Abstract
Objectives. Previous work has shown that the CO-releasing molecule CORM-2 protects against cartilage degradation. The aim of this study was to examine whether CORM-2 can control the production of inflammatory mediators in osteoarthritic chondrocytes and determine the mechanisms involved.
Methods. Primary cultures of chondrocytes from OA patients were stimulated with IL-1β. The production of reactive oxygen species, nitrite, PGE2, TNF-α and IL-1 receptor antagonist (IL-1Ra) were measured in the presence or absence of CORM-2. The expression of nitric oxide synthase-2 (NOS-2), cyclo-oxygenase-2 (COX-2) and microsomal PG E synthase-1 (mPGES-1) was followed by western blot and real-time PCR. Activation of nuclear factor-κB (NF-κB) and hypoxia inducible factor-1α (HIF-1α), and phosphorylation of NF-κB inhibitory protein α (IκBα) were determined by ELISA.
Results. CORM-2 decreased the production of oxidative stress, nitrite and PGE2. In addition, CORM-2 inhibited IL-1β-induced TNF-α but enhanced IL-1Ra production. Treatment of chondrocytes with CORM-2 strongly down-regulated NOS-2 and mPGES-1 protein expression, whereas COX-2 was reduced to a lesser extent. These changes were accompanied by a significant decrease in mRNA expression for NOS-2 and mPGES-1. CORM-2 showed a concentration-dependent inhibition of DNA-binding activity for p65 NF-κB and HIF-1α. IκBα phosphorylation was also reduced by CORM-2 treatment.
Conclusions. These data have opened new mechanisms of action for CORM-2, raising the prospect that CO-releasing molecules are an interesting strategy for the development of new treatments in articular conditions.
Introduction
Biomechanical stress likely initiates the production of inflammatory mediators leading to the dysregulation of cartilage homoeostasis in OA [1]. Therefore, increased levels of cytokines such as IL-1β and TNF-α have been found in OA tissues and are believed to play a role in the progression of this condition. These cytokines stimulate the production of reactive oxygen species (ROS), nitric oxide (NO), PGE2 and degradative enzymes [2, 3] but inhibit the synthesis of cartilage matrix proteins [4]. High levels of NO are generated in OA chondrocytes by the inducible enzyme nitric oxide synthase-2 (NOS-2) [5], which leads to interactions with superoxide to form peroxynitrite. This cytotoxic product and a number of reactive species derived from it can oxidize cellular components and deplete endogenous anti-oxidants [6]. In addition, the coordinated induction of cyclo-oxygenase-2 (COX-2) [7] and microsomal PG E synthase-1 (mPGES-1) [8] by pro-inflammatory cytokines results in the enhanced production of PGE2 in OA chondrocytes.
The synergistic effects resulting from the plethora of inflammatory mediators lead to cartilage degradation, although some endogenous molecules can regulate cytokine actions. For instance, IL-1 signalling is modulated by IL-1 receptor antagonist (IL-1Ra), thus protecting against the catabolic effects of IL-1β [9]. IL-1Ra is approved for the treatment of RA and it has shown beneficial effects in experimental OA, as IA injections of the IL-1Ra gene can prevent the progression of articular damage [10].
A major cytokine-induced signalling pathway involves the activation of nuclear factor-κB (NF-κB) which is responsible for the transcription of many genes relevant in joint inflammation and OA pathogenesis [11]. Regulation of NF-κB may prevent the progression of articular lesion. Thus, a number of drugs approved for OA treatment inhibit the activation of this transcription factor by IL-1β in OA chondrocytes [12, 13]. In addition, studies using adenoviral vector-mediated NF-κB p65-specific siRNA administration during experimental OA have corroborated the relevance of this transcription factor [14].
Other strategies for modulating the production of inflammatory mediators have been suggested by recent studies on the role of hypoxia inducible factor-1α (HIF-1α) in inflammatory pathways and OA cartilage metabolism [15]. This transcription factor is believed to be an important regulator of cellular adaptation to catabolic stress and low oxygen tension. In OA chondrocytes, catabolic or oxidative stress and pro-inflammatory cytokines such as IL-1β and TNF-α induce the expression of HIF-1α [16].
CO-releasing molecules (CO-RMs) can reproduce the biological actions of CO derived from haem oxygenase-1 (HO-1) activity [17]. These compounds have shown anti-inflammatory effects in some cell lines [18–20] and can regulate leucocyte–endothelial cell interactions [21] and collagen-induced arthritis [22]. We recently reported the protective effects of HO-1 induction [23] and tricarbonyldichlororuthenium (II) dimer (CORM-2) against OA cartilage degradation. Thus, we noted that CORM-2 significantly down-regulated MMP and aggrecanase expression, and enhanced type II collagen expression and glycosaminoglycan synthesis [24]. The contribution of articular inflammation to the joint degradation associated with OA, prompted us to consider the modification of the inflammatory process, NF-κB and HIF-1α signalling in chondrocytes as possible targets for CO-RMs in OA, which may lead us to understand the possible beneficial effects of this class of drugs in articular diseases. The objective of the present study was to investigate whether CORM-2 acts on important regulators of the inflammatory process in primary human OA chondrocytes.
Materials and methods
Reagents
IL-1β was from Peprotech EC Ltd. (London, UK). Antibodies against NOS-2, COX-2 and mPGES-1 were purchased from Cayman (Ann Arbor, MI, USA). The peroxidase-conjugated IgGs were purchased from Dako (Copenhagen, Denmark). CORM-2, RuCl3 and other reagents were from Sigma Aldrich (St Louis, MO, USA).
Chondrocyte culture
Cartilage specimens were obtained from patients with the diagnosis of advanced OA (20 females, 12 males, aged 76 ± 8 yrs, mean ± s.e.m.) undergoing total knee joint replacement. Diagnosis was based on clinical, laboratory and radiological evaluation. The design of the work was approved by the Institutional Ethical Committee (Comité Etico de Investigación Clínica del Hospital Clinic Universitari de Valencia, Spain). Samples were obtained under patient's consent according to the declaration of Helsinki. Cartilage slices were removed from the femoral condyles and tibial plateaus, and cut into small pieces. Chondrocytes were isolated by sequential enzymatic digestion: 1 h with 0.1 mg/ml hyaluronidase followed by 12 h with 1 mg/ml collagenase (type IA) at 37°C in DMEM/Ham's F-12 containing penicillin (100 U/ml) and streptomycin (100 g/ml) at 37°C in 5% CO2 atmosphere. The digested tissue was filtered through a 70 m nylon mesh, washed and centrifuged. Cell viability was >95% according to the Trypan blue exclusion test. The isolated chondrocytes were seeded at 2.5 × 105 cells/well in 6-well plates. Cells were cultured in DMEM/Ham's F-12 supplemented with 10% human serum, penicillin (100 U/ml) and streptomycin (100 g/ml) in a humidified 5% CO2 incubator at 37°C. Chondrocytes in primary culture were allowed to grow until nearly confluence and then incubated with CORM-2 at different concentrations or vehicle for 1 h before stimulation with IL-1β (100 U/ml) for different times. CORM-2 was dissolved in dimethyl sulphoxide and then diluted in culture medium (0.1% v/v). Control cells were treated with the same vehicle. Possible cytotoxicity of treatments was assessed by the mitochondrial-dependent reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) to formazan. After appropriate stimulation, cells were incubated with MTT (200 μg/ml) for 2 h. The medium was then removed and the cells solubilized in dimethyl sulphoxide (100 μl) to quantitate formazan at 550 nm.
Oxidative stress assay
Oxidative stress was assessed by laser scanning cytometry (LSC) analysis. Formation of intracellular ROS was detected using dihydrorhodamine 123 (DHR), which is oxidized to fluorescent rhodamine (excitation at 485 nm and emission at 534 nm). Chondrocytes (75 000/well) were spread in 8-well Lab-tek chambers (Nalge Nunc International, Naperville, IL, USA) with DMEM/Ham's F-12, penicillin (100 U/ml), streptomycin (100 g/ml) and 10% human serum in a humidified 5% CO2 incubator at 37°C. Confluent cells were incubated with DMEM without phenol red and DHR (5 μM) for 15 min at 37°C. After washing, fresh medium was added and cells were incubated with CORM-2 for 30 min and then with IL-1β (100 U/ml) in the presence or absence of CORM-2 for other 30 min. Cultures were washed twice with fresh medium and analysed by LSC.
Determination of nitrite and PGE2
Chondrocytes in primary culture were stimulated with IL-1β (100 U/ml) or IL-1β + CORM-2 for 24 h. Supernatants were harvested, centrifuged and frozen at −80°C until analysis. Nitrite was determined fluorometrically in microtitre plates using a standard curve of sodium nitrite [25]. PGE2 was measured by RIA [26].
Western blot analysis
After 24 h stimulation with IL-1β (100 U/ml) or IL-1β + CORM-2 (100 μM), chondrocytes in primary culture were lysed in 100 μl of buffer (1% Triton X-100, 1% deoxycholic acid, 20 mM NaCl and 25 mM Tris, pH 7.4) and centrifuged at 4°C for 10 min at 10 000g. Proteins (25 μg) in supernatants of cell lysates were separated by 12.5% SDS–PAGE and transferred onto polyvinylidene difluoride membranes (GE Healthcare Life Sciences, Barcelona, Spain). Membranes were blocked with 3% BSA and incubated with specific antibodies for 2 h at room temperature. Finally, membranes were incubated with peroxidase-conjugated goat anti-rabbit IgG and the immunoreactive bands were visualized by enhanced chemiluminescence (GE Healthcare Life Sciences) using the AutoChemi image analyser (UVP Inc., Upland, CA, USA).
Real-time PCR
Chondrocytes in primary culture were stimulated with IL-1β (100 U/ml) or IL-1β + CORM-2 (100 μM) for 12 h. Total RNA was extracted using the TRIzol reagent (Life Technologies Inc., Barcelona, Spain) according to the manufacturer's instructions. Reverse transcription was accomplished on 1 μg of total RNA using random primers (TaqMan reverse transcription reagents, Applied Biosystems Spain, Madrid). PCR reactions were performed using SYBR Green PCR Master Mix (Bio-Rad Laboratories, Madrid, Spain). PCR assays were performed in duplicate on an iCycler Real-Time PCR Detection System (Bio-Rad Laboratories) running the cycling conditions: 95°C for 10 min, 40 cycles of 95°C for 15 s and 60°C for 1 min. Reaction specificity was determined by melt curve analysis, which was performed by heating the plate from 55°C to 95°C and measuring SYBR Green I dissociation from the amplicons. Cycle threshold (CT) values for each gene were corrected using the mean CT value for β-actin. Relative gene expression was calculated using the ΔCT method and expressed as fold change (2−ΔΔCT) relative to the expression values in non-stimulated cells.
ELISA
Chondrocytes in primary culture were stimulated with IL-1β (100 U/ml) or IL-1β + CORM-2 for 24 h. Supernatants were harvested, centrifuged and frozen at −80°C until analysis. TNF-α and IL-1Ra were measured by ELISA kits from eBioscience (San Diego, CA, USA) and Biosource (Invitrogen, Barcelona, Spain), respectively, with sensitivity of 4 pg/ml. NF-κB and HIF-1α DNA binding was quantitated by ELISA in nuclear extracts from chondrocytes stimulated with IL-1β (100 U/ml) or IL-1β + CORM-2 for 1 h, using the Trans AM™ kit for p65 or HIF-1 (Active Motif Europe, Rixensart, Belgium), respectively, following the manufacturer's recommendations. NF-κB inhibitory protein α (IκBα) phosphorylation was measured by the CASE™ cellular activation of signalling ELISA kit (SuperArray, Frederick, MD, USA).
Data analysis
Results are presented as mean ± s.e.m. Statistical analyses were performed using one-way analysis of variance followed by Dunnett's t-test for multiple comparisons and unpaired Student's t-test for dual comparisons.
Results
CORM-2 inhibits the production of ROS, NO and PGE2
Upon activation by 100 U/ml IL-1β for 24 h, chondrocytes produced high levels of ROS, NO and PGE2 (Fig. 1). Figure 1A shows that CORM-2 significantly decreased IL-1β-stimulated ROS production, whereas the negative control RuCl3 was ineffective. Basal levels of ROS were obtained in the presence of the highest concentration of CORM-2 (150 μM). Up-regulation of NOS-2 resulted in the production of NO, measured as nitrite levels in the medium. As shown in Fig. 1B, CORM-2 reduced nitrite levels. In addition, chondrocytes stimulated with IL-1β released significantly more PGE2 into the medium when compared with control. The addition of CORM-2 decreased the production of this eicosanoid in a concentration-dependent manner and, at the highest concentration (150 μM) CORM-2 blocked PGE2 release induced by IL-1β to basal levels (Fig. 1C). These effects of CORM-2 were not the result of cytotoxicity as cells treated with this compound exhibited percentages of viability of 95.8 ± 0.9, 93.1 ± 2.2 and 92.1 ± 2.4 at 50, 100 and 150 μM, respectively, after 24 h incubation with OA chondrocytes in the presence of IL-1β.
Inhibition of IL-1β-induced oxidative stress (A), NO (B) and PGE2 (C) production by CORM-2. OA chondrocytes were incubated with IL-1β (100 U/ml) in the absence or presence of CORM-2 at various concentrations (50, 100 and 150 μM) or RuCl3 (150 μM) for 24 h. Oxidative stress was measured by LSC using dihydrorhodamine 123 (A), NO was measured as nitrite by a fluorometric procedure (B) and PGE2 was measured by RIA (C). Values were normalized with respect to the protein content. Data are mean ± s.e.m. of independent cultures with cells from seven different donors. *P < 0.05, **P < 0.01 with respect to IL-1β; ##P < 0.01 with respect to non-stimulated cells.
CORM-2 regulates NOS-2, COX-2 and mPGES-1 expression
Chondrocytes stimulated with IL-1β for 24 h demonstrated significant elevations in NOS-2, COX-2 and mPGES-1 protein expression (Fig. 2A). CORM-2 treatment strongly down-regulated NOS-2 and mPGES-1 protein expression, whereas COX-2 was reduced to a lesser extent. The inhibition of NO production by CORM-2 could be the result of diminished NOS-2 protein expression, whereas the inhibition of PGE2 levels could be dependent on the reduction of COX-2 and mPGES-1 protein expression by CORM-2. In chondrocytes stimulated with IL-1β for 12 h, we measured the relative gene expression of NOS-2, COX-2 and mPGES-1 by real-time PCR. Figure 2B shows that CORM-2 treatment repressed NOS-2 and mPGES-1 transcripts increased by IL-1β, while the modulation of COX-2 mRNA expression was not significant.
Effect of CORM-2 on protein (A) and mRNA (B) expression of NOS-2, COX-2 and mPGES-1. OA chondrocytes were stimulated with IL-1β (100 U/ml) for 24 h (A) or 12 h (B) in the presence or absence of CORM-2 (100 μM). The immunoblot is representative of three independent experiments. The mRNA expression was determined by real-time PCR. Data are expressed as mean ± s.e.m. of independent cultures with cells from three different donors. *P < 0.05, with respect to IL-1β.
Effect of CORM-2 on the production of TNF-α and IL-1Ra
Chondrocytes stimulated with IL-1β generated increased levels of TNF-α and IL-1Ra (Fig. 3). We found that CORM-2 concentration-dependently inhibited IL-1β-induced TNF-α production (Fig. 3A). As far as the IL-1Ra production by chondrocytes was concerned, although CORM-2 alone had no significant effect, this compound stimulated IL-1Ra production in the presence of IL-1β (Fig. 3B). Our results show a concentration-dependent effect of CORM-2.
Effect of CORM-2 on TNF-α (A) and IL-1Ra (B) production in human OA chondrocytes. OA chondrocytes were stimulated with IL-1β (100 U/ml) for 24 h in the presence or absence of CORM-2 (50, 100 and 150 μM) or RuCl3 (150 μM). TNF-α and IL-1Ra were measured by ELISA in supernatants. Data are expressed as mean ± s.e.m. of independent cultures with cells from ten different donors. *P < 0.05, **P < 0.01, with respect to IL-1β; ##P < 0.01 with respect to non-stimulated cells.
CORM-2 reduces IL-1β-dependent NF-κB activation
IL-1β-mediated induction of inflammatory mediators is associated with NF-κB activation [11]. Stimulation of primary OA chondrocytes with IL-1β resulted in an increased binding of NF-κB to its consensus sequence (Fig. 4A). Consistent with the reduction in inflammatory mediators, CORM-2 showed a concentration-dependent inhibition of p65-binding activity.
Effect of CORM-2 on NF-κB activation in OA chondrocytes. Cells in primary culture were stimulated with IL-1β (100 U/ml) for 1 h in the presence or absence of CORM-2. NF-κB binding to consensus oligonucleotide (A) was measured in nuclear extracts by ELISA. Results were normalized with respect to protein content. The rate phosphorylated IκBα/total IκBα protein (B) was measured in cellular extracts by ELISA after 30 min of cell stimulation with IL-1β. Data are expressed as mean ± s.e.m. of independent cultures with cells from four different donors. *P < 0.05, **P < 0.01 with respect to IL-1β; ##P < 0.01 with respect to non-stimulated cells.
The process of NF-κB activation depends on the phosphorylation of its inhibitory protein, IκBα which plays a key role in the maintenance of cytoplasmic localization of the inactive complexes to achieve efficient NF-κB activation by extracellular signals, and regulates nuclear translocation [27]. We investigated whether the inhibition of NF-κB–DNA binding was due to the inhibition of IκBα phosphorylation. As shown in Fig. 4B, CORM-2 reduced the ratio of phosphorylated IκBα to total IκBα protein.
CORM-2 reduces IL-1β-dependent HIF-1α activation
The transcription factor HIF-1α has been shown to support the expression of mPGES-1 in OA chondrocytes [15]. To address this potential mechanism, we analysed the effect of CORM-2 on the activation of this transcription factor. Figure 5 shows that IL-1β strongly stimulated the binding of HIF-1α to its consensus sequence, whereas in chondrocytes treated with CORM-2, a concentration-dependent inhibition was observed.
Effect of CORM-2 on HIF-1α activation in OA chondrocytes. Cells in primary culture were stimulated with IL-1β (100 U/ml) for 1 h in the presence or absence of CORM-2. HIF-1α binding to consensus oligonucleotide was measured in nuclear extracts by ELISA. Results were normalized with respect to protein content. Data are expressed as mean ± s.e.m. of independent cultures with cells from four different donors. **P < 0.01 with respect to IL-1β; ##P < 0.01 with respect to non-stimulated cells.
Discussion
This study demonstrates for the first time that a CO-RM can control in primary human OA chondrocytes the production of inflammatory mediators relevant in the progression of OA. Previous studies have demonstrated an overproduction of NO and ROS in inflammatory and degenerative joint diseases [28]. Large amounts of NO are associated with matrix degradation, which can be dependent on the suppression of glycosaminoglycan and collagen synthesis, expression of MMPs and activation of pro-enzymes [29]. ROS may also mediate the induction of degradative enzymes [30]. We have shown that CORM-2 reduces the production of oxidative stress and NO in OA chondrocytes stimulated with IL-1β, which may play a role in the protective effects of CORM-2 against cartilage degradation [24]. Our data indicate that CORM-2 controls NO production through the inhibition of NOS-2 mRNA and protein expression, which is in line with our results in the cell line caco-2 [20].
It is becoming clear that NO can produce profound effects on cytokine actions. For example, NO can reduce IL-1Ra production leading to an enhancement of cartilage matrix degradation by IL-1 [31]. We have shown that CORM-2 is able to enhance the production of IL-1Ra in primary OA chondrocytes stimulated with IL-1β. Although the levels produced are low, it is likely that in vivo they may block small amounts of IL-1 that synergize with other cytokines to induce relevant catabolic effects [32]. One of these cytokines is TNF-α, which induces the production of inflammatory mediators and catabolic enzymes in OA tissues [10]. Conversely, blockade of TNF-α inhibits extracellular matrix cleavage and increases glycosaminoglycan content in OA cartilage [33]. Therefore, the reduction of TNF-α production by CORM-2 in primary OA chondrocytes could contribute to the anti-inflammatory effects of this agent and it is of interest in the context of OA, in view of the catabolic effects of this cytokine.
On the other hand, NO may cooperate with ROS to induce chondrocyte apoptosis [34]. Our results suggest that inhibition of ROS generation by CORM-2 may contribute to the anti-apoptotic effects of this agent in primary OA chondrocytes [35]. There is compelling evidence of regulatory effects of CO on redox signalling. Exogenous CO from CO-RMs or HO-1 stimulation has been reported to inhibit NADPH oxidase and cytochrome oxidase in some cell types [36]. Further studies would be necessary to determine whether the inhibition of ROS production by CORM-2 is the result of the regulation of these enzymes in primary OA chondrocytes.
Elevated levels of PGE2 may play a role in cartilage degradation by promoting the production of MMPs [37]. In contrast, low concentrations of PGE2 may down-regulate collagenases, pro-inflammatory cytokines and collagen 2A1 cleavage [38]. This eicosanoid may also sensitize OA chondrocytes to the cell death induced by NO [39]. In addition to COX-2, mPGES-1 is significantly increased in OA cartilage [8] and chondrocytes [15]. This enzyme has been proposed as a novel target for OA [40], although mPGES-1 deficiency does not affect OA induced by joint instability [41]. Our studies have shown that CORM-2 is able to control the enhanced synthesis of PGE2 in OA chondrocytes treated with IL-1β. We have found that CORM-2 inhibits the expression of mPGES-1 protein and to a lesser extent of COX-2, suggesting that the inhibitory effects of CORM-2 on PGE2 production stimulated by IL-1β can be mainly dependent on the reduction in mPGES-1 gene expression. These results show, for the first time, that mPGES-1 can be a target for CORMs.
NF-κB is a key intracellular signal pathway for inflammatory and catabolic gene expression in articular chondrocytes [42, 43]. Therefore, the inhibition of NF-κB activation by CORM-2 may provide a basis for the observed reductions in NOS-2 expression and NO production. In addition, our results may explain in part the mechanism responsible for the inhibition of expression of key degradative enzymes such as MMP-1, MMP-3, MMP-10, MMP-13 and aggrecanase by CORM-2 [24]. Although NF-κB binding sites are involved in COX-2 activation by IL-1 in chondrocytes, other transcription factors can also play a role [7], which would explain the weak reduction in COX-2 expression achieved by CORM-2.
The phosphorylation of IκBα and subsequent proteosomal degradation are critical steps in NF-κB activation [44]. The degradation of IκB allows phosphorylation of NF-κB and its translocation to the nucleus [45]. Therefore, the inhibition of IκBα phosphorylation by CORM-2 would result in the prevention of NF-κB nuclear translocation and DNA binding. A number of enzymes are involved in IκBα phosphorylation. For instance, recent studies indicate that protein kinase Cζ mediates the activation of IκB kinases after IL-1β stimulation of human chondrocytes [43]. Further work would be necessary to identify the kinases relevant for IκBα phosphorylation that may be affected by CORM-2 treatment.
The mechanisms regulating the transcription of mPGES-1 in human OA chondrocytes remain to be determined. It is known that peroxisome proliferator-activated receptor-γ activation exerts inhibitory effects [8] and recent studies suggest the contribution of HIF-1α to the regulation of mPGES-1 and to a lower extent of COX-2 expression in primary chondrocytes [15]. Our data suggest that inhibition of NF-κB and HIF-1α activation may play a role in the inhibitory effects of CORM-2 on mPGES-1 expression. In addition, extracellular signal-regulated kinase 1/2 (ERK1/2) can participate in the induction of mPGES-1 in human chondrocytes [40]. Therefore, the reduction in ERK1/2 phosphorylation by CORM-2 [24] may be an additional mechanism leading to the inhibition of inflammatory gene expression.
Our findings indicate that CORM-2 reduces oxidative stress and the production of NO, TNF-α and PGE2 but enhances IL-1Ra in primary OA chondrocytes. Some of these effects may be dependent on the regulation of NF-κB and HIF-1α. Results from the present study establish the presence of anti-inflammatory mechanisms in the beneficial effects elicited by CORM-2 in OA chondrocytes and cartilage [24], raising the prospect that CO-RMs may be developed for the treatment of articular conditions.
Acknowledgements
Funding: This work was supported by grant SAF2007-61769 (Ministerio de Educación y Ciencia-FEDER). J.M. thanks Spanish Ministerio de Educación y Ciencia for a fellowship.
Disclosure statement: The authors have declared no conflicts of interest.
Comments