Six to seven week-old male DBA/1 mice were bought from the Chinese Academy of Sciences (Shanghai, China) and kept under standard conditions (15) at the Tongji University School of Medicine (Shanghai, China). All experimental procedures were approved by the Institutional Animal Care and Use Committee of Tongji University.

Ten newly diagnosed patients with active RA who had not receive any treatment and fulfilled the 2010 American College of Rheumatology/European League Against Rheumatism criteria for RA (16) were recruited at the Shanghai Tenth People’s Hospital of Tongji University (Shanghai, China). Patients with active disease were defined as those with a disease activity score in 28 joints of >3.2. This study was performed with the approval of the Ethics Committee of Tongji University in accordance with the Declaration of Helsinki. Informed consent was obtained from all patients.

CIA was induced in male DBA/1 mice via an intradermal injection of 100 μg bovine type II collagen (CII; Chondrex, Inc., Redmond, WA, USA) emulsified with Complete Freund’s adjuvant containing Mycobacterium tuberculosis (Chondrex, Inc.). Mice were boosted 21 days later with an injection (100 μg bovine type II collagen emulsified in Incomplete Freund’s adjuvant (Chondrex, Redmond, WA, USA).

On day 28 following the first immunization, arthritis was observed in all mice and treatment with PL was initiated. Mice were randomly divided into two groups (n=8 per group) and intraperitoneally injected with 2.4 mg/kg/day PL (Sigma, St. Louis, MO, USA) or a vehicle control (dimethylsulfoxide; DMSO; Sigma, St. Louis, MO, USA) as described by Raj et al (17). PL treatment was continued until day 47.

Clinically, arthritic severity was characterized via the observation of joint properties and the inflammation of the surrounding tissue. The clinical arthritis was assessed every three days and scored as previously described (18). The maximal arthritis score per paw is 4, and the maximal disease score per mouse is 16.

The mouse joints were fixed in 4% paraformaldehyde (Sigma), decalcified in EDTA (Sigma) and embedded in paraffin. Sections (thickness, 7 μm) were prepared and stained with hematoxylin and eosin (H&E) and Safranin O (Promega Corporation, Madison, WI, USA to detect proteoglycans. On day 47 following immunization, the mice were sacrificed under CO₂ anaesthesia and the hind paws were fixed in 10% buffered formalin, decalcified and embedded in paraffin. Joint sections (thickness, 6 μm) were prepared and stained with hematoxylin. Images were captured using an Olympus BX41 microscope (Olympus, Center Valley, PA, USA). The histological scores were evaluated, as previously described (18).

Serum concentrations of anti-CII IgG and IgG2a were measured using an enzyme-linked immunosorbent assay (ELISA). Blood was collected at the end of study, immediately following sacrifice. Microtiter plates were coated with CII (10 μg/ml), blocked and incubated with serially diluted test sera. Bound IgG was detected via incubation with horseradish peroxidase (HRP)-conjugated rat anti-mouse IgG or IgG2a, and tetramethylbenzidine substrate. Absorbance (wavelength, 450 nm) was measured with an ELISA plate reader (3550; Bio-Rad, Hercules, CA, USA), and the values were represented in arbitrary optical density (OD) units.

Serum and supernatant TNF-α, IL-1β, IL-23 and IL-17 were measured using the BD Cytometric Bead Array (CBA) and specific ELISA kits for TNF-α, IL-1β, IL-23 and IL-17 (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions. (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions.

Draining lymph node (DLN) cells from each treatment group were cut and removed following sacrifice and made into a single cell suspension. The cells were washed, incubated with antibodies for fluorescence-activated cell sorting (FACS) and analyzed with a FACS Canto II flow cytometer. The following monoclonal antibodies were obtained from eBiosciences, Inc. (San Diego, CA, USA): Anti-CD11b, anti-Gr-1, anti-CD4 (phycoerythrin; PE), anti-CD25 (fluorescein isothiocyanate; FITC), anti-Foxp3 (PE-cyanine 5) and anti-IL-17 (FITC). Cells were restimulated with 50 ng/ml phorbol 12-myristate 13-acetate and 500 ng/ml ionomycin in the presence of GolgiPlug™ for 5 h followed by intracellular staining. Cells were fixed, permeabilized and stained with fluorochrome-conjugated antibodies. The stained cells were then analyzed using a FACSCalibur or BD FACSAria instrument. Unless otherwise stated, all reagents and equipment was purchased from BD Biosciences.

Synovial tissues were obtained from patients with RA during total knee replacement surgery or arthroscopic synovectomy. The FLS from five patients were treated with PL (100 μg/ml) or a vehicle. Tissues were digested with 4 mg/ml collagenase II (Sigma) in serum-free Dulbecco’s modified Eagle’s medium (DMEM; Gibco-BRL, Grand Island< NY, USA) for ≥4 h at 37°C. Cell suspensions were passed through a nylon mesh and the cell suspensions were collected via centrifugation at 500 × g for 5 min and re-suspended in DMEM supplemented with 10% fetal bovine serum (FBS; Gibco-BRL). Harvested cells were cultured in 75-cm² culture flasks (Costar, Cambridge, MA, USA) with DMEM supplemented with 10% FBS at 37°C in 5% CO₂. Once the cells had grown to confluence they were detached with 0.25% trypsin, split at a ratio of 1:3 and re-cultured in DMEM under the same conditions. Cells obtained from the fourth to sixth passages were used in the following experiments.

In vitro migration assay of FLS chemotaxis was performed using the Boyden chamber method with an 8.0-μm pore size (Transwell; Corning Inc., Corning, NY, USA). DMEM containing 10% FBS as a chemoattractant was placed in the lower wells. FLS (5×10⁴ cells/ml) were suspended in serum-free DMEM in the upper wells. The chamber was incubated at 37°C under 5% CO₂ for 6 h. Following incubation, the non-migrating cells were removed from the upper surface of the filter using a cotton swab. The filters were fixed in methanol for 20 min and stained with 0.1% crystal violet for 20 min. Chemotaxis was quantified by counting the stained cells which had migrated to the lower side of the filter using an optical fluorescence microscope (System Microscope BX53; Olympus, Center Valley, PA, USA)

For the in vitro invasion assay, similar experiments were performed using inserts coated with a Matrigel basement membrane matrix (BD Biosciences, Oxford, UK). To investigate the effect of PL on RA FLS invasion, PL (100 μg/ml) was added to the upper and lower wells. Following treatment with TNF-α (100 ng/ml), cells were observed under a fluorescence microscope (Olympus).

The differences between groups were determined by analysis of variance and comparison between two groups was analyzed with the Student’s t-test using SPSS, version 16.0 (SPSS, Inc., Chicago, IL, USA). Data are presented as the mean ± the standard error of the mean. P<0.05 was considered to indicate a statistically significant difference.

The clinical arthritis was assessed every three days and scored. PL was observed to attenuate the clinical score of CIA (Fig. 1A). In addition, H&E and Safranin O staining were used to detect proteoglycans. It was revealed that the histopathologic damage, including inflammatory cell infiltration and cartilage destruction, was significantly attenuated in PL-treated CIA mice in comparison with that in the vehicle-treated CIA control mice (Fig. 1B and C).

PL significantly reduced the serum anti-CII IgG and IgG2a levels as compared with those of the vehicle-treated CIA control mice (Fig. 2A). In addition, the results revealed that PL inhibits the production of serum TNF-α, IL-1β, IL-23 and IL-17 in CIA mice as compared with that in the vehicle-treated CIA control mice (Fig. 2B).

To evaluate the effect of PL on immune cell responses in vivo, the frequencies of MDSCs, T_regs and Th17 cells were examined in the DLNs of the control and PL-treated CIA mice. The percentage of Th17 cells was markedly decreased in the DLNs of the PL-treated CIA mice compared with that of the vehicle-treated CIA control mice (Fig. 3A and C). Conversely, the number of CD11b⁺Gr-1⁺ MDSCs showed a marked increase in the DLN of the PL-treated CIA mice compared with that of the vehicle-treated CIA control mice (Fig. 3B and C). In addition, the T_regs CD4⁺, CD25⁺ and Foxp3⁺ frequency were not affected by PL treatment (9.8±1.27, vs. 10.1±1.31%; P>0.05).

To determine whether PL affects human RA FLS in a similar way to that observed in mice, the human RA FLS were pretreated with PL for 24 h and then stimulated by human TNF-α (100 ng/ml) for 48 h. The supernatants were then collected. It was revealed that PL reduced the levels of secretion of IL-1β, IL-23 and IL-17 by human RA FLS as detected in the supernatants compared with those in the controls (Fig. 4).

Subsequently, the effect of PL on migration and invasion of human RA FLS was investigated. Treatment with PL substantially inhibited the migration and invasion of TNF-α-induced RA FLS compared with that of the vehicle controls (Fig. 5). These results indicate that PL inhibits the migratory behavior of human RA FLS.