Objectives Mucosal immune therapy with disease-inducing antigens is an effective way to prevent experimental arthritis, but in humans these antigens are unknown. In juvenile idiopathic arthritis, however, T cell recognition of a so-called bystander antigen, heat shock protein 60 (HSP60), is associated with a good prognosis. Recently epitopes derived from HSP60, a microbial peptide (p1) and its self-homologue (p2) were reported to induce tolerogenic T cell responses in vitro in patients with arthritis. A study was undertaken to determine whether mucosal administration of these bystander epitopes can be similarly effective in suppressing arthritis.

Methods Rats were treated nasally with p1, p2 or phosphate-buffered saline before arthritis induction. Arthritis scores were assessed and peptide-specific proliferative responses, phenotypic analysis, cytokine production and in vitro suppressive capacity of cells were measured in lymph nodes and spleens. CD4 spleen T cells from p1- or p2-treated rats were adoptively transferred into naïve rats that were subsequently injected with complete Freund's adjuvant for arthritis induction.

Results Nasal administration of p1 prevented experimental arthritis whereas treatment with the self-homologue p2 did not. Adoptive transfer of CD4 T cells protected against experimental arthritis. Treatment with p1 increased peptide-specific and self-crossreactive interferon γ (IFNγ) production. Tumour necrosis factor α (TNFα) levels were reduced at the site of inflammation. Forkhead box P3 (FoxP3) expression remained stable but the suppressive capacity of T regulatory cells in p1-treated rats was enhanced.

Conclusion p1 immune therapy induces a population of CD4 T cells with reduced TNFα and increased peptide-specific IFNγ production at the site of inflammation. This population expresses FoxP3 and has potent suppressive capacity which, upon transfer, protects against arthritis. The bystander epitope p1 may therefore be a suitable candidate for antigen-specific immunotherapy in arthritis.