Fetal bovine serum (FBS), Lipofectamine 2000 reagent, non-essential amino acids, RPMI-1640 medium and TRIzol reagent were purchased from Life Technologies (Grand Island, NY, USA). Esophageal cancer cell lines, Eca109, Kyse30, Kyse140, Kyse180, Kyse410, Kyse510, Kyse520, and TE-1, as well as the normal esophageal epithelial cell (NEEC) line, were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), agar, propidium iodide, puromycin, and trypsin were purchased from Sigma-Aldrich (St. Louis, MO, USA). Antibodies against α-tubulin, Axin2, cyclin D1, phosphorylated Rb or Rb were purchased from Abcam Ltd. (Cambridge, UK). Horseradish peroxidase-coupled secondary IgGs were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). ECL reagent was purchased from Pierce (Rockford, IL, USA). The luciferase Reporter Assay system, BglII, MluI and XhoI were purchased from Promega (Madison, WI, USA). pSuper-puro and pGL3-Basic Luciferase reporter plasmid were purchased from Clontech Laboratories, Inc. (Mountain View, CA, USA). miR-374a inhibitor and its negative control were purchased from Qiagen (Valencia, CA, USA).

In total, 46 patients who received esophageal cancer surgery at the China-Japan Union Hospital of Jilin University between April, 2012 and April, 2014 were selected. The diagnosis of esophageal cancer was confirmed pathologically. These patients had not previously received chemotherapy or radiotherapy. Biopsies were obtained from the esophageal cancer tissues and normal tissues adjunct to cancer area during surgery. Tissues were used for pathology to confirm the diagnosis and for mRNA extraction and RT-PCR. All 46 patients provided written informed consent. The study protocol was approved by the China-Japan Union Hospital of Jilin University.

Esophageal cancer cell lines, Eca109, Kyse30, Kyse140, Kyse180, Kyse410, Kyse510, Kyse520, TE-1, and the NEEC line were grown in RPMI-1640 medium, supplemented with heat-inactivated FBS (10% v/v), 100 U/ml penicillin, 100 µg/ml streptomycin and non-essential amino acids (1% v/v) and cultured at 37°C in an atmosphere of 5% CO₂ with a relative humidity of 95%.

The miR-374a expression plasmid was generated by cloning the pre-miR-374a gene with BglII and XhoI endonuclease sites on each flanking side into retroviral transfer plasmid pSuper-puro to generate the pSuper-puro-miR-374a plasmid. A scrambled miRNA with no homology to any known human miRNAs was used as the negative control. These plasmids were transfected into Eca109 cells using Lipofectamine 2000 reagent according to the manufacturer's instructions. After 36 h of transfection, the cells were selected using selection medium containing 1.5 µg/ml puromycin. The surviving cells, which were considered as stably transfected cells, were maintained and passaged in culture medium containing puromycin.

The transfection of Axin2 siRNA, miR-374a inhibitor and their respective control oligonucleotides into Eca109 cells was performed using Lipofectamine 2000 reagent according to the manufacturer's instructions. The sequences of siRNA and its negative control were as follows: Axin2, 5′-GCAGAGGGACAGGAATCAT-3′, and the negative control, 5′-GCAGGG ACAAGGTAGACAT-3′.

The Axin2 3′-UTR was amplified from human genomic DNA and subcloned into the pGL3-basic luciferase reporter plasmid using MluI and XhoI sites. Axin2 3′-UTR luciferase reporter plasmids were transfected into Eca109 cells that stably expressed miR-374a using Lipofectamine 2000 reagent. Axin2 3′-UTR luciferase reporter plasmids were also co-transfected with miR-374a inhibitor into intact Eca109 cells using Lipofectamine 2000 reagent. Transfected cells were serum-shocked (20%) for 2 h and maintained in 2% media for 24 h prior to cell collection. The dual-luciferase reporter assay was performed according to the manufacturer's instructions. Data were presented as relative luciferase activity.

Cell proliferation was analyzed with MTT as previously described (21). Briefly, 1×10³ cells of each group were plated in 96-well microplates in 150 µl of medium. After different time of culture, 20 µl of MTT substrate (5 mg/ml) was added to each well, and the plates were incubated for an additional 4 h. The medium was then removed, and the cells were solubilized in 150 µl of dimethyl sulfoxide. The culture plates were then agitated for 15 min and the optical absorbance values were measured at 490 nm.

Following the treatments, the cells were trypsinized and counted. Cells (1×10³) of each group were seeded in 12-well plates. The number of colonies was counted at the 5th day after seeding. The trypsinized cells were also seeded for an anchorage-independent growth ability assay as previously described (22). One thousand cells were suspended in 2 ml complete medium plus 0.33% agar. The agar-cell mixture was plated on top of a bottom layer consisting of 0.66% complete medium agar mixture. After 5 days, the colony sizes were measured using an ocular micrometer and colonies >0.1 mm in diameter were counted.

Flow cytometric analysis was performed as previously described (22). The cells were harvested by trypsinization, washed in ice-cold PBS, and fixed in 80% ice-cold ethanol. Prior to staining, the cells were pelleted in a cooled centrifuge and resuspended in cold PBS. The cells were then incubated in 20 µg/ml propidium iodide for 20 min at room temperature, and analyzed on a flow cytometer (FACSCalibur; BD Biosciences, Bedford, MA, USA).

Total RNA was extracted from human tissues and cells using TRIzol reagent, and purified using a miRNeasy column (Qiagen). The RNA was reverse transcribed into cDNA using a Reverse Transcript reagent kit. miR-374a expression was determined using TaqMan microRNA assays (Applied Biosystems, Grand Island, NY, USA). U6, a small nuclear RNA, was used as an internal control. Specific TaqMan primers, targeting mature miR-374a or U6 were obtained from Applied Biosystems. RT-PCR was performed in an ABI 7500 Real-Time PCR system. The results were presented as the relative changes of miR-374a following normalization to U6.

The protein levels of Axin2, cyclin D1 and p-Rb were assessed using western blotting as previously described (23). Whole-cell proteins were separated electrophoretically in 4–12% SDS-PAGE gels, and transferred to nitrocellulose membranes. Following 30 min of blocking with 2.5% non-fat milk, the membranes were incubated with primary antibody (dilution at 1:1,000~1:2,000) at 4°C overnight, and followed by 1-h incubation with horseradish peroxidase-conjugated secondary antibody (1:2,000). The membranes were adequately washed with PBS containing 0.5% Tween-20 subsequent to each treatment with antibody. The membranes were developed with chemiluminescence reagent and then exposed to X-ray film. The protein levels were presented as the ratio of the band optical intensity to that of α-tubulin or total Rb.

Data are presented as mean ± SEM. Statistical analysis was performed using one-way analysis of variance, followed by Dunnett's test for multiple comparisons and by the Student's t-test for comparisons between two groups. P<0.05 was considered to indicate statistical difference.

We first examined the miR-374a levels in various esophageal cancer cells using real-time PCR. The results showed that miR-374a levels were significantly increased in eight types of esophageal cancer cells, Eca109, Kyse30, Kyse140, Kyse180, Kyse410, Kyse510, Kyse520 and TE-1, as compared to the NEEC line (P<0.01, Fig. 1A). We further examined whether the increased miR-374a level existed in human esophageal cancer tissues. The mRNA was extracted from cancer and adjunct normal esophageal tissues from 46 patients with esophageal cancer, and real-time PCR was performed to measure miR-374a expression. When compared to the normal esophageal tissues, miR-374a levels in cancer tissues were increased in each pair of samples, and were significantly higher collectively (P<0.01, Fig. 1B). These results demonstrated that miR-374a expression is increased in esophageal cancer, and suggested that miR-374a is associated with the development of esophageal cancer.

To examine the role of the increased miR-374a in esophageal cancer, esophageal cancer Eca109 cells were trans-fected with miR-374a-containing vector or scrambled miRNA. The stable overexpression of miR-374a was obtained after antibiotic selection, and was confirmed by RT-PCR (Fig. 2A). The same amount of Eca109 cells, which stably overexpressed miR-374a or scrambled control, were seeded to examine the effect of miR-274a on cell proliferation. In hte MTT assay, miR-374a overexpression yielded a more rapid increase in the MTT absorbance value with culture time, compared to scramble control (Fig. 2B). In addition, miR-374a overexpression increased the relative colony number (Fig. 2C) and the colony number with size >0.1 mm (Fig. 2D). When analyzing the cell cycle using flow cytometry, we found that ectopic overexpression of miR-374a induced a significant decrease in the percentage of cells in the G1/G0 phase and an increase in the percentage of cells in the S phase (Fig. 2E). These results suggested that the upregulation of miR-374a promotes the Eca109 cell proliferation.

We introduced an underexpression model of miR-374a to further validate the promotive role of miR-374a in esophageal cancer cells proliferation. An antisense-based specific inhibitor against miR-374a was applied to inhibit the expression of miR-374a in Eca109 cells (Fig. 3A). Unlike miR-374a overexpression, miR-374a inhibition significantly decreased the MTT absorbance value with culture time, as compared to its negative control oligonucleotides (Fig. 3B). miR-374a inhibition also reduced cell colony formation (Fig. 3C and D). Additionally, flow cytometry showed a significant increase in the percentage of cells in the G1/G0 phase and a decrease in the percentage of cells in the S phase by miR-374a inhibitor, as compared to its negative control oligonucleotides (Fig. 3E). These results suggested that the underexpression of miR-374a suppressed the proliferation of Eca109 cells.

The 3′-UTRs of Axin2 have a binding site of miR-374a, which indicates Axin2 is a putative potential target of miR-374a (Fig. 4A). To confirm the relationship of miR-374a with Axin2, a luciferase reporter plasmid containing Axin2 3′-UTR was transfected into miR-374a-overexpressed Eca109 cells. In the dual-luciferase assay, miR-374a overexpression significantly decreased luciferase activity, as compared to miR-374a-scrambled Eca109 cells. When the luciferase reporter plasmids were transfected into Eca109 cells with miR-374 inhibitor, miR-374 inhibitor control oligonucleotides or miR-374a mutant, the luciferase activity was significantly increased by miR-374 inhibitor, compared to the control oligonucleotides, while the miR-374a mutant did not change the luciferase activity (Fig. 4B). Thus, these results demonstrated that miR-374a suppressed Axin2 transcription activity. Following the Axin2 protein analysis using western blotting, miR-374a overexpression was decreased, whereas miR-374a inhibition increased the Axin2 protein level (Fig. 4C). In addition, miR-374a exhibited a promotive effect on the tumor-related factors, cyclin D1 and Rb. miR-374a overexpression increased cyclin D1 expression and Rb phosphorylation, while miR-374a inhibition decreased cyclin D1 expression and Rb phosphorylation. miR-374a over-expression and miR-374a inhibition did not affect the levels of total Rb (Fig. 4D).

Axin2 is a negative regulator of the canonical Wnt/β-catenin signaling pathway, and acts as a tumor-suppressor protein in a number of human cancers (24). However, it is less involved in the role of Axin2 in esophageal cancer. To examine whether decreased Axin2 plays a tumor-suppressor role in esophageal cancer, we applied specific Axin2 siRNA to silence Axin2 in Eca109 cells, which were simultaneously treated by the miR-374a inhibitor (Fig. 5A). In the colony formation assay, Axin2 siRNA significantly increased the relative colony number (Fig. 5B) and colony number with a size of >0.1 mm (Fig. 5C), compared to the negative control of Axin2 siRNA. These results suggested that Axin2 may exert a suppressive effect on Eca109 cell proliferation.