-glycosylation sites in the Fab domain, all of which are generated by somatic mutation. This could indicate that ACPA-expressing B cells acquire a selective survival advantage by introducing N-glycosylation consensus sequences in the Fab domain, a process that is likely to occur under the influence of T cell help and that could facilitate the break of tolerance to citrullinated antigens.

Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.