Materials and methods

Cell culture
All in vitro experiments were carried out with fibroblast-like synoviocytes (FLSs) derived from patients with RA. After informed consent was obtained, synovial tissues were collected from RA patients. They met the 1987 American College of Rheumatology criteria for the diagnosis of RA and had been treated with nonbiological disease-modifying antirheumatic drugs and had undergone therapeutic joint surgery. FLSs were isolated and grown in Dulbecco's modified essential medium (low glucose) (Gibco-BRL, now part of Invitrogen Corporation, Carlsbad, CA, USA) supplemented with 10% (vol/vol) fetal bovine serum (Invitrogen Corporation) and 1× Antibiotic-Antimycotic (Invitrogen Corporation) as described previously [19]. After the cells had grown to confluence, they were split at a 1:4 ratio. FLS passages 3 to 6 from three patients were used for all experiments. Ethical permission was obtained from the Institutional Review Board for Human Research of Kyung Hee University, Neo-Medical Center.

Measurement of gene expression by enzyme-linked immunosorbent assay
Synovial cells (2.5 × 105 cells per 60-mm dish per 2 mL of serum-free media) were treated with recombinant adiponectin (1 or 10 μg/mL) or IL-1β (0.1 or 1 ng/mL) (ProSpec, Rehovot, Israel). Conditioned media was collected 24 hours later. Briefly, FLS cultures were centrifuged and the supernatants were collected and analyzed for IL-6, IL-8, PGE2, VEGF, MMP-1, and MMP-13 with an enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Inc., Minneapolis, MN, USA). Three independent experiments were performed in quadruplicate. Each experiment was performed using synovial cells from different patients. For the assessment of MMP-1, MMP-13, VEGF, and adiponectin levels in joint fluid, the collected joint fluid from 30 patients with RA or OA was dispensed into 1-mL aliquots and treated with hyaluronadase at 50 μg/mL for 1 hour at room temperature. The joint fluid was diluted with diluent's buffer for the proper detection range with ELISA. The levels of proteins of interest in joint fluid were measured using a commercial ELISA kit from R&D Systems, Inc., as described above.

Real-time polymerase chain reaction
Real-time quantitative polymerase chain reaction (PCR) was carried out using the LightCycler PCR system (Roche Diagnostics, Meylan, France) using the DNA binding SYBR Green I dye to detect the PCR products as described previously [19]. A serial dilution was used to generate each standard curve. Each 20-μL reaction mixture contained 1× LightCycler-DNA master SYBR Green I, a specific primer, and 2 μL of cDNA. After 2 minutes of denaturation at 95°C, the MMPs, VEGF, IL-6, IL-8, cyclooxygenase-2 (COX-2), and β-actin underwent 40 reaction cycles at 95°C for 5 seconds, 55°C to 60°C for 10 seconds (annealing), and 72°C for 13 seconds. Product specificity was determined by melting curve analysis as described in the LightCycler manual. The results are calculated as ratios of gene transcripts to β-actin transcripts, with the quantity of transcripts in each sample expressed as a copy number. The ratio of MMPs, VEGF, IL-6, IL-8, and COX-2/β-actin mRNA was assigned a value of 100%, with the corresponding results calculated as relative percentages. The primers were synthesized by Bioneer Co. Ltd. (Seoul, Korea), and their sequences are listed in Table 1.
Table 1  The sequence of polymerase chain reaction primers used in this experiment  bp, base pairs; COX-2, cyclooxygenase-2; IL, interleukin; MMP, matrix metalloproteinase; VEGF, vascular endothelial growth factor.

Western blot analysis
FLSs cultured (2.5 × 105 cells) in 60-mm dishes were serum-starved overnight and stimulated by adiponectin (1 or 10 μg/mL) or IL-1β (0.1 or 1 ng/mL) for 24 hours. The cells were subsequently washed twice in phosphate-buffered saline and treated with 50 μL of lysis buffer (20 mM Tris-Cl [pH 8.0], 150 mM NaCl, 1 mM EDTA [ethylenediaminetetraacetic acid], 1% Triton X-100, 20 μg/mL chymostatin, 2 mM PMSF [phenylmethylsulphonyl fluoride], 10 μM leupeptin, and 1 mM AEBSF [4-(2-aminoethyl)benzenesulfonyl fluoride]). The samples were separated using 12% SDS-PAGE and were then transferred to Hybond-ECL [enhanced chemiluminescence] membranes (Amersham, now part of GE Healthcare, Little Chalfont, Buckinghamshire, UK). The membranes were first blocked with 6% nonfat milk dissolved in TBST buffer (10 mM Tris-Cl [pH 8.0], 150 mM NaCl, 0.05% Tween 20). The blots were then probed with various rabbit polyclonal antibodies for COX-2 and β-actin (Cell Signaling Technology, Inc., Danvers, MA, USA) diluted 1:1,000 in Tris-buffered saline at 4°C overnight and incubated with 1:1,000 dilutions of goat anti-rabbit IgG secondary antibody coupled with horseradish peroxidase. The blots were developed using the ECL method (GE Healthcare). For re-probing, the blots were incubated in the stripping buffer (100 mM 2-mercaptoethanol, 2% SDS, 62.5 mM Tris-HCl [pH 6.7]) at 50°C for 30 minutes with occasional agitation.

Statistical analysis
The in vitro experimental data are expressed as the mean ± standard error of the mean of quadruplicate samples. Differences between groups were assessed by repeated analysis of variance followed by the Dunnett multiple comparison test. The level of MMPs, VEGF, and adiponectin in the joint fluid of RA and OA patients was compared between groups with the unpaired t test. To determine the degree of linearity between two variables, data were compared using the Pearson correlation test (two-tailed). Prism software 4 (GraphPad Software, Inc., San Diego, CA, USA) was used for statistical analysis and graphing. Differences were considered significant at a P value of less than 0.05.