Chondroprotective effects and mechanisms of resveratrol in advanced glycation end products-stimulated chondrocytes

Polyclonal antisera against iNOS, COX-2 and IκBα were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Polyclonal anti-collagen II antibodies were purchased from Chemicon International (Temecula, CA, USA). The antibodies recognizing ERK, JNK, unphosphorylated IKKα or phospho-IKKα/β were purchased from Cell Signaling (Danvers, MA, USA). Resveratrol (R5010) was purchased from Sigma-Aldrich Chemical Company (St. Louis, MO, USA) and was prepared as a 200 mM stock concentration with ethanol. The required concentrations of resveratrol for individual experiments were made by further dilution of the stock preparation with culture medium when needed. Unless otherwise specified, the rest of the reagents were purchased from Sigma-Aldrich Chemical Company.

Methylglyoxal-modified albumin, as the source of AGEs examined in this study, was prepared as described [17]. In brief, by incubating 10 mg/ml of bovine serum albumin (BSA) (USB Corporation, Cleveland, OH, USA) in phosphate-buffered saline (PBS) containing 50 mM methylglyoxal, 0.1% sodium azide, and 1 mM phenylmethylsulphonyl fluoride (PMSF) at 37°C for seven days, the crude preparation of AGEs was obtained. After dialysis against PBS, the AGEs preparation was filtrated and stored at -80°C until use. The AGEs preparation using methylglyoxal-modified albumin contains N(epsilon)-carboxyethyl-lysine (CEL) and other AGEs [18]. To enhance the significance of this study, the AGEs preparation using glycoaldehyde-modified albumin that contains N(epsilon)-carboxymethyllysine (CML), pentosidine and other AGEs was also prepared according to the report [19, 20].

Porcine cartilage was obtained from the hind leg joints of pigs. The preparation of chondrocytes from cartilage was performed according to our previous report [17]. After enzymatic digestion of articular cartilage with 2 mg/ml protease in serum-free Dulbecco's modified Eagle's medium (DMEM) containing antibiotics and 10% fetal bovine serum (FBS), the specimens were then digested overnight with 2 mg/ml collagenase I and 0.9 mg/ml hyaluronidase in DMEM/antibiotics. The cells were collected, passed through a cell strainer (Beckton Dickinson, Mountain View, CA, USA) and cultured in DMEM containing 10% FBS and antibiotics for three to four days before use.

Evaluation of potential cytotoxic effects of resveratral was performed by using 3-(4,-Dimethylthiazol-2-y)-2,5-diphenyl-tetrazolium bromide (MTT) colorimetric assays as described [21]. In brief, chondrocytes were incubated in the presence or absence of resveratrol for 48 h. Then, 25 μl of MTT (5 mg/ml in H₂O) was added, and cells were incubated at 37°C for 2 h followed by the addition of 100 μl of lysis buffer containing 20% sodium dodecyl sulfate and 50% dimethylformamide. After incubation at 37°C for another 6 h, the content of dissolved reduced MTT crystals was measured with an ELISA reader (Dynatech, Chantilly, VA, USA).

Porcine chondrocytes were seeded one day before the transfection experiment at a density of 3 × 10⁵ cells/6-cm dish. TransIT^®-LT1 transfection reagent (Pan-vera, Madison, WI, USA) was used to co-transfect 1.5 μg/ml of the reporter plasmid pNF-κB-luciferase or pAP-1-luciferase with the internal control plasmid pTK-Renilla-luciferase (Promega, Madison, WI, USA) at a ratio of 100:1 into cells in serum- and antibiotic-free DMEM. Four hours after transfection, the culture medium was replaced with fresh DMEM supplemented with 2% FBS. For luciferase assay, total cell lysates were prepared and the firefly luciferase activity was measured and normalized to Renilla luciferase activity according to the dual luciferase reporter system (Promega).

The measurement of NO release was reflected by determination of its stable end product, nitrite, in supernatants [17]. The Griess reaction was performed with the concentrations of nitrite measured by a spectrophotometer. In brief, an aliquot (100 μl) of culture supernatant was incubated with 50 μl of 0.1% sulfanilamide in 5% phosphoric acid and 50 μl of 0.1% N-1-naphthyl-ethylenediamine dihydrochloride. After 10 minutes of incubation at room temperature, the absorbance was measured at 550 nm wavelength with a plate reader (Tecan, Grodig, Australia). PGE₂ concentrations were determined by ELISA according to the manufacturer's protocol (Cayman, Ann Arbor, MI, USA).

ECL Western blotting (Amersham-Pharmacia, Arlington Heights, IL, USA) was performed as described [22]. Briefly, equal amounts of whole cellular extracts were analyzed on 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to the nitrocellulose filter. For immunoblotting, the nitrocellulose filter was incubated with Tris-buffered saline with 1% Triton X-100 containing 5% non-fat milk for 1 h and then blotted with antibodies against specific proteins for another 2 h at room temperature. After washing with milk buffer, the filter was incubated with rabbit anti-goat IgG or goat anti-rabbit IgG conjugated to horseradish peroxidase at a concentration of 1:5000 for 30 minutes. The filter was incubated with the substrate and then exposed to X-ray film (Kodak, Rochester, NY, USA).

Nuclear extract preparation and EMSA analysis were performed as detailed in our previous report [22]. The oligonucleotides containing NF-κB-binding site, SP-1 binding site or AP-1-binding site were purchased and used as DNA probes. The DNA probes were radio-labeled with (γ-³²P)ATP using the T4 kinase (Promega). For the binding reaction, the radio-labeled pr)be was incubated with 4 μg of nuclear extracts. The binding buffer contained 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 0.5 mM ethylenediaminetetraacetic acid (EDTA), 1 mM dithiothreitol, 1 mM MgCl₂, 4% glycerol and 2 μg poly(dI-dC). The reaction mixture was left at room temperature to proceed with binding reaction for 20 minutes. The final reaction mixture was analyzed in a 6% non-denaturing polyacrylamide gel with 0.5 × Tris/Borate/EDTA as an electrophoresis buffer.

Gelatin zymography was performed according to the description by other researchers [20, 23]. The culture supernatant (16 μl) was mixed with 4 μl buffer containing 4% SDS, 0.15 M Tris (pH 6.8), and 20% glycerol that contains 0.05% bromophenol blue and analyzed on a 10% polyacrylamide gel with copolymerized 0.1% gelatin (Sigma-Aldrich). The positive control of MMP-13 was purchased (EMD Chemicals, Gibbstown, NJ, USA). After electrophoresis, gels were washed with 2.5% Triton X-100 for three times with 20 minutes each. After incubation with the gelatinase buffer (50 mM Tris-HCl (pH 7.6), 10 mM CaCl₂, 50 mM NaCl, and 0.05% Brij-35) at 37°C for 24 h, the gel was stained with 0.1% Coomassie blue, and the clear bands that indicate gelatinolytic activity were visualized under the background of uniform light blue staining. The localization of MMP-13 was identified as judged by the molecular weight of the standards and the report from other researchers [24] under Casio Ex Z60 digital camera (Shibuy-ku, Tokyo, Japan).

This assay has been described in our previous work [22]. In brief, the whole cellular extract 50 to 100 μg was incubated overnight with 5 μl of specific antibodies in incubation buffer containing 25 mM HEPES (pH 7.7), 300 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.1% Triton-X-100, 20 mM β-glycerophosphate, 0.1 mM Na3VO4, 2 μM leupeptin, and 400 μM PMSF. The mixture was then immunoprecipitated by the addition of protein A beads and rotated at 4°C for 2 h. After extensive wash, the beads were resuspended in 40 μl kinase buffer with addition of cold ATP (30 μM), 8 ng of substrate (GST-c-Jun), and 10 μCi of (γ-32P)ATP. The mixture was incubated at 30°C with occasional gentle mixing for 30 minutes. The reaction was then terminated by resuspending in 1% SDS solubilizing buffer and boiling for five minutes; then the samples were analyzed in SDS-PAGE.

Total RNA was isolated after lysing the cells by Trizol (Invitrogen, Carlsbad, CA, USA). After reverse transcription of RNA to cDNA, samples were subjected to PCR reactions. Real-time quantitation of MMP-13 and GAPDH messenger RNA (mRNA) was performed according to the manufacturer's instructions (CYBR Green Master Mix, Applied BioSystems, Foster City, CA, USA). In brief, 10 ng of cDNA was amplified in a total volume of 20 μl consisting of 1× Master Mix and the gene-specific primers, which were added at a final concentration of 100 nM. The primer sequences for both MMP-13 and GAPDH were described by other reports [25, 26]: 5'-CCA AAG GCT ACA ACT TGT TTC TTG -3' and 5'-TGG GTC CTT GGA GTG GTC AA -3'(for MMP-13); 5'-TGT C AT CCA TGA CAA CTC GG -3' and 5'-GCC ACA GTT TCC CAG AGG -3' (for GAPDH). The reactions were performed for 50 cycles with 95°C for denaturation and 60°C for annealing and extension on the ABI 7300 real-time PCR system (Applied BioSystems). The data were collected and the fold changes in gene expression following stimulation in the presence or absence of resveratrol was calculated with the following formula: fold changes = 2 − Δ ( Δ C t ), where C_t = C_{t stimulated} - C_{t GAPDH}, and (C_t) = C_{t stimulated} - C_{t control}.

The preparation of cartilage explants was performed according to our previous report with mild modification [17]. In brief, articular cartilage from the femur head of hind limb joint of pigs was excavated by a stainless-steel dermal-punch 3 mm in diameter (Aesculap, Tuttlingen, Germany) and weighed. After dissection, each cartilage explant was placed in a 24-well plate and cultured for 24 h in DMEM containing antibiotics and 10% FBS. After resting for 72 h in serum-free DMEM, the cartilage explants were used for further study.

Cartilage degradation was assessed by measuring the amount of proteoglycan released into the culture medium as described [17]. In brief, culture medium was added to 1,9-dimethylmethylene blue (DMB) solution (Sigma-Aldrich) in which the metachromatic dye can bind sulfated glycosaminoglycan (GAG), a major component of proteoglycan. The amount of the formation of GAG-DMB complex was measured in a 96-well plate using a plate reader (TECAN, Grödig, Austria) at a wavelength 595 nm. The loss of GAG was calculated and expressed as the total GAG (μg) released per mg (wet weight) of the cartilage weight.

Cartilage explants were mounted in embedding medium (Miles Laboratories, Naperville, IL, USA) and rapidly frozen at -80°C. Serial but incontinuous microscopic sections (7 μm) of cartilage explants were cut at -20°C on a Microm cryostat and mounted on Superfrost Plus glass slides (Menzel-Gläser, Braunschweig, Germany). Tissue sections were then stained with Safranin O/fast green countered with Weigert's iron hematoxylin to assess the changes of proteoglycan content [17]. In parallel, the expression of aggrecan NITEGE neoepitope recognized by NITEGE antibodies in tissue sections was determined as described by other researchers [27].

Because both iNOS-NO and COX-2-PGE₂ pathways are critical in AGEs-induced cartilage damage [17], we determined whether resveratrol could provide any protective effect through regulating these pathways. The AGEs preparation using methylglyoxal-modified albumin that contains CEL and other AGEs was chosen as the stimulant [18]. When chondrocytes were pretreated with various concentrations of resveratrol, the AGEs-induced expression of iNOS was suppressed (Figure 1a). Meanwhile, resveratrol also inhibited AGEs-induced COX-2 expression. The concentrations of NO and PGE₂, both downstream products of iNOS and COX-2, respectively, were measured. The results show that resveratrol inhibited AGEs-induced NO and PGE₂ production (Figure 1b, c). Importantly, resveratrol at a low concentration 25 μM was enough to inhibit PGE₂ production (Figure 1c). By both MTT and LDH release assays, we confirmed that at the examined concentrations, resveratrol did not have detectable cytotoxic effects in chondrocytes (Figure 1d and data not shown).

We next investigated the mechanisms of resveratrol-mediated inhibition of iNOS and COX-2 in AGEs-activated chondrocytes. We focused on examining the activity of NF-κB and AP-1, two families of transcription factors that regulate the activation of iNOS and COX-2 genes and are critical in many inflammatory responses [28, 29]. Pretreatment with resveratrol resulted in a dose-dependent inhibition of AGEs-induced DNA-binding activity of NF-κB (Figure 2a). Because the process of IκBα degradation proceeds before NF-κB activation, we examined whether resveratrol had any effect on IκBα degradation. The results show that AGEs treatment caused degradation of IκBα with the effect observed 2 h after treatment (Figure 2b). In the presence of resveratrol, there was dose-dependent reversal of the IκBα levels to a certain extent (Figure 2c). Since degradation of IκBα requires its phosphorylation by IκB kinases, we evaluated whether resveratrol could inhibit IKKα/β activities. By Western blotting with antibodies specifically recognizing phosphorylated forms of both IKKα and IKKβ, the results reveal that AGEs effectively induced phosphorylation of IKKα/β but had no effect on total IKKα (Figure 2d). In the presence of resveratrol, the levels of AGEs-induced phosphorylated forms of IKKα/β decreased (Figure 2e). To determine whether suppression of IKKα/β-IκBα-NF-κB DNA-binding activity could finally lead to decreased transcriptional activation in target genes, luciferase reporter assays were applied. The reporter plasmid pNF-κB-luciferase and the internal control plasmid pTK-Renilla-luciferase were co-transfected into chondrocytes. After transfection, chondrocytes were equally distributed and were stimulated with AGEs in the presence or absence of various doses of resveratrol. Total cell lysates were collected for luciferase activity measurement. We show that consistent with suppression of DNA-binding activity, resveratrol effectively reduced AGEs-induced NF-κB transcriptional activity in a dose-dependent manner (Figure 2f).

In parallel to its suppressive effects on NF-κB activation, we demonstrated that resveratrol inhibited AGEs-induced AP-1 DNA-binding activity (Figure 3a). As a comparison, resveratrol did not affect SP-1 DNA-binding activity. Since AGEs can induce activation of both ERK and JNK but not p38 in porcine chondrocytes [17], we determined whether resveratrol had any effect on these AP-1 upstream kinase activities. The results show that AGEs-induced phosphorylation of ERK was noticeably suppressed by resveratrol (Figure 3b). In addition, by immunoprecipitation kinase assays, we demonstrated that resveratrol could suppress JNK activity in a dose-dependent manner (Figure 3c); however, resveratrol did not affect the total amount of JNK. Instead, treatment with resveratrol effectively reduced AGEs-stimulated transcriptional activity of AP-1 (Figure 3d).

Different from rheumatoid arthritis, in which infiltrating inflammatory cells and synovial pannus are responsible for joint damage, chondrocytes by themselves are primarily the cells causing cartilage damage through generating and secreting cartilage-breakdown enzymes like MMPs. Because MMP-13, regulated by NF-κB and AP-1 signaling pathways [30], is directly responsible for damaging cartilage matrix, we examined the effects of resveratrol on AGEs-induced MMP-13 activity. As shown in Figure 4a, b, AGEs-induced MMP-13 enzyme activities were suppressed by resveratrol. By quantitative real time RT/PCR analysis, we demonstrate that resveratrol significantly decreased the levels of MMP-13 induced by AGEs treatment (Figure 4c). Since cartilage collagen II is preferentially cleaved by MMP-13, we determined whether resveratrol could affect AGEs-mediated destruction of collagen II. As shown in Figure 4d, resveratrol treatment prevented AGEs-mediated decreases of collagen II by Western blotting. Furthermore, as shown in Additional file 1, resveratrol also suppressed the expression of iNOS and COX-2, the production of PGE₂ and NO, the DNA-binding activities of NF-κB and AP-1 as well as the enzyme activity of MMP-13 in chondrocytes stimulated by another preparation of AGEs using glycoaldehyde-modified albumin that contains CML, pentosidine and other AGEs [19, 20]. Altogether, the results indicate that resveratrol was able to provide cartilage protection at least by down-regulating AGEs-induced MMP expression and enzyme activity as well as by preventing AGEs-mediated decrease of collagen II in chondrocytes.

To further address the chondroprotective effect of resveratrol as a reflection of its anti-inflammatory property, we investigated whether resveratrol could regulate AGEs-induced degradation of cartilage matrix. When the cartilage explants were stained, a significant reduction of Safranin O-positive proteoglycan and an increase of cleavage products of aggrecan (NITEGE) became evident in AGEs-treated samples. Such findings were successfully prevented by resveratrol treatment (Figure 5, upper and middle panels). Consistently, we demonstrated that treatment with resveratrol was effective to prevent AGEs-mediated release of proteoglycan into the supernatants of culture of cartilage explants (Figure 5, lower panel).