Differential ACPA binding to nuclear antigens reveals a PAD-independent pathway and a distinct subset of acetylation cross-reactive autoantibodies in rheumatoid arthritis

Katy A. Lloyd, Gustaf Wigerblad, Peter Sahlström, Manasa G. Garimella, Karine Chemin, Johanna Steen, Philip J. Titcombe, Bianka Marklein, Diana Zhou, Ragnhild Stålesen, Elena Ossipova, Christina Lundqvist, Olov Ekwall, Johan Rönnelid, Daniel L Mueller, Mikael C.I. Karlsson, Mariana J. Kaplan, Karl Skriner, Lars Klareskog, Fredrik WermelingVivianne Malmström, Caroline Grönwall

Research output: Contribution to journalArticlepeer-review

21 Scopus citations

Abstract

Rheumatoid arthritis (RA) associated anti-citrullinated protein autoantibodies (ACPA) target a wide range of modified proteins. Citrullination occurs during physiological processes such as apoptosis, yet little is known about the interaction of ACPA with nuclear antigens or apoptotic cells. Since uncleared apoptotic cells and neutrophil extracellular trap (NET) products have been postulated to be central sources of autoantigen and immunostimulation in autoimmune disease, we sought to characterize the anti-nuclear and anti-neutrophil reactivities of ACPA. Serology showed that a subset of anti-CCP2 seropositive RA patients had high reactivity to full-length citrullinated histones. In contrast, seronegative RA patients displayed elevated IgG reactivity to native histone compared to controls, but no citrulline-specific reactivity. Screening of 10 single B-cell derived monoclonal ACPA from RA patients revealed that four ACPA exhibited strong binding to apoptotic cells and three of these had anti-nuclear (ANA) autoantibody reactivity. Modified histones were confirmed to be the primary targets of this anti-nuclear ACPA subset following immunoprecipitation from apoptotic cell lysates. Monoclonal ACPA were also screened for reactivities against stimulated murine and human neutrophils, and all the nuclear-reactive monoclonal ACPA bound to NETs. Intriguingly, one ACPA mAb displayed a contrasting cytoplasmic perinuclear neutrophil binding and may represent a different NET-reactive ACPA subset. Notably, studies of CRISPR-Cas9 PAD4 KO cells and cells from PAD KO mice showed that the cytoplasmic NET-binding was fully dependent on PAD4, whilst nuclear- and histone-mediated NET reactivity was largely PAD-independent. Our further analysis revealed that the nuclear binding could be explained by consensus-motif driven ACPA cross-reactivity to acetylated histones. Specific acetylated histone peptides targeted by the monoclonal antibodies were identified and the anti-modified protein autoantibody (AMPA) profile of the ACPA was found to correlate with the functional activity of the antibodies. In conclusion, when investigating monoclonal ACPA, we could group ACPA into distinct subsets based on their nuclear binding-patterns and acetylation-mediated binding to apoptotic cells, neutrophils, and NETs. Differential anti-modified protein reactivities of RA-autoantibody subsets could have an important functional impact and provide insights in RA pathogenesis.

Original languageEnglish (US)
Article number3033
JournalFrontiers in immunology
Volume10
Issue numberJAN
DOIs
StatePublished - 2018

Bibliographical note

Funding Information:
This work was supported by the Swedish Research Council, the Swedish Rheumatism Association, King Gustaf V's 80-year Foundation, the Region Västra Götaland (ALFGBG-718021), IngaBritt, and Arne Lundbergs Research Foundation, NIH/NIGMS (MSTP T32 GM008244, PT), Rheumatology Research Foundation (Disease Targeted Innovative Research Grant, Within our Reach Award, DM), and the EU/EFPIA Innovative Medicines Initiative (IMI) 2 Joint Undertaking projects BTCure 115142 and RTCure 777357. Funded in part by NIAMS (ZIAAR041199). We thank UCB Pharma for support with technology and antibody production as well as financial support for KL postdoctoral training within the IMI BTCure project, and Stephen Rapecki (UCB Pharma) for scientific discussions. We would also like to thank Lena Israelsson and Dr. Monika Hansson (Karolinska Institutet) for support in antibody validation and characterization. In addition, we thank Lena Israelsson, Eva Jemseby, and Julia Boström for managing the cohorts biobanking and handling of blood samples. We thank Johanna Källström (Uppsala University Hospital) for help with ANA-line blot analysis and Dr. Radha Thyagarajan (Karolinska Institutet) for phagocytosis experiments. We thank Drs. Heidi Wähämaa, Anca Catrina, and Karin Lundberg (Karolinska Institutet) for the CCP2-purified polyclonal ACPA pool. We are grateful to Dr. Scott Coonrod (Cornell University) for generating and generously sharing the PAD KO mice.

Funding Information:
This work was supported by the Swedish Research Council, the Swedish Rheumatism Association, King Gustaf V’s 80-year Foundation, the Region Västra Götaland (ALFGBG-718021), IngaBritt, and Arne Lundbergs Research Foundation, NIH/NIGMS (MSTP T32 GM008244, PT), Rheumatology Research Foundation (Disease Targeted Innovative Research Grant, Within our Reach Award, DM), and the EU/EFPIA Innovative Medicines Initiative (IMI) 2 Joint Undertaking projects BTCure 115142 and RTCure 777357. Funded in part by NIAMS (ZIAAR041199).

Keywords

  • ANA
  • Acetylation
  • Anti-CCP
  • Anti-citrullinated protein autoantibodies
  • Apoptosis
  • Neutrophil extracellular traps (NETs)
  • PAD4
  • Rheumatoid arthritis

Fingerprint Dive into the research topics of 'Differential ACPA binding to nuclear antigens reveals a PAD-independent pathway and a distinct subset of acetylation cross-reactive autoantibodies in rheumatoid arthritis'. Together they form a unique fingerprint.

Cite this