Background/Purpose Regulatory T cell (Treg) dysfunction has been documented in juvenile idiopathic arthritis (JIA), but the basis for this lapse in suppressive capacity is incompletely understood.  Animal models of autoimmunity demonstrate that a diverse and polyclonal Treg repertoire is essential for immune system homeostasis and Treg cell function.  We therefore employed next generation sequencing (NGS) to analyze the repertoires of synovial fluid (SF) and peripheral blood (PB) Treg and effector T cells (Teff) in JIA. 

Methods Paired SF and PB samples were obtained from patients with JIA.  Control samples included PB from age-matched, healthy children and SF from patients with Lyme arthritis.  Treg (CD4⁺CD25⁺CD127^lo) and Teff (CD4⁺CD25^–) cells were isolated from SF and PB mononuclear cells by fluorescence-activated cell sorting.  The T cell receptor (TCR) β chain was amplified by multiplex PCR with genomic DNA serving as the template (ImmunoSEQ^TM).  The PCR products were sequenced using the Illumina HiSeq platform.  Subsequently, the data was analyzed using the ImmunoSEQ^TM set of online tools, the International ImMunoGeneTics system (IMGT) HighV-QUEST platform, and the Immunoglobulin Analysis Tool (IgAT).  The pre-specified primary aim was to evaluate the clonality of SF and PB Treg and Teff populations in JIA patients.  Based on data evaluating the Treg repertoire and autoimmunity in animal models, we calculated that a sample size of 5 JIA patients would provide sufficient power for this study. 

Results  5 patients with JIA, 3 healthy controls, and 2 patients with Lyme arthritis were studied.  In the PB of controls, the Treg and Teff repertoires were equally polyclonal.  In contrast, JIA PB Treg cells were more clonal than control PB Treg, control PB Teff, and JIA PB Teff cells (ANOVA p<0.0001 with Bonferroni correction).  JIA disease severity, as measured by the active joint count, correlated with the PB Treg clonality (ρ=0.95, p=0.005).  Clonal abnormalities were not observed in the JIA PB Teff repertoire.  In SF, JIA Treg cells were significantly more clonal than JIA PB Treg and Teff cells, JIA SF Teff cells, and control PB Treg and Teff cells (ANOVA p<0.0001with Bonferroni correction).  JIA patients shared a substantial portion of SF Treg clonotypes, significantly more than SF Teff cells (Wilcoxon p=0.0002).  In some JIA patient pairs, up to 19% of SF Treg clonotypes were shared.  Further, these shared SF Treg clonotypes were private to JIA patients and were not identified in Lyme arthritis.  In further support of disease specific TCRs in JIA SF, convergent recombination was seen preferentially in the SF Treg compared to the PB Treg compartment (Wilcoxon p=0.002).  Skewed TCRβ variable family and joining gene usage, including overuse of gene segments that have been associated with other autoimmune conditions, was observed in JIA Treg and Teff cells. 

Conclusion Our data identified an unexpected restriction of the SF and PB Treg repertoires in JIA with sharing of SF Treg clonotypes across arthritis patients.  These findings suggest that inadequacy in the Treg repertoire may contribute to Treg cell dysfunction and the perpetuation of inflammation in JIA, possibly in response to shared antigenic triggers.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/next-generation-sequencing-reveals-restriction-of-the-treg-cell-repertoire-and-an-abundance-of-shared-synovial-treg-clonotypes-in-jia/