A total of 60 2-month-old male Sprague-Dawley (SD) rats (180±20 g) were obtained from the Laboratory Animal Center of Anhui Medical University (Hefei, China). The present study was approved by the Medical Ethics Committee of the Academic Committee at Anhui Medical University. The animals were acclimatized in a room at constant temperature (22–23°C) under a 12-h light-dark cycle (light on at 7:00 AM), and the relative humidity was controlled within the range 50–70%. The rats were provided with access to standard mice chow and water ad libitum. Aliquots of 150 µl Freund's complete adjuvant (FCA; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) were injected into the left hind toe of the male SD rats once, and evident arthritis was observed 20 days later. Control rats were injected with 150 µl physiological saline in the identical region of the limb. Subsequently, the AA model rats were randomly divided into five groups that were respectively treated with 0, 5, 15 or 45 mg/kg resveratrol and 200 mg/kg N-acetyl-L-cysteine (NAC; Merck KGaA) for 12 days by continuous intragastric administration. The selection of the doses used in the present study were based on a study described previously (23). All animals were sacrificed on the 12th day upon completion of the above-mentioned treatments. Blood samples were centrifuged at 1,500 × g for 20 min at −4°C to obtain the supernatant fluid, which was stored at −80°C prior to further analysis. The degree of paw swelling and the arthritis index scores were determined to evaluate the severity of AA. Following the injection of the FCA emulsion, the hind paw volume (ml) of all rats was measured using a plethysmometer at 3-day intervals. The arthritis index was classified using a four-points scale (0 points, red spots or mild swelling; 2 points, moderate joint swelling; 3 points, severe joint swelling; 4 points, joint rigidity, deformity or severe dysfunction), totaling 16 points for each rat. The degree of paw swelling was also used to evaluate the severity of the lesion in the affected limb. Prior to the induction of inflammation, an animal volume detector (Jinan Yanyi Biotechnology Corporation, Jinan, China) was applied to detect the affected hind paw volume of each rat (F0). From the onset of inflammation, the left hind paw volume was measured every 4 days (F1), and the degree of paw swelling of the affected paw was calculated according to F1/F0.

FLSs were isolated from the rats in the AA model group as described previously (24). Sterile synovial tissue samples were separated into 1-mm³-sized pieces, mixed with twice the volume of 0.2% type II collagenase (Merck KGaA) containing 10% Gibco^® fetal bovine serum (FBS; Thermo Fisher Scientific, Inc., Waltham, MA, USA), and the tissues were digested for 2–2.5 h at 25°C (percussing the samples once every 30 min). Finally, 0.25% trypsin was added for a further digestion step for 30 min, and the isolated FLSs were cultured in Gibco^® Dulbecco's modified Eagle medium (Thermo Fisher Scientific, Inc.) supplemented with 15% FBS in an incubator at 37°C with 5% CO₂ prior to performing the following experiments.

Blood from the AA rats was obtained via removal of the eyes, and serum was isolated via centrifugation (2,000 × g) at 4°C for 20 min. Serum levels of IL-1, IL-6, IL-8, IL10 and TNF-α were detected using ELISA kits (all Abcam, Cambridge, MA, USA) for mouse IL-1 (cat. no. ab100704), IL-6 (cat. no. ab100712), IL-8 (cat. no. ab46032), IL10 (cat. no. ab108870) and TNF-α (cat. no. ab208348) according to the manufacturer's protocol. Cell apoptosis was detected using a terminal deoxynucleotidyl-transferase-mediated dUTP nick end labeling (TUNEL) assay (Roche Diagnostics, Indianapolis, IN, USA) in situ cell death detection kit (cat. no. 11684817910), according to the manufacturer's protocol. Briefly, FLSs were washed 3 times with phosphate-buffered saline (PBS) and incubated with reaction buffer at 37°C for 30 min in the dark. FLS were then stained with DAPI at room temperature for 5 min in the dark to visualize the nuclei, following which slices were mounted (cat. no. ab64230; Abcam) at room temperature for ~5 min. The number of apoptotic cells, and the total numbers of cells, were counted from five random fields in each slide under a light microscope (magnification, ×200). The results are presented as the ratio of the apoptotic cell number to the total cell number (n=3 for each group) (25).

Apoptosis-associated proteins, including Bcl-2, caspase-3, Bax, caspase-12 and Chop, were detected via western blot analysis according to a protocol described previously (23). Briefly, cells were lysed using RIPA buffer (Beyotime Institute of Technology, Shanghai, China) and centrifuged at 12,000 × g for 10 min at 4°C. The protein concentration was determined using a BCA Protein Assay kit (cat. no. P0011; Beyotime Institute of Technology). The supernatants were degenerated via heating at 100°C for 5 min with 1/5 volume of loading buffer (Beyotime Institute of Technology), and 30 µg samples were loaded in each well, separated via 10% SDS-PAGE and transferred to PVDF membranes (EMD Millipore, Billicera, MA, USA). Next, 5% nonfat milk in washing buffer was used to block the PVDF membranes for 2 h at room temperature, which were then incubated at 4°C overnight with primary antibodies (all from Abcam) specific for Bax (1:1,500; cat. no. ab32503), Bcl-2 (1:2,000; cat. no. ab182858), caspase-3 (1:2,000; cat. no. ab13847), caspase-12 (1:2,500; cat. no. ab62484), Chop (1:3,000; cat. no. ab179823) and β-actin (1:10,000; cat. no. ab115777). On the following day, membranes were washed and incubated with horseradish peroxidase-conjugated anti-rabbit Immunoglobulin G secondary antibody (1:10,000; cat. no. A0208; Beyotime Institute of Biotechnology) for 1 h at room temperature. Finally, protein bands were visualized using enhanced chemiluminescence reagent (Boster Biological Technology, Pleasanton, CA, USA), imaged using a Gel-Dox XR+ imager (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and quantified using Image Lab v4.0 (Bio-Rad Laboratories, Inc.).

2′,7′-Dichlorodihydrofluorescein diacetate (DCFH-DA) was utilized as the fluorescent probe to detect intracellular ROS in the FLSs, also according to the manufacturer's protocol (Bioluminor Biotechnology Co., Ltd., Xiamen, Fujian, China). Briefly, 1×10⁵/cm² FLSs grown on glass coverslips were cultured with 5 µM DCFH-DA at 37°C for 20 min. Following staining, the slides were rinsed three times with PBS. Subsequently, the fluorescence intensity was determined using a confocal laser scanning microscope (Leica SP5-DMI6000-DIC; Leica Microsystems GmbH, Wetzlar, Germany); its in-built evaluation software (Leica LAS AF Lite 2.6.0 build 7288) was used to detect the quantitative analysis of the green fluorescence signal with an excitation wavelength of 488 nm and an emission wavelength of 522 nm. The ROS level was positively correlated with fluorescent intensity.

Following corresponding animal experiments, the rats were sacrificed via cervical dislocation. The knee joint was extracted and fixed in 4% paraformaldehyde at 4°C. Subsequently, the tissues were dehydrated in ethanol and finally embedded in paraffin. Histologic cuts from the paraffin blocks (5-mm thickness) were obtained and stained with hematoxylin and eosin as previously described (22). Briefly, the sections were attached to the slide and heated at 60°C. Next, the sections were dewaxed with xylene twice for 5 min, rehydrated in a graded ethanol series and then distilled water for 3 min, and then stained with hematoxylin for 5 min and with eosin for 20 sec (both steps at room temperature). Sections were then dehydrated in a graded ethanol series and subsequently incubated with xylene for 5 min. Finally, sections were sealed with neutral paraffin. The images of the stained tissue were captured via a light microscope (magnification, ×100 and ×400). A total of six fields were randomly selected from each group, and the fields of view were analyzed by two different observers.

According to the manufacturer's protocol, a Δψm assay kit with JC-1 (Thermo Fisher Scientific, Inc.) was used to detect Δψm. In terms of the functioning of the assay, the JC-1 probe accumulates in the mitochondrial matrix to form a polymer, which results in marked red fluorescence in normal mitochondria; however, the JC-1 probe exists as a monomer in damaged mitochondria, and appears as marked green fluorescence. Following the specific experimental treatments with resveratrol and H₂O₂, 1×10⁵/cm² FLSs, grown on glass coverslips, were cultured with 5 µg/ml JC-1 at 37°C for 20 min. Following staining, the slides were rinsed three times with PBS. As described above, the fluorescence intensity was subsequently determined using a Leica SP5-DMI6000-DIC confocal laser scanning microscope (Leica Microsystems GmbH), and its in-built evaluation software (Leica LAS AF Lite) was used to detect the quantitative analysis of the green fluorescence signal, with an excitation wavelength of 507 nm and an emission wavelength of 529 nm. Measurements for Δψm were determined as the ratios of red-to-green fluorescence intensity.

The measurement of [Ca²⁺]_i was performed as previously described (26). Briefly, 1×10⁵/cm² FLSs were seeded on circular coverslips and incubated with 10 µmol/l Fluo-8 combined with 0.02% pluronic acid F-127 at 37°C for 20 min. Ca²⁺ release was triggered via treating FLSs with 4 µmol/l thapsigargin (TG) or 100 µmol/l ATP for 5 min in Ca²⁺-free PBS containing 140 mmol/l NaCl, 5 mmol/l KCl, 1 mmol/l MgCl₂, 10 mmol/l glucose, 0.2 mmol/l EGTA and 5 mmol/l 4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid (pH 7.4). Real-time fluctuations in fluorescence representing [Ca²⁺]_i (the internal concentration of Ca²⁺) were recorded using a Leica SP5-DMI6000-DIC confocal laser scanning microscope (Leica Microsystems GmbH) with an excitation wavelength of 488 nm and a long-pass emission wavelength of 515 nm. Variations in [Ca²⁺]_i are shown as the ratio of fluorescence relative to the intensity prior to the administration of TG or ATP (F1/F0). The fluorescence intensities were measured, based on an average of 20–30 cells for each measurement.

The results were analyzed using SPSS 19.0 statistical software (IBM Corp., Armonk, NY, USA), and data are shown as the mean ± standard deviation of experiments performed in triplicate. All datasets were analyzed using one-way analysis of variance followed by Tukey's post hoc test to compare two groups. P≤0.05 was considered to indicate a statistically significant difference.

Following the injection of FCA in SD rats, marked paw swelling was observed. In order to identify the association between the dose and pharmacological action of resveratrol, three experimental groups, treated with low (5 mg/kg), middle (15 mg/kg) and high (45 mg/kg) doses of resveratrol, were established in the present study (23,27). NAC, the precursor of reduced glutathione, is an important non-enzymatic antioxidant in the cell that eliminates reactive oxygen groups, and has been demonstrated to be effective in suppressing AA and RA. Therefore, NAC was applied as a positive control in order to examine the functional role of resveratrol (28–30). Compared with the AA model group, treatment with resveratrol markedly reduced paw swelling and the arthritis score in AA rats (Fig. 1A and B). In addition, the effects of resveratrol on AA rats were further demonstrated by H&E staining. FCA injection triggered mass mononuclear cell infiltration, and hyperplasia of synovial tissue. Upon microscopic examination, eroded cartilage and thickened synovial tissue were clearly observed, with a large number of fragments gathered in the synovial cavity of the AA model group. Following treatment with resveratrol, the synovial samples from rats revealed evidently attenuated inflammatory cell infiltration and synovial hyperplasia; however, all the above symptoms were relieved, or even eliminated, compared with the AA model group (Fig. 1C).

The rats in the respective groups were fed for 27 days, and subsequently received intragastric administration of 5, 15 or 45 mg/kg resveratrol and 200 mg/kg NAC for 12 days. Subsequently, the serum levels of IL-1, IL-6, IL-8, IL-10 and TNF-α were detected using ELISA assay. Administration of resveratrol led to a marginal decrease in the levels of IL-1, IL-6, IL-8 and TNF-α, although the level of IL-10 in AA rats was increased compared with that in the non-immunized controls in a dose-dependent manner. This indicated that resveratrol was able to reduce the level of inflammatory injury in the AA rats (Fig. 2A-E). Among the three experimental groups treated with resveratrol, administration of 45 mg/kg resveratrol exhibited the most marked suppression, which approached the level of normal controls.

Considering the anti-inflammatory and anticancer effects of resveratrol reported previously in the literature, the hypothesis of the present study was that resveratrol triggers the apoptosis of FLSs. In order to test this hypothesis, a TUNEL assay was used to examine the apoptotic rate of the FLSs. As shown in Fig. 2F and G, no evident differences in the apoptotic rates were observed when comparing between the normal controls and FLSs; only when the cells were incubated with 5 µM H₂O₂ did treatment with resveratrol lead to an evident increase in the apoptotic rate, and this increase occurred in a dose-dependent manner. Subsequently, DCFH-DA was utilized to assess the levels of intracellular ROS, and these results demonstrated that resveratrol led to an increase in the level of intracellular ROS in FLSs in a dose-dependent manner, and therefore it may be potential marker for apoptosis and ER stress (Fig. 2H).

As resveratrol induced the apoptosis of FLSs in an environment containing 5 µM H₂O₂, further elucidation of the intracellular mechanism became the subsequent objective. It has been well established that the mitochondrial signaling pathway has a vital role in cell apoptosis (31). In order to examine whether the mitochondrial signaling pathway is involved in resveratrol-induced apoptosis, the expression levels of mitochondria-associated apoptotic proteins were initially examined. The immunoblotting results suggested that, in the presence of 5 µM H₂O₂ treatment, resveratrol led to a notable increase in the expression levels of pro-apoptotic proteins, including Bax and active caspase-3, whereas the anti-apoptotic protein, Bcl-2, was evidently suppressed compared with that in normal control cells and FLSs treated only with 5 µM H₂O₂ (Fig. 3C-F).

A reduction in Δψm or enhanced mitochondrial membrane depolarization have been demonstrated to provide an early sign of cell apoptosis, even in advance of DNA damage (32). JC-1, a widely used fluorescent probe to detect Δψm, forms aggregates in the mitochondrial matrix that yield red florescence when the value of Δψm is relatively high, whereas it forms monomers, giving rise to green florescence, when the value of Δψm is comparatively low (33). As shown in Fig. 3A and B, resveratrol was able to suppress the ratio of the red-to-green fluorescent intensity in FLSs in the presence of 5 µM H₂O₂ in a dose-dependent manner compared with normal control cells and FLSs treated only with 5 µM H₂O₂, indicating that resveratrol was able to lead to a reduction in Δψm, which further confirmed the pro-apoptotic effects of resveratrol.

Apart from the mitochondria-associated apoptotic pathway, the ER, which is known as an intracellular Ca²⁺ store, also serves a crucial role in cell apoptosis. Previous reports have indicated that the depletion of ER Ca²⁺ stores may trigger cell apoptosis and growth arrest (34,35), and the hypothesis of the present study was that resveratrol may affect the depletion of the Ca²⁺ store. As Ca²⁺ stores may be depleted by both ATP and TG, the effect of resveratrol on TG- and ATP-induced Ca²⁺ release from the ER Ca²⁺ stores was subsequently examined (36,37). Unexpectedly, the calcium imaging experiments revealed that higher doses of resveratrol led to a suppression of TG- and ATP-induced Ca²⁺ release in FLSs in an environment containing 5 µM H₂O₂ (Fig. 4A-D). By contrast, continuous and heavily applied ER stress has been suggested to be another possible means of inducing cell apoptosis (38). Two essential terminal apoptosis-associated proteins located downstream of the ER stress pathway, Chop and caspase-12, have been revealed to be elevated under aggravated ER stress (39). In the present study, the immunoblotting data obtained revealed that resveratrol led to an increase in the expression levels of Chop and caspase-12 in FLSs in the presence of 5 µM H₂O₂, in a dose-dependent manner, compared with levels in normal control cells and FLSs treated only with 5 µM H₂O₂ (Fig. 4E-G).