1810: Destabilized Treg Cells Predominant in Severe Forms of Juvenile Idiopathic Arthritis

Background/Purpose: T peripheral helper (Tph) cells stimulate excessive B cell responses in the joints of patients with autoantibody-positive arthritis, including seropositive RA in adults and ANA-positive oligoarticular (oligo) JIA in children. Despite this potentially shared disease pathogenesis, a substantial proportion of children with oligo JIA enter long-term remission while this is rare in seropositive arthritis. To better understand these divergent outcomes, we studied the ability of regulatory T (Treg) cells to restrain Tph-B cell interactions in oligo and seropositive polyarticular (poly) JIA.

Methods: Synovial fluid (SF) from children with oligo and RF+ poly JIA (ILAR criteria) and peripheral blood (PB) from healthy controls were studied. Mononuclear cells were stained for surface markers, fixed/permeabilized for FOXP3 staining, and evaluated by flow cytometry. Bisulfite conversion pyrosequencing was done for methylation analysis of cytosine guanine dinucleotide (CpG) sites in the conserved noncoding sequence 2 (CNS2) region of FOXP3 (EpigenDx). SF Tph cells were co-cultured in the presence of SEB for 5 days with memory B cells (Bmem) obtained from controls with/without a given Treg population at a ratio of 1:10:1. The percentage of plasmablasts measured by flow cytometry and total IgG level measured by ELISA served as co-culture readouts.

Results: Table 1 summarizes the study participants. FOXP3, the lineage defining transcription factor for Treg cells, was measured in Treg and T effector (Teff) cells oligo JIA SF. We identified a population of SF Tregs that co-expressed Treg (CD4⁺CD25⁺CD127^loFOXP3⁺) and B cell help (PD1^int) markers, which we termed T peripheral regulatory (Tpr) cells (Fig 1A-B). By contrast, T cells in the Treg gate that were PD1^hi expressed low levels of FOXP3. Similar to PB Tregs, Tpr cells maintained a Treg-specific methylation pattern while PD1^hi T cells did not (Fig 1C-D). Confirming their suppressive capacity, the addition of Tpr cells to Tph-Bmem co-cultures reduced plasmablast differentiation and IgG production (Fig 1E-G). Next, we compared RF+ poly vs. oligo JIA and found that SF Tph cells were significantly more frequent in seropositive patients (Fig 2A-B). In the SF Treg compartment, PD1^hi T cells predominated in RF+ poly JIA patients while PD1^int Tpr cells were more plentiful in oligo JIA patients (Fig 2C-D). While not statistically significant, there was a trend towards reduced plasmablast differentiation in Tph-B cell co-cultures with PD1^intTpr cells compared to PD1^hi T cells from RF+ poly JIA patients (Fig 2E-F).

Conclusion: Tpr cells expressing intermediate levels of PD1 were enriched in the joints of oligo vs. RF+ poly JIA patients and displayed features of stable Treg cells, including robust FOXP3 expression, Treg-specific methylation patterns, and intact suppressive capacity. By contrast, the Treg compartment of seropositive JIA patients was characterized by PD1^hi cells that had low FOXP3 expression, lack the Treg epigenic signature, and may be less suppressive. These findings suggest that Treg fitness and ability to suppress Tph-B cell interactions may play a role in disease course.