CXB, one type of COX-2 inhibitor, was purchased from Pfizer Corporation Inc. (New York, NY, USA). CET was provided by Merck Serono (Darmstadt, Germany).

HSC3, an OSCC cell line, was obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen) in a humidified atmosphere of 5% CO² at 37°C.

HSC3 cells grown in monolayers were harvested and dispensed in 96-well culture plates in 100 μl of DMEM at a concentration of 5×10³ cells per well. After 24 h, different drug concentrations of CET (0–400 μg/ml), CXB (0–40 μM), or both (0–200 μg/ml CET plus 20 μM CXB) were added to the cells. Cell viability was assessed using 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) colorimetric assay at 490 nm with minor modifications according to a previous study (23). This assay was carried out in triplicate. The inhibition rate was calculated according to the following formula: Inhibition rate (%) = [1 - (average absorbance of the experimental group/average absorbance of the blank control group)] × 100%.

To evaluate the change in the apoptotic index, terminal transferase dUTP nick end labeling (TUNEL) assay was used. HSC3 cells (1×10⁵ cells/ml in 6-well plates) were cultured and treated with their respective half maximal inhibitory concentration (IC₅₀) values for CET, CXB, or both for 48 h. After fixation in 4% paraformaldehyde/PBS (pH 7.2) for 15 min at room temperature and permeabilization in a permeabilization buffer for 2–5 min, the assay was performed on cells on coverslips using a commercially available in situ apoptosis detection kit (In situ Cell Death Detection kit, POD, Roche Diagnostic, Branchburg, NJ, USA). After applying anti-FITC HRP conjugate at 37°C for 30 min, diaminobenzidine (DAB) was applied to generate an insoluble colored substrate at the site of DNA fragmentation for 10 min. Apoptosis indices of both cell types for each treatment were calculated as the percentage of cells displaying TUNEL labeling out of the total number of nuclei. In addition, we also detected caspase-3 and -8 activity by ELISA as an additional indicator of apoptosis.

The activity of caspase-3 and -8 was measured using the Caspases Colorimetric Protease Assay kits (Millipore, Billerica, MA, USA) as per the manufacturer’s instructions. Briefly, cells were treated with their respective half maximal inhibitory concentration (IC₅₀) values for CET, CXB, or both for 24 h, then washed twice with ice-cold PBS (pH 7.2) and harvested by centrifugation at 700 × g for 10 min. The cell pellets were then lysed in 150 μl buffer provided in the kit. Protein concentrations of the lysates were determined using the Lowry method. Then, an aliquot of lysates (80 μl) was incubated with 10 μl substrate of each caspase at 37°C for 2 h. Samples were analyzed at 405 nm by a microplate reader (Thermo Fisher Scientific Inc., Waltham, MA, USA).

To assess the effect of CET combination with CXB on cell migration, a wound-healing assay was performed. In brief, migration chambers (8-μm pores; Corning) were coated for 1 h at 37°C with either 10 μg/ml purified plasma fibronectin or PBS and blocked with migration buffer [3:1 of DMEM:F12 and 0.5% (w/v) BSA] for 1 h. Cells were harvested following treatment with their respective half maximal inhibitory concentration (IC₅₀) values for CET, CXB, or both for 48 h. Then, 1×10⁵ cells were resuspended in 100 μl of migration buffer, placed in the chambers and incubated at 37°C for 4 h. Non-migrated cells were removed, samples were fixed in 10% (v/v) buffered formalin, and stained with haematoxylin and mounted on glass slides. Using an inverted phase-contrast microscope (Leica DMR, Germany), the number of migrated cells was determined by counting 5 randomly selected fields.

The invasiveness of the HSC3 cells following treatment with CET, CXB or CET combined with CXB in vitro was measured using BD BioCoat™ Matrigel invasion chambers (Becton Dickinson Labware, Bedford, MA, USA) according to the manufacturer’s instructions. In brief, filters were precoated on the upper side with Matrigel provided in the kits (1 mg/ml). The lower chamber was filled with culture media containing 10% FBS. Cells (3×10⁵) treated with the indicated drugs in low serum media [3:1 of DMEM:F12 with 0.5% (v/v) FCS and 2% (w/v) L-glutamine] in the inner chamber and KGM was used as a chemo-attractant. Plates were incubated for 16 h at 37°C, and the invaded cells were observed with an immunofluorescence microscope by counting the cells that had invaded into the bottom of the cell culture insert. We also detected MMP-9 and MMP-2 protein expression by western blotting as an additional indicator of invasion and migration.

PGE2 synthesis was determined by competitive enzyme-linked immunosorbent assay (ELISA) as previously described with minor modification (23). In brief, HSC3 cells were treated with their respective half maximal inhibitory concentration (IC₅₀) values for CET, CXB, or their combination for 48 h in 12-well plates, and then the culture media were centrifuged to remove cell debris. Cell-free culture media were collected at the indicated time, then PGE2 levels were measured by competitive ELISA as described by the kit manufacturer (Cayman Chemical, Ann Arbor, MI, USA) using an ELISA reader (μQuant; BioTek Instruments, Inc., Winooski, VT, USA).

HSC3 cells were treated with their respective half maximal inhibitory concentration (IC₅₀) values for CET, CXB, or their combination for 48 h in 24-well plates, and then the culture media were centrifuged to remove cell debris. Cell-free culture media were collected at indicated time. Protein levels of vascular endothelial growth factor (VEGF) in the cell supernatant were determined by the Human VEGF ELISA kit (Yanyu, Shanghai, China) according to the manufacturer’s instructions. Samples were measured in triplicate and were properly diluted to ensure that the measured values were within the concentration range of the standard curve.

To assess in vivo the combined effect of CET and CXB on OSCC, we used an OSCC xenotransplanted nude mouse tumorigenesis model. For inoculation of OSCC cells, the detached HSC3 cells were diluted and emulsified with medium to a final cell concentration of 5×10⁷ cells/ml, and the 200 μl emulsion was then inoculated subcutaneously into the right flank of a total of 80 female BALB/c nude mice (6–8 weeks of age; weight ~200 mg; Laboratory Animal Center of Jilin University, Changchun, China). The mice were maintained in an environment complying with the NIH guidelines for the care and use of laboratory animals, following a protocol approved by the Ethics Committees of the Disease Model Research Center, Jilin University. All animals were divided into 4 groups (10 mice/group) when the volume of the ensuing mass reached 75–100 mm³. The control group received 1% polysorbate resuspended in deionized water. The other three groups were treated with CXB (4.56 mg/kg body weight), CET (1 mg/kg body weight), or CXB plus CET (2.5 mg/kg plus 0.5 mg/kg body weight, respectively) intraperitoneally on alternative days for 3 weeks. Each mouse was weighed every day to evaluate the side effects of the administrations. The lengths and widths of the tumors were measured with a caliper every 7 days, and the tumor volume (in cubic millimeters) was calculated. At the end of 7 days, the animals were euthanized using chloroform and their spleen tissues were collected and cultured for a splenocyte surveillance study (23). In addition, each tumor was excised and weighed when the mice were sacrificed. Parts of each tumor tissue were wax embedded for H&E staining to study cell apoptosis in vivo by TUNEL.

HSC3 cells were treated with their respective IC₅₀ values for CET, CXB or their combination for 48 h. The cells were then homogenized in lysis buffer (Tris-HCl 50 mmol/l, EDTA 5 mmol/l, NaCl 150 mmol/l, sodium deoxycholate 1%, Na₃VO₄ 500 μmol/l, Triton X-100 0.5%, AEBSF 10 μmol/l, NaF 10 mmol/l) on ice. The homogenates were then centrifuged at 14,000 rpm at 4°C for 30 min, and the supernatants were collected for protein concentration determination using the Bradford reagent (Sigma). Cell extracts (50 μg of protein) were separated on a sodium dodecyl sulfate-polyacrylamide electrophoretic gel (SDS-PAGE) and transferred to nitrocellulose membranes, which were blocked in 3% bovine serum albumin (BSA) for 2 h. After blocking, the membranes were incubated with primary antibodies overnight at 4°C for 2 h, and then with horseradish peroxidase-conjugated secondary antibody for 2 h at room temperature. Protein bands were visualized with enhanced chemiluminescence reagent (ECL; GE Healthcare, Velizy-Villacoublay, France). Blots were stripped and reprobed with anti-β-actin to control for loading variations. Quantity One^® Software (Bio-Rad) was used for quantification of the protein bands. For western blot analysis, the following antibodies were used: a mouse monoclonal anti-β-actin, a mouse monoclonal anti-MMP-9 and a mouse monoclonal anti-MMP-2 (Sigma Aldrich, St. Louis, MO, USA), and mouse monoclonal anti-AKT, mouse monoclonal anti-phosphorylated (p)-AKT, mouse monoclonal anti-EGFR, mouse monoclonal anti-p-EGFR, mouse monoclonal anti-PI3K, mouse monoclonal anti-p-PI3K, and horseradish peroxidase-conjugated goat anti-mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA, USA).

Data from at least three independent experiments are expressed as mean ± SD. Statistical comparison of more than two groups was performed using one-way ANOVA followed by a Tukey’s post hoc test. All statistical tests were two-sided, and a P-value of <0.05 was considered indicative of a statistically significant difference. Statistical analyses were undertaken using the SPSS^® Statistical Package, version 13.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism, version 5.01 (GraphPad^® software, San Diego, CA, USA) for Windows.

To evaluate the effect of CET, CXB, and the combination on the cell viability of OSCC cancer cells in vitro, HSC3 cells were treated with increasing concentrations of CET (0–400 μg/ml), CXB (0–40 μM), or both (0–200 μg/ml CET plus 20 μM CXB). Treatment with CTX alone resulted in an IC₅₀ value of 210.35±15.28 μg/ml. Treatment with CXB alone also resulted in an IC₅₀ value of 25.5±1.780 μM. Combination treatment (0–200 ng/ml CET plus 20 μM CXB) resulted in a leftward shift in the concentration-response curve such that the IC₅₀ value was reduced to 100.88±7.98 μg/ml. In addition, our results also demonstrated that CET, CXB and the combination inhibited cell proliferation dose-dependently Based on the results, we selected the respective IC₅₀ values of the drugs for further treatments throughout the study.

We next examined whether the combination of relatively low concentrations of CET and CXB could additively or synergistically inhibit OSCC cell proliferation in vitro with their respective IC₅₀ values for CET, CXB or the combination. As shown in Fig. 1, the inhibitory rate of the combination treatment group was higher than the rates of the single drug groups (P<0.01). There was no statistically significant difference between the CET group and the CXB group (P>0.05). In addition, the inhibitory rate of the combination treatment group was higher than that of the control group (all P<0.05).

To verify whether CET and CXB exert an anticancer effect via induction of apoptosis, apoptotic cancer cells in vitro were detected with a light microscope using a TUNEL technology. The assays were performed on HSC3 cells treated with CET or CXB alone and the combination of the two agents at their respective IC₅₀ values. HSC3 cells treated with CET, CXB or the combination had an increased percentage of apoptotic cells compared with the untreated cells (P<0.05, Fig. 2A). In addition, the low-dose combination resulted in an even higher percentage of apoptotic cells than did the higher doses of either drug alone. There was no significant difference between the CET group and the CXB group in the induction of OSCC apoptosis. These data were consistent with the results from the MTT assay, which indicated an additive effect of CET and CXB in inducing cell death via apoptosis.

To explore the possible mechanism involved in the induction of cell apoptosis of CTX in combination with CXB, caspase-3, -8 and -9 activity was determined by ELISA. The results showed that caspase-3, -8 and -9 activity was significantly increased in the treatment group compared to the control group (P<0.01, Fig. 2B–D). Importantly, caspase-3, -8 and -9 activity was significantly increased in the combination group compared to the activity in the single treatment groups (P<0.05, Fig. 2B–D).

To ascertain the inhibitory effect of CET in combination with CXB on OSCC cell motility in vitro, a wound-healing assay was performed. After 48 h, cells in the CET group, the CXB group and the combination CET and CXB group exhibited a significantly reduced migration capacity compared to the control group. Compared to the CET or the CXB group, the cells in the combination group exhibited significantly reduced migration of HSC3 cells (P<0.05; Fig. 3A).

The ability of CET in combination with CXB to reduce the invasiveness of OSCC was further investigated by the Transwell system assay. Cell invasion was also significantly decreased in the treatment groups compared to the control group (P<0.05, Fig. 3B). Compared to the CXB or the CET group, the combination treatment group greatly inhibited HSC3 cell invasion (P<0.05, Fig. 3B).

To determine the potential mechanisms involved in the inhibition of cell migration and invasion in vitro, MMP-9 and MMP-2 protein expression was examined by western blotting. Western blot analysis revealed a significant decrease in MMP-2 and MMP-9 proteins in all treatment groups compared to the control group (Fig. 3C). The combination group showed a maximally reduced expression compared to either the CET or CXB group (P<0.05, Fig. 3D).

To examine the effect of CET in combination with CXB on PGE2 production in HSC3 cells, ELISA was performed. As shown in Fig. 4A, all treatment groups inhibited PGE2 production when compared with the control group. CXB in combination with CET significantly inhibited PGE2 production compared to the single drug groups (P<0.05).

We also determined VEGF protein expression by ELISA to evaluate the effect of CET in combination with CXB on VEGF expression. As shown in Fig. 4B, ELISA analysis revealed that VEGF excretion in the supernatant from all treatment groups was significantly decreased compared to the control group (P<0.05). Compared to single drug treatments, CET in combination with CXB significantly inhibited VEGF expression (P<0.05).

The EGFR signaling pathway plays a crucial role in cell proliferation and survival in various types of cancer; thus, we evaluated the effect of CET or CXB alone and in combination on several key downstream molecules involved in the EGFR signaling pathway by western blotting. CET and CXB alone and the combination inhibited the phosphorylation of PI3K, p-AKT and p-EGFR (Fig. 5). In addition, treatment of cells with CET combined with CXB further reduced the expression levels of the activated forms of EGFR (p-EGFR), p-PI3K and p-Akt, in the HSC3 cells.

We assessed the in vivo therapeutic efficacy of CET and CXB in female BALB mice bearing HSC3 tumor cells. Mice were sacrificed and tumor tissues were excised at the end of 7 days. Tumor weight of the animals was then measured. It was found that the tumor weight of all treatment groups was lower than that of the untreated group. Compared to the single drug groups, the tumor weight of the combination group was the lowest (P<0.05, Fig. 6A). In addition, we also found that the tumor volume in the combination group was significantly lower compared with that in the single drug groups and the untreated group (P<0.05, Fig. 6B).

To assess the efficacy of CET and CXB in modulating splenocyte proliferation, MTT assay was performed. As shown in Fig. 6C, the inhibitory rate of the CET combination treatment group was higher than that of the control group and single drug groups (P<0.05). In addition, we also determined tumor tissue cell apoptosis in vivo by TUNEL. The cell apoptosis ratios of all treatment groups in vivo were significantly higher than the ratio of the control group (P<0.05, Fig. 6D). Compared to the single drug groups, the cell apoptosis ratio in the combination groups was markedly increased. Taken together, these results demonstrated that the CET and CXB combination suppressed tumor growth of oral squamous cell carcinoma in vivo.