Session Type: ACR Concurrent Abstract Session
Session Time: 9:00AM-10:30AM
Background/Purpose: Antibody to citrullinated peptides/proteins (ACPA) is specific feature of rheumatoid arthritis (RA) and seen in 50 to 70% of RA patients. Amino acid (AA) position 11/13 in the HLA-DRβ1 is the strongest genetic risk factor in the HLA locus to ACPA(+) RA. We previously showed that HLA-DRB1 alleles interact to confer risk of ACPA(+) RA (Lenz et al Nat Gen 2015). However, genetic architectures underlying individual ACPA are still poorly understood. Here, we investigated the extent to which different individual ACPA are driven by distinct genetic factors within HLA locus.
Methods: We phenotyped RA cases with a multiplex peptide array to quantify 18 fine-specific ACPA as well as 2nd generation cyclic ctrullinated peptide antibody (CCP2), the representative method to quantify ACPA. We queried 6,267 patients and 12,054 controls across 3 independent cohorts. We performed clustering and principal component analysis to identify subgroups of ACPAs. We imputed HLA genotypes and tested for MHC associations with specific ACPA with logistic regression models correcting for 10 principal components and cohort effects. We also evaluate interactive effects of combinations of HLA-DRB1 alleles on fine-specific ACPA(+) RA susceptibility to assess whether allelic interactions influence expression of specific ACPAs.
Results: Clustering and principal component analysis of 18 fine ACPAs and CCP2 demonstrated that ACPAs could be classified into 2 distinct clusters, which we refer to as ACPA-A and ACPA-B. ACPA-A consisted of a total of 12 antibodies including CCP2 and an antibody recognizing a peptide derived from fibrinogen beta AA position 60 to 74 with positions 60, 72 and 74 citurillnated (Fibbeta60-74Ab) which was more strongly correlated with CCP2 than any other ACPA. ACPA-B contained a total of 7 antibodies including Fibbeta62-81cit72Ab, which reacts to a peptide sharing 12 AA sequences to the Fibbeta60-74Ab peptide but differs only in citrullination at one AA position. Position 11 of HLA-DRβ1 protein, which has six possible AA residues, showed the strongest associations in 14/18 ACPAs in intra-case analysis (omnibus p≤1.1×10-33). Intriguingly, Fibbeta62-81cit72Ab showed distinct association patterns of AA residues at the position 11 (p=1.1×10-50) from CCP2 associations. When we examined the subset of 175 case individuals that were positive for ACPA-B, but negative for all ACPA-A antibodies, we observed only weak evidence of association to position 11 (p=0.066), but did observe a significant association with HLA-B AA position 9 (p=1.2×10-7). We identified two novel allelic combinations specifically showing interactive effects on fine-specific ACPAs, for example DRB1*13:01 and 11:01 interact to influence expression of an antibody recognizing a peptide derived from heterogeneous ribonucleotide protein (p=3.1×10-4).
Conclusion: These findings suggest that specific HLA-DRB1 alleles, and their interacting combinations lead to a different repertoire of antibodies, with reactivities to different citrullinated peptide sequences. They suggest that reactivity to multiple antigen sets might independently lead to RA.
To cite this abstract in AMA style:Terao C, Brynedal B, Chen Z, Jiang X, Westerlind H, Hansson M, Mathsson-Alm L, Jakobsson PJ, Skriner K, Serre G, Rönnelid J, Padyukov L, Worthington J, Alfredsson L, Klareskog L, Raychaudhuri S. Specific Antibody Subphenotypes in Rheumatoid Arthritis Are Associated with Unique HLA-DRB1 Residues at Position 11 [abstract]. Arthritis Rheumatol. 2016; 68 (suppl 10). https://acrabstracts.org/abstract/specific-antibody-subphenotypes-in-rheumatoid-arthritis-are-associated-with-unique-hla-drb1-residues-at-position-11/. Accessed December 1, 2020.
« Back to 2016 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/specific-antibody-subphenotypes-in-rheumatoid-arthritis-are-associated-with-unique-hla-drb1-residues-at-position-11/