Background Multimeric adiponectin isoforms of high-, middle-, and low-molecular weight (HMW, MMW and LMW) exist in body fluids. Their implication in rheumatoid arthritis (RA) pathogenesis is suggested but not fully confirmed.

Objectives (1) To evaluate adiponectin isoform concentrations in sera and synovial fluids (SF) of RA patients. (2) To search for relationship between adiponectin isoform levels and clinical data. (2) To investigate the effect of adiponectin isoforms on rheumatoid fibroblast-like synoviocytes (FLS) function.

Methods The concentrations of HMW, MMW, LMW isoforms and total adiponectin (T) were measured in sera (n=44) and SF (n=12) of patients with established RA, using ELISA (ALPCO Diagnostic). Body composition was measured by dual-energy X-ray absorptiometry (DXA); total (tFM) and visceral (vFM) fat mass were assessed. Rheumatoid FLS were treated for 18 hrs with human HMW, LMW or LMW+MMW isoforms, applied at 0.1-10 ng/ml. Untreated and cytokine-treated (1 ng/ml of IL-1ß or 10 ng/ml of TNF) cells were used as negative and positive controls, respectively. The concentrations of proinflammatory cytokines (IL-6, IL-8) and connective tissue degrading enzyme, matrix metalloproteinase-3 (MMP-3), were measured in cell culture supernatants by specific ELISA. The Wilcoxon signed-rank test was applied to evaluate the effect of stimuli and to compare adiponectin concentrations in sera and SF. The correlation was assessed using a Spearman test.

Results All adiponectin isoform concentrations were higher in sera than in SF, but in SF the LMW/T and LMW+MMW/T ratios were significantly higher than in sera of the same patients. There was a correlation between tFM/vFM and serum ratios of HMW/T (inverse correlation) and LMW+MMW/T (positive correlation). Moreover, serum HMW/T and LMW+MMW/T ratios showed inverse and positive correlation, respectively, with atherogenic index (AI = total cholesterol/high density lipoprotein-cholesterol, HDL-C). In addition, serum MMW concentrations correlated inversely with disease activity score (DAS₂₈) and ESR (erythrocyte sedimentation rate), a marker of systemic inflammation. In vitro, both HMW and mixture of LMW+MMW significantly up-regulated IL-6, IL-8 and MMP-3 production by FLS, while pure LMW had no effect.

Conclusions In RA patients circulating adiponectin isoform pool is dependent on body FM; obesity/overweight favours low HMW and high LMW+MMW levels. Circulating HMW and LMW+MMW have opposite systemic effects, i.e. anti-atherogenic and proatherogenic, respectively, which is consistent with observations of others. However, the role of adiponectin isoforms in joint inflammation and destructions is more complex. HMW and MMW (more likely than LMW), isoforms may support these pathological processes. Nevertheless, serum high MMW level is associated with less active/inflammatory disease. Thus, we report that HMW and MMW may have some protective, i.e. antiatherogenic (HMW) or anti-inflammatory (MMW) systemic action, but their local effect is disadvantageous (proinflammatory, prodestructive).

Acknowledgements Sponsored by grant No NN402 369938 from the National Science Centre.

Disclosure of Interest None Declared