COX-2 expression has been reported to stimulate directional migration and invasion of human cancer cells 23 24 . We used the IPTG-inducible COX-2 gene expression vector to examine the role of COX-2 in chondrosarcoma cells. JJ012 cells were transfected with IPTG-inducible COX-2 gene expression vector or control vector, and then IPTG (5 mM) was added for 24 hr. By Western blot analysis and ELISA, respectively, we found that IPTG induced COX-2 and PGE₂ expression (Fig. 1A&1B). Furthermore, over-expression of COX-2 enhanced cell migration in chondrsarcoma cells (Fig. 1C). To confirm IPTG-inducible COX-2-mediated cell migration, the COX-2 specific inhibitors (celebrex and NS-398) were used. Celebrex and NS-398 but not COX-1 specific inhibitor (valeryl salicylate) reduced IPTG-inducible COX-2-mediated cell migration (Fig. 1D). We then directly exposed JJ012 cells to PGE₂ and examined the migration activity. Stimulation of cells with PGE₂ increased the migration activity in chondrosarcoma cells dose-dependently (Fig. 1E). We also examined human chondrosarcoma tissues for the expression of the COX-2 using qPCR. Expression of mRNA levels of COX-2 in human chondrosarcoma tissues (Fig. 1F, lines 5-8) and chondrosarcoma cell lines (SW1353 and JJ012) were significantly higher than those in normal cartilage (Fig. 1F, lines 1-4). Compared with normal cartilage, human chondrosarcoma tissues expressed a higher level of COX-2 mRNA (Table 1). In addition, primary chondrosarcoma cells and SW1353 or JJ012 cell lines were more migratory than normal chondrocyte (Fig. 1G). Thus, expression of COX-2 was associated with a metastatic phenotype of chondrosarcoma cells.

PGs exert their effects through interaction with specific EP1-4 subtype receptors 10 11 12 13 14 . To investigate the role of EP1-4 subtype receptors in COX-2-mediated increase of cell migration, we assessed the distribution of these EP subtype receptors in human chondrosarcoma cells by qPCR analysis. The mRNAs of EP1, EP2, EP3, and EP4 subtype receptors could be detected in human chondrosarcoma cells (Fig. 2A). After IPTG/COX-2-transfected JJ012 cells were treated for 24 hr with IPTG, the mRNA level of EP1 subtype receptor was increased, whereas EP2 and EP4 receptor mRNA remained un-changed (Fig. 2A). In addition, a similar induction of EP1 receptor mRNA, but not EP2 and EP4 receptor subtypes, was observed in JJ012 cells treated with PGE₂ (Fig. 2B). However, over-expression of COX-2 and exogenous PGE₂ slightly increased expression of EP3 receptor (Fig. 2A&2B). On the other hand, the mRNA levels of EP1 receptor in human chondrosarcoma tissues (Fig. 2C, lines 5-8) and chondrosarcoma cell lines (SW1353 and JJ012) were significantly higher than those in normal cartilage (Fig. 2C, lines 1-4). Compared with normal cartilage, human chondrosarcoma tissues expressed a higher level of EP1 mRNA (Table 1). To determine the role of EP1 receptor-dependent signaling in the regulation of cell migration in chondrosarcoma cells, the cells were treated with EP1-4-specific agonists, and then the cell migration activity was examined. Of the agonists tested, only the EP1/EP3-selective receptor agonist, 17-phenyl trinor PGE₂ (3 μM), significantly increased the migration activity (Fig. 2D). In contrast, butaprost (EP2 agonist; 10 μM) and 11-deoxy-PGE₁ (EP2/EP4-selective agonist; 10 μM) failed to up-regulate cell migration. Sulprostone (EP3 agonist; 10 μM) slightly increased cell migration in JJ012 cells (Fig. 2D). In addition, treatment with EP1 receptor antagonist SC19220 (10 μM) effectively antagonized the potentiating effect of PGE₂ on cell migration activity (Fig. 2D). To further confirm this stimulation-specific mediation by EP1 receptor without EP3 receptor contamination, we assessed the role of EP1 and EP3 by using ON-TARGET smart pool EP1 and EP3, which decreases nonspecific effects by chemical modification and pooling 26 . Transfection of cells with ON-TARGET smart pool EP1 and EP3 siRNA reduced EP1 and EP3 expression, respectively (Fig. 2E; upper panel). Transfection of cells with EP1 but not EP3 siRNA effectively inhibited the PGE₂-mediated migration of chondrosarcoma cells (Fig. 2E; lower panel). These results indicate that PGE₂ increased cell migration in human chondrosarcoma cells via EP1 receptor.

Previous studies have demonstrated significant expression of integrins in human chondrosarcoma cells 27 . We therefore, hypothesized that integrins may be involved in PGE₂-directed migration of chondrosarcoma cells. Flow cytometry analysis showed that PGE₂ induced the cell surface expression of α2 and α2β1 integrin in JJ012 cells (Fig. 3A). To confirm this finding, expression of mRNAs in the integrins in response to PGE₂ was analyzed by qPCR. Treatment of JJ012 cells with PGE₂ induced the mRNA expression of α2 and β1 integrins (Fig. 3B). In addition, treatment of IPTG/COX-2-transfected cells with IPTG increased mRNA expression of α2 and β1 integrins (Fig. 3C). Furthermore, compared with normal cartilage, human chondrosarcoma tissues expressed higher levels of α2 and β1 integrin mRNA (Table 1). Therefore, the α2β1 integrin plays an important role in PGE₂-induced migration of human chondrosarcoma cells. Stimulation of cells with PGE₂ also increased mRNA expression and cell surface expression of α2 and β1 integrin time-dependently (Fig. 3D&3E). Pretreatment of cells for 30 min with anti-α2β1 monoclonal antibody (mAb) (3 μg/ml) markedly inhibited the PGE₂-induced cell migration (Fig. 3F). On the other hand, EP1/3 agonist enhanced the cell surface expression of α2β1 integrin (Fig. 3G). Pretreatment of cells with SC19220 reduced PGE₂-mediated α2β1 integrin expression (Fig. 3G). These data suggest that PGE₂-induced cancer migration may occur via activation of the α2β1 integrin.

It has been reported that PLC/PKC/c-Src dependent pathway is involved in EP1-mediated bone formation 28 . We therefore directly measured the phosphorylation of PLC, PKC and c-Src in response to PGE₂. Treatment of JJ012 cells with PGE₂ induced the phosphorylation of PLCβ3, PKCα and c-Src time-dependently (Fig. 4A). In addition, PKCα activity was also increased by PGE₂ treatment of human chondrosarcoma cells time-dependently (Fig. 4E). Furthermore, pretreatment of cells with PI-PLC inhibitor (U73122), PKC inhibitor (GF109203X) and c-Src inhibitor (PP2) reduced PGE₂-increased cell migration and integrin up-regulation (Fig. 4B&4D). Transfection of cells with PKCα and c-Src mutant or PLCβ siRNA also inhibited PGE₂-mediated migration activity (Fig. 4C). Transfection of cells with PLC siRNA reduced PLC expression (Fig. 4C; upper panel). Based on these results, it appears that PGE₂ acts through the PLCβ, PKCα and c-Src dependent signaling pathway to enhance α2β1 integrin expression and cell migration in human chondrosarcoma cells.

As previously mentioned, NF-κB activation is necessary for the migration and invasion of human chondrosarcoma cells 27 . To examine whether NF-κB activation is involved in PGE₂-induced cancer migration, an NF-κB inhibitor, PDTC, was used. Fig. 5A&5B show that chondrosarcoma cells pretreated with PDTC inhibited the PGE₂-induced migration and integrin expression of chondrosarcoma cells. Furthermore, cells pretreated with TPCK (3 μM), an IκB protease inhibitor, also reduced PGE₂-induced migration of cancer cells (Fig. 5A&5B). Therefore, the NF-κB pathway has a role in PGE₂ induced migration of chondrosarcoma cells. We further examined the upstream molecules involved in PGE₂-induced NF-κB activation. Stimulation of cells with PGE₂ induced IKKα/β phosphorylation in a time-dependent manner (Fig. 5C). Furthermore, transfection with IKKα or IKKβ mutant markedly inhibited the PGE₂-induced cell migration (Fig. 5D). These data suggest that IKKα/β activation is involved in PGE₂-induced the migration activity of human chondrosarcoma cells. Treatment of chondrosarcoma cells with PGE₂ also caused IκBα phosphorylation in a time-dependent manner (Fig. 5C). Previous studies showed that p65 Ser⁵³⁶ phosphorylation increases NF-κB transactivation 29 . Therefore, the antibody specific against phosphorylated p65 Ser⁵³⁶ was employed to examine p65 phosphorylation. Treatment of cells with PGE₂ for various time intervals resulted in p65 Ser⁵³⁶ phosphorylation (Fig. 5C). To directly determine NF-κB activation after PGE₂ treatment, chondrosarcoma cells were transiently transfected with κB-luciferase as an indicator of NF-κB activation. As shown in Fig 5E, PGE₂ treatment of chondrosarcoma cells for 24 hr caused increase in κB-luciferase activity. In addition, U73122, GF109203X, PP2, PDTC and TPCK or PLCβ siRNA, PKCα, IKKα and IKKβ mutant reduced PGE₂-mediated NF-κB activity (Fig. 5E). Furthermore, U73122, GF109203X and PP2 reduced PGE₂-mediated p65 phosphorylation (Fig. 5F). Taken together, these data suggest that activation of EP1 receptor, PLC, PKC and c-Src pathway is required for PGE₂-induced NF-κB activation in chondrosarcoma cells.

Anti-mouse and anti-rabbit IgG-conjugated horseradish peroxidase, rabbit polyclonal antibodies specific for PLCβ, PKCα, c-Src, IKK, p-IκBα, IκBα, p65 and the siRNAs against PLCβ and control (negative control for experiments using targeted siRNA transfection; each consisted of a scrambled sequence that would not lead to the specific degradation of any known cellular mRNA) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rabbit polyclonal antibody specific for p-IKKα/β, p-PLCβ3, p-PKCα, p-c-Src and p-p65 were purchased from Cell Signaling and Neuroscience (Danvers, MA). Rabbit polyclonal antibodies specific for α2, α5, β3, αvβ3, α5β1 and α2β1 integrin were purchased from Chemicon (Temecula, CA). PGE₂, 17-phenyl trinor PGE₂, butaprost, sulprostone, 11-deoxy-PGE₁, SC19220 and rabbit polyclonal antibody specific for COX-2, EP1 and EP3 were purchased from Cayman Chemical (Ann Arbor, MI). Valeryl salicylate, NS398, U73122, GF109203X, PP2, PDTC, TPCK and IPTG (isopropyl-β-D-thiogalactopyranoside) were obtained from Calbiochem (San Diego, CA). Celebrex was obtained from Pharmacia Co. (Piscataway, NJ). ON-TARGET smart pool EP1 and EP3 siRNA and ON-TARGET plus siCONTROL Nontargeting pool siRNA were purchased from Dharmacon. The COX-2 IPTG-induced expression plasmid, p-NLR-COX2 was a gift from Dr. W.M. Fu (National Taiwan University) 41 . The IKKα(KM) and IKKβ(KM) mutants were gifts from Dr. H. Nakano (Juntendo University, Tokyo, Japan). The PKCα dominant negative mutant was a gift from Dr. V. Martin (Louis Pasteur de Strasbourg University, France). The c-Src dominant negative mutant was a gift from Dr. S. Parsons (University of Virginia Health System, Charlottesville, VA). The NF-κB luciferase plasmid was purchased from Stratagene (La Jolla, CA) and luciferase assay kit was purchased from Promega (Madison, MA, USA). All other chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA).

The human chondrosarcoma cell line (JJ012) was kindly provided by the laboratory of Dr. Sean P Scully (University of Miami School of Medicine, Miami, FL, USA). The JJ012 cells were cultured in DMEM/α-MEM supplemented with 10% fetal bovine serum (FBS) and maintained at 37°C in a humidified atmosphere of 5% CO₂. The human chondrosarcoma cell line (SW1353) was obtained from the American Type Culture Collection. The cells were cultured in DMEM/α-MEM supplemented with 10% FBS and maintained at 37°C in a humidified atmosphere of 5% CO₂.

The migration assay was performed using Transwell (Costar, NY; pore size, 8-μm) in 24-well dishes. Before the migration assay was performed, cells were pretreated for 30 min with different concentrations of inhibitors, including the U73122, GF109203X, PP2, PDTC, TPCK or vehicle control (0.1% DMSO). Approximately 1×10⁴ cells in 100 μl of serum-free medium were placed in the upper chamber, and 300 μl of the same medium containing PGE₂ was placed in the lower chamber. The plates were incubated for 24 h at 37°C in 5% CO₂, then cells were fixed in methanol for 15 min and stained with 0.05% crystal violet in PBS for 15 min. Cells on the upper side of the filters were removed with cotton-tipped swabs, and the filters were washed with PBS. Cells on the underside of the filters were examined and counted under a microscope. Each clone was plated in triplicate in each experiment, and each experiment was repeated at least three times. The number of migrating cells in each experiment was adjusted by the cell viability assay to correct for proliferation effects of the PGE₂ treatment (corrected migrating cell number = counted migrating cell number/percentage of viable cells) 42 .

Human chondrosarcoma cells were plated in six-well dishes. The cells were then washed with PBS and detached with trypsin at 37°C. Cells were fixed for 10 min in PBS containing 1% paraformaldehyde. After rinsing in PBS, the cells were incubated with rabbit anti-human antibody against α2, α5, β3, αvβ3, α5β1 or α2β1 integrin (1:100) for 1 h at 4°C. Cells were then washed again and incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit secondary IgG (1:150; Leinco Tec. Inc., St. Louis, MO, USA) for 45 min and analyzed by flow cytometry using FACS Calibur and CellQuest software (BD Biosciences, Palo Alto, CA, USA).

The cellular lysates were prepared as described previously 43 . Proteins were resolved on SDS-PAGE and transferred to Immobilon polyvinyldifluoride (PVDF) membranes. The blots were blocked with 4% BSA for 1 hr at room temperature and then probed with rabbit anti-human antibodies against p-PLCβ3, p-PKCα, p-c-Src or p-p65 (1:1000) for 1 hr at room temperature. After three washes, the blots were subsequently incubated with a donkey anti-rabbit peroxidase-conjugated secondary antibody (1:1000) for 1 hr at room temperature. The blots were visualized by enhanced chemiluminescence using Kodak X-OMAT LS film (Eastman Kodak, Rochester, NY, USA).

PKC activity was assessed by a PKC Kinase Activity Assay Kit according to manufacturer's instructions (Assay Designs, MI). The PKC activity kit is based on a solid-phase ELISA that uses a specific synthetic peptide as a substrate for PKC and a polyclonal antibody that recognized the phosphorylated form of the substrate.

The chondrosarcoma cells were transfected with reporter plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer's recommendations. Twenty-four hours after transfection, the cells were treated with inhibitors for 30 min, and then PGE₂ or vehicle was added for another 24 hr. Cell extracts were then prepared, and luciferase and β-galactosidase activities were measured 42 44 .

Total RNA was extracted from cancer cells using a TRIzol kit (MDBio, Taipei, Taiwan). The reverse transcription reaction was performed using 2 μg of total RNA that was reversely transcribed into complementary DNA using oligo(dT) primer. The quantitative real time PCR (qPCR) analysis was carried out using a Taqman^® one-step PCR Master Mix (Applied Biosystems, CA, USA). One hundred ng of total cDNA were added per 25-μl reaction with sequence-specific primers and Taqman^® probes. Sequences for all target gene primers and probes were purchased commercially [GAPDH was used as internal control (Applied Biosystems, CA, USA)]. qPCR assays were carried out in triplicate with an StepOnePlus sequence detection system. The cycling conditions were 10-min polymerase activation at 95°C followed by 40 cycles at 95 °C for 15 s and 60 °C for 60 s. The threshold was set above the non-template control background and within the linear phase of target gene amplification in order to calculate the cycle number at which the transcript was detected (denoted C_T).

Upon approval by the local ethics committee, specimens of tumor tissue or normal cartilage tissue were obtained from patients, who had been pathologically diagnosed with chondrosarcoma or knee osteoarthritis (the articular cartilage was collected) and had undergone surgical resection at the China Medical University Hospital. The chondrosarcoma patient group consisted of 6 females and 11 males and ranged in age from 22 to 68 years (mean, 48 years). The histologic grade of chondrosarcoma was checked. Cytologically, increased cellularity and cytological atypia are the most important features, and these characteristics are used to determine the grade of the chondrosarcoma 45 . Eleven patients were alive without disease, one was alive with disease, two died without disease, and three died of causes secondary to progressive disease. The overall survival estimate at 5 years was 90%. The recurrence rate for patients with adequate surgical margins was 10%, compared with 75% for patients with inadequate margins. Tissue specimens were ground and then sonicated in a TRIzol kit. The mRNA level was analyzed using qPCR analysis.

For primary cultures from human chondrosarcoma cells, the samples were cleaned of surrounding fat, connective tissue and blood vessels. Thereafter, tissue samples were minced into pieces using a razor blade. Minced samples were transferred into 50 ml Falcon tubes, spun down at 1000 r.p.m. for 5 min and rinsed twice with fresh PBS. Digestion was performed with 1 mg collagenase II/ml PBS at 37°C for 50 min in a shaking water bath. Cell suspension was pipetted up and down at least twice during incubation. After digestion, pure FBS was added to a minimum concentration of 10% to inactivate the collagenase, followed by a centrifugation step as described above. Cells were then cultured in Dulbecco's modified Eagle's medium/α-modified Eagle medium supplemented with 10% FBS and maintained at 37°C in a humidified atmosphere of 5% CO₂ 46 .

Primary cultures of human chondrocytes were isolated from articular cartilage as described previously 47 . The cells were grown in plastic cell culture dishes in 95% air-5% CO2 with Dulbecco's modified Eagle's medium which was supplemented with 10% FBS, 2 mM-glutamine, penicillin (100 U/ml) and streptomycin (100 μg/ml).

The values given are means ± S.E.M. The significance of difference between the experimental groups and controls was assessed by Student's t test. The difference was considered significant if the p value was < 0.05.