Recent evidence suggests an important role for IL-7 in the pathogenesis of rheumatoid arthritis (RA) and a therapeutic potential for IL-7 blockade [1]. IL-7 is produced by synovial stromal cells (StrC) in RA but is barely detectable in osteoarthritis [2]. Our published data showed reduced production of IL-7 by bone marrow (BM) StrC and reduced levels of circulating IL-7 in RA compared with health [3]. IL-7 was also detected in synovial tissue in active RA but levels were below detection in clinical remission. These conflicting observations suggest that the regulation of IL-7 expression is tightly controlled at the level of tissue specificity. To support this hypothesis, we showed that several cytokines have a different effect on IL-7 production in BM StrC, epithelial cells from the liver and gut. This work aims at identifying the factors that regulate IL-7 production in StrC isolated from synovial tissue of RA patients.

Synovial fluid and tissue biopsy were obtained from RA (n = 6) and osteoarthritis (n = 4) patients (documented with inflammation scores, visual analogue score) and BM aspirates from healthy donors (n = 6). Skin fibroblasts were used as negative control. We quantified IL-7 in freshly isolated StC (using an advanced cell-sorting strategy [4]) from healthy BM and showed that IL-7 expression is high. However expression declined rapidly with tissue culture (20-fold by passage 2), suggesting that environmental factors are required to sustain IL-7 expression in vitro. IL-7 expression was threefold lower in expanded StrC from RA synovial tissue compared with healthy BM; however, levels were also directly correlated with a visual analogue score of inflammation performed during arthroscopy (ρ = 0.930, P < 0.001).

IL-7 expression is regulated by several cytokines. IFNγ induced IL-7 by ninefold in health but this regulation is lost in RA in direct relationship with levels of inflammation (ρ = 0.842, P < 0.001). In contrast, TNFα and IL-1β induced IL-7 by an average of threefold and 15-fold, respectively, in RA but had no effect on healthy BM StrC. The effect of these cytokines was again correlated with exposure to inflammation (ρ = 0.883, P < 0.001 for TNFα and ρ = 0.964, P < 0.001 for IL-1β). TGFβ₁ reduced IL-7 expression in health (fivefold) but had no effect in RA. Therefore, the promoter of IL-7 is only sensitive to regulation by other cytokine at low levels of exposure to inflammation. In contrast, IL-7 expression is maximal at high levels of inflammation and cannot be further increased by other cytokines.

The expression of IL-7 is tightly controlled in health, the accessibility of the IL-7 promoter to regulation by other cytokines probably being restricted by tissue specificity. In contrast, in RA joints this regulation is abrogated in direct relation with exposure to inflammation. The specific effect of cytokine is therefore only observed when inflammation is low, suggesting promiscuity on the promoter of the IL-7 gene. This may be resulting from an alteration of the IL-7 promoter at the epigenetic level possibly through the effect of Line 1-mediated hypomethylation of chromatin [5].