Background Identification of autoantigens has been a great challenge in autoimmune diseases such as rheumatoid arthritis (RA). To address this issue, we have developed a new method that combines discovery-based proteomics to identify HLA-DR-presented peptides in patients' synovial tissue [1], joint fluid or peripheral blood (PB); and translational research to determine the immunogenicity of the identified peptides and their source proteins. With this approach, we recently identified endothelial cell growth factor as a target of T and B cell responses in Lyme arthritis (LA) [2] and annexin A2 as such a target in both LA and RA [3].

Objectives To identify novel disease-associated autoantigens which induce T and B cell responses in patients with RA.

Methods HLA-DR-presented self-peptides were isolated and identified from RA patients' tissue or fluids by tandem mass spectrometry, synthesized and tested for T cell autoreactivity with the matching patient's peripheral blood mononuclear cells (PBMC). Immunoreactive peptides or their source proteins were then tested for T cell autoreactivity by IFN-g ELISpot assay and for autoantibody responses by ELISA using cells and sera from our large cohort of RA patients and control subjects. All RA patients met the 2010 ACR/EULAR criteria for RA.

Results In the RA patient presented here, 1 of 89 HLA-DR-presented peptides identified from her synovial tissue and 1 of 15 HLA-DR-presented peptides identified from her PB induced her PBMC to secrete IFN-g. These reactive peptides were derived from the proteins N-acetylglucosamine-6-sulfatase (GNS), a lysosomal enzyme, and from filamin A (FLNA), an actin-binding protein, respectively. We then found that 8 of 25 RA patients (32%) had T cell reactivity with GNS that was >3 SD above the mean value of 10 healthy controls (HC) (p=0.004), and 13 of the 25 patients (52%) had T cell reactivity to FLNA (p=0.002). In addition, 35% of 92 RA patients had elevated autoantibody responses to GNS that were >3 SD above the mean value of 50 HC (p<0.0001), and 32% of the patients had high autoantibody responses to FLNA (p<0.0001). Of the 92 patients, 52 (56%) had a response to one of these autoantigens, and 10 (11%), including the case patient, had reactivity with both autoantigens. Compared with patients with RA, none of 94 patients with other rheumatic diseases, including systemic lupus (SLE), psoriatic arthritis, spondyloarthritis, Lyme arthritis, or osteoarthritis had B cell reactivity with GNS (in each instance, p<0.0001), and only 3 patients with SLE (p=0.003) and 1 with psoriatic arthritis (p<0.0001) had responses to FLNA that were slightly above the cut-off value.

Conclusions This approach is an effective way to identify novel autoantigens associated with autoimmune forms of arthritis. With this method, we identified two previously unrecognized autoantigens, each of which is a target of T and B cell responses in about 1/3 of patients with RA, but rarely in those with other types of arthritis. These findings suggest that testing for these newly recognized autoantibodies may have diagnostic utility in RA.

References

1. Seward RJ et al., Mol Cell Proteomics, 2011.

2. Drouin EE et al., Arthritis Rheum, 2013.

3. Pianta A et al., Arthritis Rheum, 2014 66:S437.

Disclosure of Interest None declared