All experiments were performed in the Central Laboratory of the North Sichuan Medical College (Sichuan, China). Human NPC 5–8F cells were provided by the Shanghai Institute for Biological Sciences (Shanghai, China). Cells were cultured in RPMI 1640 medium (Hyclone; GE Healthcare Life Sciences, Logan, UT, USA) with 10% fetal bovine serum (FBS; Hyclone; GE Healthcare Life Sciences) and maintained with 5% CO₂ at 37°C. A total of 20 female BALB/C athymic nude mice, aged 4–6 weeks and weighing 18–20 g, were purchased from the Institute of Zoology, Chinese Academy of Sciences (Beijing, China) and housed in a sterile environment at 25°C with a 12-h light/dark cycle and 40% humidity, and food and water sterilized by high pressure used to feed the mice ad libitum. Curcumol was dissolved to a concentration of 10 mg/ml in absolute ethyl alcohol and then diluted with the 10% FBS RPMI-1640 medium (adjustment based on the required concentration of the drug). Cells without Curcumol treatment (0 µΜ/ml) served as a control group. Animal experiments were approved by the Affiliated Hospital of North Sichuan Medical College ethics committee (CBYXY0022).

Curcumol (cat. no. 100185-200506) was purchased from MedChemExpress (Monmouth Junction, NJ, USA). A cell counting kit-8 (CCK-8) assay kit was obtained from Dojindo Molecular Technologies, Inc. (Kumamoto, Japan). Dimethyl sulfoxide (DMSO) was purchased from Beyotime Institute of Biotechnology (Haimen, China) and used for the storage of cells. The purified recombinant human TGF-β1 powder (R&D Systems, Inc., Minneapolis, MN USA) was dissolved in 4 mM HCl containing 1 mg/ml bovine serum albumin (Shanghai Yi Sheng Biotechnology Co., Ltd., Shanghai, China; storage concentration: 50 ng/ml). Curcumol was diluted with media to different concentrations prior to use. Matrigel was purchased from BD Biosciences (Franklin Lakes, NJ, USA). Transwell inserts were provided by Corning Costar Co., Ltd. (Corning, NY, USA). Rabbit polyclonal antibodies against E-cadherin (cat. no. 3195), N-cadherin (cat. no. 13116) and β-actin (cat. no. 3700) were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Goat anti-TGF-β1 (cat. no. TA130016), rabbit monoclonal antibody for β-actin (cat. no. TA130008), horseradish peroxidase (HRP)-conjugated goat anti-mouse (cat. no. TA130010) and goat anti-rabbit (cat. no. TA130020) immunoglobulin G were purchased from OriGene Technologies, Inc. (Rockville, MD, USA). The enhanced chemiluminescence (ECL) reagent was purchased from EMD Millipore (Billerica, MA, USA). The reagents for quantitative polymerase chain reaction (qPCR) were obtained from Takara Biotechnology Co., Ltd. (Dalian, China). Other chemicals were of standard analytical grade.

RPMI-1640 culture medium without FBS was used as the primary solvent with the following treatments: i) Control group, cells without Curcumol (0 µM/ml) treatment; ii) 0.1 µM/ml Curcumol group; iii) 0.2 µM/ml Curcumol group; iv) 0.4 µM/ml Curcumol group. Following mixing of the medium, serum and Curcumol according to the concentration requirement, cells were cultured in 37°C with 5% CO₂.

Cell viability was detected using a CCK-8 assay. Cells were seeded in 100 µl complete medium in 96-well plates at a density of 5×10⁵/ml and triplicate wells were used for each group. Following 8 h incubation at 37°C, cells were exposed to different doses of Curcumol for 48 h in 96-well plates. A solution of 10 µl CCK-8 and 90 µl FBS-free medium was added to each well and incubated for a further 10 min at 37°C. A wavelength of 450 nm was then used to measure cell viability.

Curcumol-induced cell apoptosis was detected by FCM using an Annexin V-fluorescein isothiocyanate (FITC) apoptosis detection kit (Beyotime Institute of Biotechnology). Briefly, following conditions of cell culture described above, all cells were harvested and washed with cold PBS twice and then treated with different concentrations of Curcumol for 48 h at 37°C. Cells were stained with Annexin propidium iodide and V-FITC following centrifugation (100 × g for 5 min at 25°C) and using a flow cytometer and BD CellQuest Pro software version 5.1 (BD Biosciences, Franklin Lakes, NJ, USA).

A wound-healing assay was performed using 5–8F treated cells. A total of 8×10⁵ cells were incubated in 6-well plates and grown to confluence. An artificial homogenous wound was created in the cell monolayer with a sterile 100 µl tip and was then washed with PBS to remove cell debris. Cells were then treated with different concentrations of Curcumol for 24 and 48 h. The wound healing rate following treatment was evaluated using ImageJ software version k1.45 (National Institutes of Health, Bethesda, MD, USA). Transwell chambers (8 µm pore size; EMD Millipore) with Matrigel were used to detected cell invasion. Cells (1×10⁴) were treated with Curcumol for 48 h and were then plated in the top chambers with an 8 µm membrane pore in FBS-free medium. The medium in the bottom chamber was supplemented with 10% FBS to allow for invasion for 24 h at 37°C. At the end of the experiment, cells at the bottom of the membrane were fixed in 4% paraformaldehyde for 10 min at 37°C and stained with 1% crystal violet at 25°C (Sigma-Aldrich; Merck KGaA, Darmstadt Germany) for 10 min. Images were captured using a light microscope at a magnification of ×100. The specific number of cells per filter in five predetermined fields was then counted and analyzed.

Total RNA was extracted from 5–8F cells using TRIzol (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), according to the manufacturer's protocol. RNA was then reverse transcribed into cDNA using an M-MLV kit (Takara Biotechnology Co., Ltd.). All PCR primers were as follows: E-cadherin, forward, 5′-GGTTATTCCTCCCATCAGCT-3′ and reverse, 5′-CTTGGCTGAGGAGGGTGTA-3′; N-cadherin, forward, 5′-GGTGGAGGAGAAGAAGACCAG-3′ and reverse, 5′-GGCATCAGGCTCCACAGT-3′; GAPDH, forward, 5′-TTCGTCATGGGTGTGAAC-3′ and reverse, 5′-AGTGAGCTTCCCGTTCAGC-3′. The qPCR assay (95°C for 30 sec, then 40 cycles of 95°C for 5 sec and 60°C for 30 sec) was performed as the instruction in triplicate using SYBR Green PCR Master mix and Applied Biosystems Prism 7500 (Applied Biosystems; Thermo Fisher Scientific, Inc.). The relative fold change among groups was analyzed using 2^−ΔΔCq quantitative analysis (20).

Cells treated with Curcumol were lysed in radioimmunoprecipitation lysis buffer (Beyotime Institute of Biotechnology) on ice for 30 min. Cells were then centrifuged at 100 × g for 15 min at 4°C and the protein concentration was determined using a BCA protein assay kit (Beyotime Institute of Biotechnology). Equal amounts (25 µg) of total protein samples were separated using 10% SDS-PAGE and different protein bands were transferred to polyvinylidene fluoride membranes (EMD Millipore). Membranes were blocked using 0.1% TBST with 5% non-fat dry milk at room temperature for 1 h and incubated with E-cadherin (dilution: 1:1,000), N-cadherin (dilution: 1:1,000) and anti-β-actin (dilution: 1:4,000) at 4°C overnight. Membranes were then washed using 0.1% TBST four times and incubated with HRP-conjugated secondary antibodies for 1.5 h at room temperature. The target protein was detected by ECL and the bands were quantified using ImageJ software version k1.45.

Pretreated serum samples and cell supernatants were dissolved on ice and a TGF-β1 ELISA kit (cat. no. ab100647; Abcam, Cambridge, UK) was used to measure the TGF-β1 concentrations in accordance with the manufacturer's protocols. Each sample was run in triplicate. Standards, serum samples or cell supernatants were added to 96-well plates, respectively. Subsequently, the enzyme conjugation solutions were added and the mixtures were co-incubated for 60 min at 37°C. Every well was washed 5 times with wash buffer. Substrate I and II were then added to each well respectively and the plate was kept away from light at room temperature for 15 min. The plate was then read at a wavelength of 450 nm using an ELISA reader.

NPC 5–8F cells (1×10⁷) were resuspended with 200 µl PBS and Matrigel and subcutaneous injections were performed on both sides of the nude mice following anesthesia with ether. Mice were randomly assigned to a control group and a treated group when the tumors reached an average volume of 50 mm³ (volume=length × width 2/2). The treated group were administered with Curcumol at 15 µg/kg crude drug (sterile water solution, equivalent to the human dosage) by gavage twice a day for 35 days and the control group mice were given an equal volume of sterile water. The tumor volume was measured every 7 days. Mice were sacrificed by cervical dislocation and tumors and peripheral blood were collected. Following weighing of the tumor, the tumors were stored at −8°C. In addition, blood samples were centrifuged at 100 × g for 15 min at 25°C in order to obtain the serum and were then stored at −80°C.

Data are presented as the mean ± standard error of the mean. Each experiment was repeated at least three times. Data were analyzed using SPSS 13.0 (SPSS Inc., Chicago, IL, USA). A Student's t-test was used to analyze differences between two groups, while the differences between >2 groups were detected using one-way analysis of variance followed by Student-Newman-Keuls test. P<0.05 was considered to indicate a statistically significant difference.

The effect of Curcumol on NPC 5–8F cell viability was investigated by determining the growth of 5–8F cells using a CCK-8 assay. The results demonstrated that Curcumol significantly inhibited cell growth at doses of 0.1, 0.2 and 0.4 µM/ml (Fig. 1A). 5–8F cell apoptosis was quantified using FCM. The results indicated that the rate of apoptosis was significantly increased following Curcumol treatment in a dose-dependent manner (Fig. 1B and C).

A wound-healing assay and Transwell assays were performed to investigate whether Curcumol treatment affected the migration and invasion ability of NPC cells (Fig. 2). The migration ability of cells was significantly decreased with an increase of the concentration of Curcumol compared with that in the control group (Fig. 2A and C). Cell invasion was significantly decreased in a dose-dependent manner at different concentrations of Curcumol treatment (Fig. 2B and D).

The mechanism of action of Curcumol on inhibition of 5–8F cell development was investigated by performing RT-qPCR and western blotting (Fig. 3). The results of RT-qPCR suggested that the expression of E-cadherin mRNA was significantly increased in 5–8F cells treated with Curcumol at doses of 0.1, 0.2 and 0.4 µM/ml (Fig. 3A). By contrast, the expression of N-cadherin was significantly inhibited in 5–8F cells treated with Curcumol in a dose-dependent manner (Fig. 3B). The results of western blotting were consistent with this (Fig. 3C-F). Thus, this suggests that Curcumol may regulate EMT-associated proteins in a dose-dependent manner.

Based on the results of experiments on 5–8F cells, further investigation was performed into the effect of Curcumol on NPC in vivo. The results indicated that the tumor volume of the Curcumol group on day 35 was significantly decreased compared with that in the control group (Fig. 4A and B). The tumor weight of the control group was also significantly increased compared with that in the Curcumol group (Fig. 4C), suggesting that Curcumol significantly inhibited the growth of tumors in vivo.

Tumors and serum were extracted from nude mice and western blotting and ELISA assays were performed to detect the expression of EMT-associated proteins and the secretion of TGF-β1. The expression of E-cadherin was significantly increased in the Curcumol group compared with the control group (Fig. 5A), while the expression of N-cadherin was significantly decreased in the Curcumol group (Fig. 5B). The expression of TGF-β1 in tumors and serum was measured by western blotting and ELISA. The concentration of TGF-β1 was significantly decreased in the Curcumol group compared with the control group (Fig. 5C) and the results of tumor analysis were consistent with this (Fig. 5D).