The JNK pathway represents a novel target in the treatment of rheumatoid arthritis through the suppression of MMP-3

Background and aim The pathophysiology of rheumatoid arthritis (RA) is characterized by excess production of pro-inflammatory cytokines, including tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6) by neutrophils and macrophages in synovium. Additionally, these cytokines promote the production of reactive oxygen species (ROS), and increased production of matrix metalloproteinases (MMPs), including MMP-3, in synoviocytes that result in joint destruction. There is limited information on how proteolytic enzymes such as MMP-3 can be regulated. We evaluated the effect of the antioxidant N-acetylcysteine (NAC) on RA and identified the relationship between the c-Jun N terminal kinase (JNK) pathway and MMP-3. We hypothesized that elucidating this relationship would lead to novel therapeutic approaches to RA treatment and management. Methods We investigated the effect of administering a low dose (1000 μM or less) of an antioxidant (NAC) to human rheumatoid fibroblast-like synoviocytes (MH7A cells). We also investigated the response of antioxidant genes such as nuclear factor erythroid -derived 2-related factor 2 (Nrf2) and Sequestosome 1 (p62). The influence of MMP-3 expression on the JNK pathway leading to joint destruction and the mechanisms underlying this relationship were investigated through primary dispersion culture cells collected from the synovial membranes of RA patients, consisting of rheumatoid arthritis-fibroblast-like synoviocytes (RA-FLS). Results Low-dose NAC (1000 μM) increased the expression of Nrf2 and phospho-p62 in MH7A cells, activating antioxidant genes, suppressing the expression of MMP-3, and inhibiting the phosphorylation of JNK. ROS, MMP-3 expression, and IL-6 was suppressed by administering 30 μM of SP600125 (a JNK inhibitor) in MH7A cells. Furthermore, the administration of SP600125 (30 μM) to RA-FLS suppressed MMP-3. Conclusions We demonstrated the existence of an MMP-3 suppression mechanism that utilizes the JNK pathway in RA-FLS. We consider that the JNK pathway could be a target for future RA therapies.

In human synoviocytes stimulated with TNFα, high dose of NAC (30 mM) administration suppressed nuclear factor-kappa B (NF-κB ) activation and production of TNFα and IL-6 proteins [10]. High doses of NAC have an anti-inflammatory effect; however, Sadowska et al. indicated that doses of 10 mM and higher are cytotoxic. Zafarrullah et al. found that low doses of NAC (0.1-1 mM) regulated the redox state but doses higher than 10 mM resulted in structural alterations of TGF-β [4,11]. In a clinical setting, caution is required when determining the appropriate NAC dose.
The mitogen-activated kinase (MAPK) signal transduction pathway is associated with ROS activity and a p38 inhibitor suppressed ROS in HeLa cells treated with H 2 O 2 [12]. JNK pathway stimulates inflammatory activity. It was prominently suppressed by administering NAC after hepatic ischemia-reperfusion injury in mice [7]. NAC treatment eliminates ROS and suppresses the JNK pathway and thereby protects granulosa cells from H 2 O 2 -induced apoptosis [13]. NAC affects the activity of MAPKs (mainly JNK); however, human synoviocytes display large individual differences in the expression of interleukins and MMPs. Owing to these differences, human synoviocytes were considered unsuitable for clarifying the mechanisms, including signal transduction and influence of the JNK pathway. For this reason, experiments were carried out on cell lines with less individual variation in the expression of interleukins and MMPs, particularly MH7A cells, which are human fibroblast-like synoviocytes. Our primary objective was to investigate the effect of low-dose NAC on RA. To achieve this, we attempted to confirm that MMP-3 expression is linked to anti-oxidative effects, anti-inflammatory activity, and joint destruction and determine its underlying MAPK signal transduction pathway. Our ultimate goal was to utilize the knowledge of signal transduction pathways to establish novel RA therapies using specific inhibitors.

Cell culture and chemicals
The MH7A rheumatoid fibroblast-like synoviocyte cell line was obtained from the Institute of Physical and Chemical Research (RIKEN, Tsukuba, Ibaraki, Japan). The pelleted MH7A cells were stored at − 80°C and cultivated in RPMI 1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Biowest, Nuaillé, France) and 1% antibiotic/antimycotic solution (Invitrogen, Carlsbad, CA, USA) in a humidified incubator with 95% air and 5% CO 2 at 37°C. After the fifth passage, MH7A cells were seeded into 3.5 cm dishes at a concentration of 3 × 10 5 cells/well and cultured for three days until 80-90% confluency was achieved. These cells were then examined.
Synovial tissues were obtained from two RA patients undergoing total knee arthroplasty and one patient undergoing synovectomy of the wrist. These patients had been diagnosed with RA according to the revised criteria of the American College of rheumatology [14] and had been treated with biologics, methotrexate, prednisolone, immunosuppressants, disease-modifying antirheumatic drugs (DMARDs), and nonsteroidal antiinflammatory drugs (NSAIDs) ( Table 1).
Written informed consent was obtained from each patient before the specimens were collected in accordance with the protocols of the Niigata University Medical and Dental Hospital ethics committee. RA-FLS were isolated using the methods of Rosengren et al. [15] and Sano et al. [16]. Briefly, synovial tissues were cut into small pieces and digested with RPMI 1640 medium mixing collagenase (1 mg/mL) (Worthington Biochemical Corporation, Lakewood, NJ, USA) for 3 h. The tissue was then filtered using a 70 μM nylon cell strainer, washed, and suspended in RPMI 1640 medium. Dissociated cells were then centrifuged at 1500×g for 3 min twice and resuspended in RPMI 1640 medium supplemented with 10% FBS and 1% antibiotic/antimycotic solution. Cells were cultured overnight, the non-adherent cells were removed, and the adherent cells were cultivated in RPMI 1640 medium supplemented with 10% FBS and 1% antibiotic/antimycotic solution. After the fifth passage, RA-FLS were seeded into 3.5 cm dishes at a concentration of 3 × 10 5 cells/well and cultured for 3 days until 80-90% confluency was achieved. These cells were then examined.

Evaluation of cell viability
The effect of NAC on cell viability was determined using the XTT assay (Cell Proliferation Kit II, Roche Diagnostics, Basel, Switzerland), which is based on the reduction of a tetrazolium salt by mitochondrial dehydrogenase in viable cells. Cells were seeded into a 96-well plate at a density of 5 × 10 4 cells/mL and treated with different concentrations of NAC ranging from 10 μM to 10 mM for 24 h at 37°C in 5% CO 2 . Then, 50 μL of XTT stock solution (0.3 mg/mL) was added to each well to attain a total volume of 150 μL. After incubation for 18 h, the optical density (OD) 450-500 was read on a scanning multi-well spectrophotometer (Model 680, Bio-Rad, Hercules, CA, USA).

Western blotting
MH7A cells in 3.5 cm dishes were incubated with medium containing NAC or SP600125 for 3 and 24 h. RA-FLS in 3.5 cm dishes were incubated with medium containing SP600125 for 3 and 24 h. Treated cells were washed with phosphate-buffered saline (PBS) (non-Ca and Mg) and harvested with a cell scraper. To prepare whole cell lysates, cell pellets were extracted with lysis buffer containing 1× Laemmli/urea (62.5 mM Tris, pH 6.8, 2% sodium dodecyl sulfate, 5% glycerol, and 6 M urea) and proteinase inhibitor (4 μL). After measuring the protein concentration in the supernatant using the Pierce TM BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA), the supernatants were mixed with 5% (v/v) 1 M dithiothreitol and 5% (v/v) bromophenol blue and heated at 98°C for 5 min. Equal amounts (50 μg/lane) of proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then electro-transferred onto nitrocellulose membranes. The membranes were incubated with the indicated primary antibodies (IL-6, MMP-3, Nrf2, phosphorylated p62, phosphorylated JNK) and further incubated with secondary G-horseradish peroxidase conjugates (Amersham TM , GE Healthcare, Little Chalfont, UK). Protein bands were visualized by Western blotting detection solution using an enhanced chemiluminescence Western blotting detection solution (Hi-RENDOL®, Hi-RENFIX®, Fujifilm, Tokyo, Japan) and exposing the membranes to Xray film or the protein signals were detected with an ECL system (BioRad, Hercules, CA, USA) and visualized using a charge-coupled device (CCD) cooled camera (Gene Genome; Syngene, Cambrigde, UK). These X-ray film data were digitalized and graphs were individually created using public domain image processing software, ImageJ (U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/).

Evaluation of ROS formation
Quantitative ROS measurements for MH7A cells incubated in medium containing NAC or SP600125 for 3 and 24 h was performed using the Muse™ Oxidative Stress kit (EMD Millipore Bioscience, Billerica, MA, USA). This provided the relative percentages of cells that are ROS-negative and ROS-positive. After treatment with NAC or SP600125, MH7A cells were harvested, incubated with oxidative stress reagent (dihydroethidium), and analyzed on the Muse Cell Analyzer according to the manufacturer's protocol. To measure ROS, a reflection of oxidative stress, MH7A cells were incubated in medium containing NAC or SP600125 for 3 and 24 h each, and H 2 O 2 was added to the medium for 1 h because the intracellular ROS level was highest 1 h after the addition of H 2 O 2 [17,18].

Chemiluminescent enzyme immunoassay
The concentrations of IL-6 in the MH7A cell culture supernatants were measured using a Fully Automated Chemiluminescent Enzyme Immunoassay system (LUMI-PULSE® G1200, Fujirebio, Inc., Tokyo, Japan). IL-6 in

Statistical analysis
All measurements were replicated three or four times, and all values are expressed as the means ± the standard error of the mean (SEM). Statistical analyses were performed with one-way analysis of variance to analyze (ANOVA) followed by Turkey's multiple comparisons test, and twoway ANOVA followed by Dunnett's, Turkey's, or Bonferroni's multiple comparisons test using GraphPad Prism software (GraphPad, Inc., San Diego, CA, USA). P value < .05 was considered statistically significance.

Determination of working concentration of NAC and H 2 O 2 in MH7A cells
To determine the appropriate experimental concentrations of NAC and H 2 O 2 for the experiment, we determined the cytotoxicity of these solutions. At a concentration of 10 mM (10,000 μM) NAC, cell viability of MH7A was 30% after 24 h of treatment ( Fig. 1a(1)).
At a concentration of 1000 μM or lower NAC, cell viability of MH7A was greater than 90%. We, therefore, chose 1000 μM or lower NAC as our working concentration ( Fig. 1.a(2)). On the other hand, H 2 O 2 is typically used at concentrations from 100 to 1500 μM [13,17,18]. MH7A cell lived about only 40% by administering 100 μM of H 2 O 2 for 24 h. We decided to use 100 μM of H 2 O 2 for a brief time (Fig. 1a(3)).
To investigate the change of IL-6 expression under treatment of NAC in MH7A cells, we performed the same experiments as MMP-3 and JNK proteins. Expression of IL-6 protein was not significantly changed at 3 and 24 h after the administering of NAC (10, 100, and 1000 μM) compared with the condition without NAC treatment ( Fig. 2c(1, 2)). H 2 O 2 (100 μM) administration for 24 h significantly increased IL-6 concentration in supernatant of MH-7A cells compared with untreated condition (743 vs. 601 pg/ml, P = .0047). NAC treatment (1000 μM) for 24 h slightly decreased IL-6 concentration compared with the condition under H 2 O 2 (100 μM) but no significant difference was detected (699 vs. 743 pg/ml). These findings indicated that NAC was not able to reduce IL-6  (Fig. 2d).
Next, we measured the rate of ROS positive cells to clarify whether JNK inhibitor has an antioxidative effect like NAC.
These findings indicated that SP100625 was able to reduce IL-6 concentration in supernatants in MH-7A cells (Fig. 3c).

Discussion
We found that low doses of NAC (1000 μM) and SP600125 (15 and 30 μM) were effective in suppressing the production of the proteolytic enzyme MMP-3, which, through suppression of JNK pathway component phosphorylation, causes joint destruction. The concentration of NAC required to suppress all proinflammatory cytokines and NF-κB is at least 5 mM. We determined that after 24 h of treatment, 10 mM NAC is cytotoxic to MH7A cells with only 30% of MH7A cells remaining viable. We utilized a 1000 μM concentration of NAC, which was cytotoxic to 10% or less of the MH7A cells in our sample. Low-dose NAC can reduce H 2 O 2 -induced ROS and increases in Nrf2 and p62 expression, which have antioxidant effects, and induce antioxidant genes (Fig. 1b, c, d) [21][22][23][24][25][26]. The expression of NF-κB is related to IL-6 production. Fujisawa et al. reported that low-dose NAC (1000 μM) is unable to suppress the transcriptional activity of NF-κB, which was consistent with our findings (Fig. 2c, d) [10]. However, we found that low-dose NAC (1000 μM) could significantly suppress the phosphorylation of JNK and downstream MMP-3 protein expression (Fig. 2a, b). These findings confirm that low-dose NAC is linked to the suppression of JNK phosphorylation and MMP-3 expression.
Findings acquired in our study demonstrated that low-dose NAC suppressed MMP-3 and ROS, but did not inhibit IL-6 and that SP600125 suppressed all the three MMP-3, ROS, and IL-6 in MH-7A cells. In particular, we confirmed that SP600125 (30 μM) had an antioxidant, anti-inflammatory effect that suppressed ROS and inhibited IL-6 production (Fig. 3b, c, and S1). Experiments using RA-FLS obtained through the primary dispersion culture of synoviocytes taken from RA patients found that MMP-3 was significantly suppressed at 3 and 24 h following the administering of SP600125 (30 μM; Fig. 4b), which was similar results in the case of MH7A cells. We believe that this indicates the presence of an MMP-3 suppression mechanism that utilizes the JNK pathway in RA-FLS.
We finally considered the explainable mechanism of MMP-3 suppression via JNK pathway by low-dose NAC and JNK inhibitor (Fig. 5). MMP-3 expression is regulated by JNK pathway. Once ROS activate JNK-pathway, phosphorylated JNK activates nuclear transcription factor AP-1 and MMP-3 protein production is promoted. Low-dose NAC or JNK inhibitor (SP600125) inhibits ROS production itself and specifically inhibits phosphorylation of JNK protein so MMP-3 protein production is prominently suppressed.

Conclusions
MMP-3, which causes IL-6 production and joint destruction in RA patients, is produced by MH7A cells. In MH7A cells, NAC reduces ROS and MMP-3. A JNK inhibitor reduces ROS production, decreases IL-6, and downregulates MMP-3. MMP-3 is also reduced in human synoviocytes collected from RA patients following the JNK inhibitor treatment. We believe that JNK pathway will be a novel therapeutic target for the treatment of RA.