Background and objectives Adenosine is an important regulator of immune responses. There is accumulating evidence that methotrexate, the anchor drug for rheumatoid arthritis (RA), mediates its anti-inflammatory effects by increasing adenosine levels at inflamed sites. Adenosine modulates (immune) cell responses through four receptors: ADORA1,-2A, -2B, and -3. ADORAs are linked to different intracellular signalling pathways: ADORA1/3 to pro-inflammatory and ADORA2A/2B to anti-inflammatory pathways. Here, we present data investigating the effects of adenosine and receptor expression in two of the major cell types involved in the pathogenesis of RA, monocytes and fibroblast-like synoviocytes (FLS).

Materials and methods Primary monocytes were isolated from healthy donors blood (n = 3) and FLS from synovial biopsies of RA patients (n = 6). After stimulation with LPS or TNF/IL-1β, cells were lysed at several time points and ADORA expression was determined by QPCR. Supernatants were taken after O/N stimulation and cytokines (CCL2, IL-6 and TNF) were determined by ELISA.

Results In resting cells, monocytes and FLS showed a distinct ADORA expression profile, with ADORA1/2B predominant on FLS, and ADORA3/2B predominant on monocytes. After stimulation with LPS or TNF/IL-1β, both cell types showed a striking upregulation of ADORA2A peaking at 4h (monocytes: 90-fold, FLS: 450 fold) but not of ADORA-1, -2B or -3. When monocytes were stimulated with LPS, co-incubation with adenosine significantly reduced pro-inflammatory cytokine production, resulting in a >70% decrease in CCL2 and TNF (p < 0.05). In similar studies using FLS, adenosine was found to have a much more modest effect, showing only a <20% decrease in CCL2 (p < 0.01) and no change in IL-6 production in TNF/IL-1β stimulated cells. In contrast, when no inflammatory stimuli were added to cell cultures, adenosine enhanced CCL2 and IL-6 production in both FLS and monocytes. In RA-FLS, the effect of adenosine was dependent on the dose of TNF/IL-1β. These results indicate that in a pro-inflammatory environment, adenosine will mainly signal through ADORA2A and repress immune responses. Interestingly, in the absence of inflammatory stimuli, adenosine will have pro-inflammatory properties, likely through A1/A3 signalling.

Conclusion These data show that the primary cells involved in the pathogenesis of RA dynamically change their ADORA expression following inflammatory stimulus, and that this change is reflected in the cells response to adenosine.