Article Text

OP0199 Immune Responses To Prevotella copri in Patients with Rheumatoid Arthritis
  1. A. Pianta1,
  2. E.E. Drouin1,
  3. S. Arvikar1,
  4. Q. Wang2,
  5. C.E. Costello2,
  6. A.C. Steere1
  1. 1Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Harvard Medical School
  2. 2Center for Biomedical Mass Spectrometry, Boston University School of Medicine, Boston, United States


Background Recently, Scher et al. showed that Prevotella copri, an intestinal microbe, was over-expanded in stool samples from patients with new-onset RA (NORA). However, it has been unclear whether over-expansion of P. copri is simply a marker for dysbiosis in the bowel, or whether the organism itself actually has a pathogenic role in RA.

Objectives To determine whether P. copri may have a pathogenic role in RA, we investigated whether patients had T and/or B cell responses to the organism and looked for P. copri DNA in synovial fluid (SF).

Methods To identify immunogenic proteins of P. copri, we identified microbial HLA-DR-presented peptides (T cell epitopes) in patients' synovial tissue, SF or peripheral blood (PB), using tandem mass spectrometry. The identified peptides were synthesized and tested for T cell reactivity with the matching patient's PBMC by IFN-g ELISpot assay. Immunoreactive peptides or source proteins were then tested for T and B cell responses using cells and sera from our cohort of DMARD-naive NORA patients, chronic RA (CRA) patients, and controls. All RA patients met the 2010 ACR/EULAR criteria for RA. Available serum and SF samples were analyzed for the presence of 16S rDNA of P. copri by nested PCR.

Results From the HLA-DR-presented peptides isolated from the PB of one RA patient (HLA-DRB1*0401/0101), we identified an immunogenic T cell epitope derived from bacterial signal sequence of a 27-kDa protein of P. copri (Pc-p27), implying that this organism or components of it were present in this patient's circulating phagocytic cells. When testing was done in additional patients, 24 of 40 (60%) NORA patients had T cell reactivity with epitopes of Pc-p27 compared with none of 10 patients with Lyme arthritis (P=0.0007) and none of 15 healthy control subjects (P<0.0001). In addition, 10 of 78 NORA patients (13%) and 10 of 49 CRA patients (20%) had IgG antibodies to Pc-p27; 8 of 78 NORA patients (10%) and 6 of 49 CRA patients had IgA antibodies to this protein. Only 2 patients had both IgG and IgA antibody responses. Overall, 31 of 127 NORA or CRA patients (24%) had IgG or IgA antibody responses to P. copri, whereas low-level antibody responses to the organism were found in only 2 of 131 patients with Lyme arthritis, spondylarthropathy or lupus (P<0.0001) and in 2 of 50 healthy control subjects (P=0.001). Finally, synovial fluid yielded positive results for P. copri 16S rDNA in 3 of 5 RA patients with IgG antibody responses to P. copri, but in none of 13 patients with IgA P. copri antibodies or no response to the organism (P=0.01).

Conclusions A subgroup of RA patients had P. copri-specific IgA or IgG antibody responses early or late in the disease, whereas patients with other types of arthritis rarely did, implying specificity of these responses in RA. IgA responses may occur because of interactions between host and immune cells in the gut mucosa, whereas IgG responses may be a sign of systemic spread of the organism to joints. These observations suggest that P. copri is more than a bystander in gut dysbiosis, and implies that the organism itself may have a pathogenic role in RA patients, creating a new paradigm to consider in RA pathogenesis.

  1. Scher J. U. et al, Elife 2013

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.