Introduction
Osteoarthritis (OA) is a degenerative joint disease with abnormal alterations in the structure, composition and function of articular tissues. Morphological changes in OA include cartilage destruction, osteophyte formation and synovial inflammation [
1]. It is believed that overproduced proinflammatory cytokines, such as IL-1β and TNFα, are involved in the pathogenesis of OA by metalloproteinase (MMP) upregulation and collagen downregulation. Moreover, IL-1β and TNFα stimulate the production of prostaglandins, such as prostaglandin E
2 (PGE
2), whose roles in the inflammatory process and proteoglycan degradation in human OA cartilage have been reported [
2].
PGE
2, a major prostaglandin produced via arachidonic acid (AA) metabolism, is involved in many physiological events, such as cell growth, immune regulation, inflammation and arthritis [
3]. Cyclooxygenase (COX) and prostaglandin E
2 synthase (PGES) are key enzymes for PGE
2 biosynthesis under inflammatory conditions. COX-1 and COX-2 convert AA into prostaglandin H
2, which is subsequently converted to PGE
2 by PGES. Among all forms of PGES, microsomal PGES-1 (mPGES-1) has been studied the most because of its inducible characteristic and collaboration with inducible COX-2, leading to PGE
2 production. It has been shown that mPGES-1 expression is upregulated in animal models of rheumatoid arthritis and in patients suffering from OA [
4,
5] and is downregulated by anti-inflammatory drugs [
6,
7]. mPGES-1 expression is regulated by several transcription factors, such as early growth response-1 (Egr-1) and NF-κB, in different cell types [
8,
9].
Leukotrienes, other end-products of AA metabolism, are potent mediators of inflammation that increase the activation, migration and adhesion of immune cells [
10]. In particular, leukotriene B
4 (LTB
4) promotes the production and release of proinflammatory cytokines from synovial membranes, drawing more attention to its role in OA. Clinical investigations have revealed that long-term COX-2 inhibition causes a switch to the 5-lipoxygenase (5-LOX) pathway, leading to LTB
4 production [
11]. It is now clear that prostaglandins and leukotrienes have additional outcomes; blocking the production of these mediators might therefore have synergistic effects and achieve optimal anti-inflammatory activity [
3]. Functional 5-LOX requires binding to 5-lipoxygenase-activating protein (FLAP), which helps 5-LOX binding to AA at the nuclear membrane and enhances the efficiency of leukotriene synthesis [
12]. 5-LOX gene expression at the transcriptional level is regulated by several transcription factors, such as smad3/4 and vitamin D receptors [
13,
14].
Reactive oxygen species produced inside the joints also are associated with the pathogenesis of OA [
15]. They induce lipid peroxidation of membrane polyunsaturated fatty acids, eliciting the production of aldehydes from AA [
16,
17]. 4-Hydroxynonenal (HNE) has been found to be the predominantly produced and most reactive aldehyde in OA articular tissues [
18]. We recently reported, for the first time, that free HNE induces cartilage degradation in isolated OA chondrocytes and alters the cellular phenotype of OA osteoblasts. These responses are mediated by the modulation of a panoply of signaling pathways, including mitogen-activated protein kinases and NF-κB [
18‐
20]. By binding to proteins, HNE activates MMP-13 and increases the susceptibility of type II collagen to proteolytic cleavage by MMP-13 [
18]. In another study, we demonstrated that HNE contributes to inflammatory responses in OA chondrocytes through the transcriptional upregulation of COX-2 via the p38 mitogen-activated protein kinase signaling pathway. We observed that HNE-induced COX-2 declines rapidly after 8 hours of incubation. HNE most probably represents one of the main lipid peroxidation products that can modulate physiological as well as pathological processes, as depicted beautifully in a recent, dedicated review [
21].
The objectives of the present project were to investigate whether COX-2 and mPGES-1 downregulation, attributed to HNE depletion, is responsible for the switch from COX-2 and mPGES-1 to 5-LOX and FLAP, and to elucidate the molecular mechanisms underlying their expression in HNE-treated human OA chondrocytes.
Materials and methods
Tissue samples
Postsurgery, discarded human OA articular cartilage was obtained from OA patients (mean ± standard deviation age, 67 ± 9 years) who underwent total knee arthroplasty. Informed consent had been obtained from patients with OA for the use of their tissues for research purposes. All patients were evaluated by rheumatologists who followed American College of Rheumatology criteria [
22]. The Clinical Research Ethics Committee of the Hôpital du Sacré-Cœur de Montréal, including clinicians, researchers and jurists, approved the study protocol and the use of human articular tissues.
Chondrocyte culture
OA cartilage (femoral condyles and tibial plateaus) was obtained under aseptic conditions and carefully dissected from the underlying bone in each specimen [
23]. OA chondrocytes were extracted by sequential enzymatic digestion with 1 mg/ml pronase (Sigma, Oakville, ON, Canada) for 1 hour at 37°C, and then with 2 mg/ml type IV collagenase (Sigma) for 6 hours in DMEM (Invitrogen, Burlington, ON, Canada) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen), 100 units/ml penicillin and 100 μg/ml streptomycin (Invitrogen). The cells were seeded at high density in culture flasks at 37°C in a humidified atmosphere of 5% CO
2/95% air until they were confluent and ready for the experiments. First-passage cells were employed to ensure their phenotype and were seeded at 10
5 cells/cm
2 in culture tissue plates. The DMEM containing 2% fetal bovine serum and antibiotics was replaced 24 hours before the experiments were performed in this medium with the factors under study for different incubation time periods.
For COX-2 and mPGES-1 studies (up to 24 hours), chondrocytes were treated with a single addition of 10 μM HNE or with repeated treatments by adding 10 μM HNE to the cultures at 0, 2, 4, 6, 8, 10, 12 and 14 hours. For 5-LOX and FLAP studies (up to 72 hours), cells were treated with a single addition of 10 μM HNE in the presence or absence of 50 μM naproxen or 100 μg/ml anti-transforming growth factor-beta 1 (TGFβ1).
Protein detection by western blotting
Twenty micrograms of total proteins from chondrocyte lysates treated with HNE under the indicated conditions were loaded for discontinuous 4 to 12% SDS-PAGE. Protein transfer, immunodetection and semiquantitative measurements were performed as described previously [
18]. The primary antibodies were rabbit anti-COX-2 (Cayman Chemical, Hornby, ON, Canada), anti-mPGES-1 (Cayman Chemical), anti-β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and anti-Egr-1 (Santa Cruz Biotechnology). After serial washes, primary antibodies were detected by goat anti-rabbit IgG conjugated to horseradish peroxidase (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA). Immunoreactive proteins were detected with SuperSignal blotting substrate (Pierce Biotechnology, Inc., Rockford, IL, USA) and were exposed to clear-blue X-ray film (Pierce).
RNA extraction and RT-PCR
Total RNA was isolated with TRIzol reagent according to the manufacturer's instructions. RNA was quantitated with the RiboGreen RNA quantitation kit (Molecular Probes, Eugene, OR, USA), dissolved in diethylpyrocarbonate-treated H2O, and stored at -80°C until use. One microgram of total RNA was reverse-transcribed with Moloney murine leukemia virus reverse transcriptase (Fermentas, Burlington, ON, Canada), as detailed in the manufacturer's guidelines. One-fiftieth of the reverse transcriptase reaction product was analyzed by traditional PCR or by real-time quantitative PCR. The following sense and antisense specific primers (Bio-Corp, Inc., Montreal, QC, Canada), were tested: human mPGES-1, 5'-GAA GAA GGC CTT TGC CAA C-3' (sense) and 5'-GGA AGA CCA GGA AGT GCA TC-3' (antisense); human 5-LOX, 5'-CTG TTC CTG GGC ATG TAC CC-3' (sense) and 5'-GAC ATC TAT CAG TGG TCG TG-3' (antisense); human FLAP, 5'-AAT GGG AGG AGC TTC CAG AG-3' (sense) and 5'-ACC AAC CCC ATA TTC AGC AG-3' (antisense); and human GAPDH, 5'-CAG AAC ATC ATC CCT GCC TCT-3' (sense) and 5'-GCT TGA CAA AGT GGT CGT TGA G-3' (antisense).
Quantitative PCR analysis was performed in a total volume of 50 μl containing template DNA, 200 nM sense and antisense primers, 25 μl SYBR Green Master Mix (Qiagen, Mississauga, ON, Canada), and 0.5 units uracil-N-glycosylase (Epicentre Technologies, Madison, WI, USA). After incubation at 50°C for 2 minutes (uracil-N-glycosylase reaction) and at 95°C for 10 minutes (uracil-N-glycosylase inactivation and activation of AmpliTaq Gold enzyme), the mixtures were subjected to 40 amplification cycles (15 seconds at 95°C for denaturation and 1 minute for annealing and extension at 60°C). Incorporation of SYBR Green dye into the PCR products was monitored in real time with a Mx3000 real-time PCR system (Stratagene, La Jolla, CA, USA), to determine the threshold cycle (Ct) at which exponential amplification of PCR products begins. After PCR, dissociation curves were generated with one peak, indicating amplification specificity. A Ct value was obtained from each amplification curve with the software provided by the manufacturer (Stratagene).
Relative mRNA expression in chondrocytes was quantified according to the ΔΔCt method, as detailed in the manufacturer's guidelines (Stratagene). A ΔCt value was first calculated by subtracting the Ct value for the housekeeping gene GAPDH from the Ct value for each sample. A ΔΔCt value was then calculated by subtracting the ΔCt value for the controls (unstimulated cells) from the ΔCt value for each treatment. Fold changes compared with the controls were then determined by 2-ΔΔCt. Each PCR generated only the expected specific amplicon, as shown by melting temperature profiles of the final product and gel electrophoresis of the test PCRs. Each PCR was performed in triplicate on two separate occasions for each independent experiment.
Prostaglandin E2, leukotriene B4 and TGF-β1 enzyme immunoassay
After incubation, the culture medium from cultured OA chondrocytes was collected and the PGE2, LTB4 and TGF-β1 levels were measured with specific commercial kits from Cayman Chemical and R&D Systems (Minneapolis, MN, USA), according to the manufacturer's instructions. Detection sensitivity was 9, 13.7 and 4.6 pg/ml, respectively. All assays were performed in duplicate.
Plasmids and transient transfection
The human mPGES-1 promoter construct (-538/-28) was a gift from Dr Terry J. Smith (University of California, Los Angeles, CA, USA) [
24]. The pEgr-1Mutx3-TK-Luc reporter construct was generously provided by Dr Yuqing E. Chen (Morehouse School of Medicine, Atlanta, GA, USA) [
25]. The 5-LOX promoter construct was kindly donated by Dr Dieter Steinhilber (University of Frankfurt, Frankfurt, Germany).
Subconfluent human OA chondrocytes were transiently transfected in 12-well cluster plates with lipofectamine 2000™ reagents (Invitrogen Life Technology, Inc.), according to the manufacturer's protocol. Briefly, transfections were conducted for 6 hours with DNA lipofectamine complexes containing 10 μl lipofectamine reagent, 2 μg DNA plasmid and 0.5 μg pCMV-β-galactosidase (as a transfection efficiency control). In cotransfection experiments the amount of transfected DNA was kept constant using a corresponding empty vector. Using the method of Howcroft and colleagues [
26], the transfection efficiency was measured and the average rate of transfected cells was 48%. After washing, the cells were incubated for different incubation periods in the presence or absence of 10 μM HNE (single addition) in fresh DMEM culture medium containing 2% fetal bovine serum in a humidified atmosphere of 95% air/5% CO
2 at 37°C. Luciferase activity was determined in cell lysate with commercially available kits (Luciferase assay system; Promega Corporation, Madison, WI, USA). The data were normalized to the β-galactosidase level, which was quantified in cell lysate by a specific ELISA (Roche Applied Science, Laval, QC, Canada).
Statistical analysis
All quantitative results were calculated as the mean ± standard error of the mean. The data were assessed by Student's unpaired t test. P < 0.05 was considered statistically significant.
Discussion
HNE, a product of lipid peroxidation, has been identified as a modulator of signal transduction, gene transcription and protein modification [
21]. Our previous study demonstrated that the level of HNE is significantly higher in synovial fluids and in chondrocytes of OA patients compared with normal subjects. HNE induces cartilage degradation by transcriptional and post-transcriptional changes of type II collagen and MMP-13 in OA [
18]. Recently, HNE was found to be an inflammatory factor that stimulates the production of proinflammatory mediators, including PGE
2, but its role in the pathology of joint and inflammatory diseases remains unclear. In the present study, we investigated the effects of HNE on COX-2/mPGES-1 and 5-LOX pathways regulation in human OA chondrocytes. COX-2/mPGES-1 and 5-LOX are primary proteins in the synthesis of PGE
2 and LTB
4, respectively.
In the present study, we observed that single treatment of OA chondrocytes with 10 μM HNE induced PGE
2 release as well as COX-2 and mPGES-1 expression, which plateaued at 8 hours of incubation before declining at 16 and 24 hours, respectively. These data are in concordance with our previously reported COX-2 findings [
20] and are supported by other investigations indicating that HNE is a potent inducer of COX-2 and PGE
2 release in various cell lines, such as rat liver epithelial RL34 cells and the RAW264.7 macrophage cell line [
29‐
31]. Interestingly, when HNE was maintained for 14 hours in cultured chondrocytes, through repeated treatments at intervals of 2 hours, the aldehyde induced both COX-2 and mPGES-1 expression up to 24 hours and consequently abolished their decrease. It has been shown that the maintaining of HNE levels at 1 μM through repetitive treatments regulated cell proliferation and differentiation [
32]. Our observations support the view that decreased COX-2 and mPGES-1 expression is attributed to HNE depletion via its metabolism in chondrocytes. Mammalian cells have developed multiple enzymatic pathways for HNE detoxification. The best characterized of these enzymes include glutathione S-transferases, aldehyde dehydrogenase, and alcohol dehydrogenase [
33]. The half-life of HNE in cells was estimated to be ~1 hour [
34]. On the contrary, we do not exclude the possibility that the increase of COX-2 and mPGES-1 expression up to 24 hours may coincide with the inflammatory resolution. As demonstrated by Gilroy and colleagues [
35], COX-2 plays a role as a proinflammatory factor in the early phase of inflammation. During the late phase, however, this enzyme regulates resolution of acute inflammation by generating an alternate set of prostaglandins, such as the cyclopentenone family.
Since mPGES-1 regulation by HNE has not yet been reported, we extended our investigation to determine whether changes in mRNA levels may be ascribed to alterations in promoter activity by transiently transfecting chondrocytes with human mPGES-1 promoter-luciferase reporter genes. The treatment of transfected cells with single addition of 10 μM HNE to cultures led to a time-dependent increment of mPGES-1 promoter activity. These data are consistent with the regulation of mPGES-1 expression by HNE at the transcription level. Emerging evidence has disclosed that the transcriptional induction of mPGES-1 is primarily controlled by Egr-1 through two Egr-1 binding motifs identified in the proximal promoter region of mPGES-1 [
36,
37]. Numerous cytokines and growth factors are known to upregulate mPGES-1 production via Egr-1 activation [
38]. We hypothesized that the induction of Egr-1 activity by HNE could be the mechanism by which HNE exerts its stimulatory effect on mPGES-1 transcription. In transfected cells, HNE evoked the activation of a synthetic luciferase reporter construct containing three tandem repeats of Egr-1 motif. Our western blot data demonstrated that 10 μM HNE heightened Egr-1 protein expression in a time-dependent manner. Collectively, these data suggest the involvement of this transcription factor in HNE-induced transcriptional activity of the mPGES-1 promoter in OA chondrocytes. Further experiments are needed, however, to confirm the direct implication of this transcription factor in HNE-induced mPGES-1.
Thereafter, we explored 5-LOX and FLAP regulation in human OA chondrocytes by HNE, examined the relative involvement of these two proteins in LTB
4 production, and studied the factors that might be responsible for enhanced LTB
4 production in these cells incubated for long periods. Our data indicated that 5-LOX and FLAP upregulation by HNE occurred, at least in part, at the transcription level, as determined by real-time quantitative RT-PCR and transient transfection experiments. HNE-induced 5-LOX and FLAP gene activation is differentially time dependent. FLAP expression is activated earlier than 5-LOX, which occurs only after 48 hours of HNE stimulation. These results are consistent with a previous report of Martel-Pelletier and colleagues showing that 5-LOX expression is activated after that of FLAP, leading to the late increase of LTB
4 production [
28]. The FLAP mRNA level was significantly enhanced after a short period (20 hours) of treatment with TGF-β1 or 1,25-dihydroxyvitamin D
3 alone or combined, whereas the 5-LOX mRNA level rose only after 72 hours [
28]. These authors postulated that the reason for the late increment of 5-LOX mRNA may be that TGF-β1 induces only 5-LOX mRNA accumulation and not true upregulation of gene expression
per se. Heightened LTB
4 production upon HNE stimulation occurred 48 hours after the beginning of stimulation, coinciding with the time at which the 5-LOX mRNA level was elevated, and FLAP gene expression remained high. These data confirm that LTB
4 production is dependent on the regulation of both 5-LOX and FLAP. Other studies, however, have demonstrated that the increase in LTB
4 production was mainly related to upregulation of the FLAP gene [
39] or to a combination of an increment in the activity and/or expression of 5-LOX [
40,
41].
Several investigations have shown that 5-LOX and FLAP expression is regulated by PGE
2 [
28,
42]. Our results put this observation in evidence. PGE
2 inhibition regulates 5-LOX and FLAP expression differentially. Using the nonspecific inhibitor of COX, naproxen, we abolished HNE-induced FLAP expression in OA chondrocytes after short or long periods of incubation. In contrast, naproxen induced 5-LOX expression after short or long stimulation time periods. The combination of naproxen with HNE, however, had no additive effect on 5-LOX mRNA expression. Significant advances have been made in understanding the role of PGE
2 in the metabolism of articular tissues. PGE
2 has been found not only to be involved in inflammatory responses but may also regulate the effects of other inflammatory mediators. More pharmaceutical companies therefore focus on mPGES-1 to treat arthritis because of its inducible form and functional linkage with COX-2 [
38].
Current therapies to treat OA are limited to reducing the pain in affected joints by with either nonselective nonsteroidal anti-inflammatory drugs or selective COX-2 agents. The production of both PGE
2 and LTB
4 requires AA; therefore, one of the pathways is used favorably, depending on the condition. Long-term COX inhibition causes the switch from COX to the 5-LOX pathway. This is observed in OA patients treated with COX-2 inhibitors [
43]. In our study, we also demonstrated that lower PGE
2 production favors the 5-LOX pathway and LTB
4 synthesis in HNE-treated OA chondrocytes. Leukotrienes have greater potential in the inflammatory process, however, and may be more harmful than PGE
2 in various tissues due to their chemotactic properties [
44]. For this reason, drugs that can inhibit both COX and lipoxygenase have attracted pharmaceutical attention in the treatment of inflammatory diseases. Tepoxalin and licofelone (ML3000) have been identified to exert this function [
45,
46]. Today, licofelone is in phase III clinical development. Licofelone not only lowers PGE
2 and LTB
4 production, but can also modify abnormal bone remodeling in OA; thus, it was thought to be useful in treating OA [
46].
In addition to the involvement of PGE
2 in HNE-induced 5-LOX and FLAP, we found data supporting the role of TGF-β1 in this process. Our observations showed that HNE evoked TGF-β1 synthesis and that TGF-β1 neutralization by anti-TGF-β1 antibody attenuated HNE-induced 5-LOX and FLAP after 72 hours of incubation. These results suggest that 5-LOX and FLAP regulation by HNE is partially mediated via TGF-β1 production. Leonarduzzi and colleagues were the first to demonstrate that HNE stimulated TGF-β1 production [
47]. Furthermore, a number of reports support the role of TGF-β1 in 5-LOX and FLAP expression through smad3/4 activation [
13,
28,
40].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
S-HC performed the experimental study, contributed to preparation of the manuscript and undertook the statistical analysis. HF evaluated and interpreted the data and assisted with preparation of the manuscript. QS assisted in the experiments and in the isolation of chondrocytes from human cartilage. MB designed the study, supervised the project, evaluated and interpreted the data, and prepared the manuscript. All authors read and approved the final manuscript.