A total of 20 pairs of ESCC tumor tissues and normal adjacent tissues were used in the present study. Tissues were isolated from newly diagnosed ESCC patients who had not received chemotherapy or radiotherapy. A total of 20 patients (8 male, 12 female; mean age, 57.6 years) were recruited from the Department of Cardiothoracic Surgery in the Third Affiliated Hospital of Soochow University (Changzhou, China) from March 2012 to June 2013. The diagnosis was confirmed based on histological examination according to the American Joint Committee on Cancer 2010 TNM staging system for EC (23,24). The samples were collected with written informed consent from the patients and stored in liquid nitrogen until assays were performed. The study protocol was approved by the Ethics Board of The Third Affiliated Hospital of Soochow University.

The human ESCC cell line EC9706 was purchased from American Type Culture Collection (Manassas, VA, USA). The cells were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% (V/V) fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.) in a humidified hood with 5% CO₂ at 37°C.

The cells were seeded into 96-well plates to 60% confluency in RMPI-1640 media supplemented with 10% FBS (antibiotic-free medium) for 24 h at 37°C prior to transfection. Lipofectamine 2000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was used to transfect 1,000–10,000 cells/well with 50 nM miR-148a mimic. The experiment was performed in triplicate for each group. Non-homologous RNA duplex was used as a negative control (NC) and empty vector was used as a blank control (BC). The nucleotides were chemically synthesized by Shanghai GenePharma Co., Ltd. (Shanghai, China). The nucleotide sequences used were as follows: For miR-148a mimic, forward, 5′-UCAGUGCAUGACAGAACUUGG-3′ and reverse, 5′-AAGTTCUGUCAUGCACUGAUU-3′; and for NC, forward, 5′-UUCUCCGAACGUGUCACGUTT−3′ and reverse, 5′-ACGUGACACGUUCGGAGAATT-3′. The cells were harvested for reverse transcription-quantitative polymerase chain reaction (RT-qPCR), western blot analysis and apoptosis assays 72 h after transfection under 37°C.

Total RNA was extracted from tissues and cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Synthesis of cDNA was performed using a PrimeScript II First Strand cDNA Synthesis kit (Takara Biotechnology Co., Ltd., Dalian, China). Target gene DNA was amplified with specific primers and a Fast SYBR Green Master mix (Applied Biosystems; Thermo Fisher Scientific, Inc.). Mature miR-148a, U6 control and HLA-G were detected using the primers reported previously (14). The primers for the internal control GAPDH were: Forward, 5′-AGCGCGTGCCTTATACCAAG-3′ and reverse, 5′-GCCGCTCAGAGAGATTCGT-3′. All amplifications were conducted with a 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The thermocycling conditions were as follows: 94°C for 1 min, followed by 45 cycles of 94°C for 20 sec, 58°C for 30 sec and 72°C for 30 sec. Relative mRNA expression was calculated using the ΔΔC_q method as described previously (25).

Cells were lysed with RIPA lysis buffer (R0020, Beijing Solarbio Science and Technology Co., Ltd). Protein was isolated by centrifuging the sample at 20,000 × g at 4°C for 15 min. Protein concentration was determined using a BCA Protein assay kit (Beyotime Institute of Biotechnology, Haimen, China). Each lane was filled with 1 ng protein and 10 µl 6× SDS loading dye. Proteins were separated on 12.5% SDS-PAGE gels and transferred to polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA. The membrane was blocked with 5% skimmed milk at room temperature for 1 h, then probed with monoclonal mouse anti-HLA-G (sc-21799; Santa Cruz Biotechnology, Inc., Dallas, TX, USA; 1:1,000), or monoclonal mouse anti-GAPDH (sc-32233; Santa Cruz Biotechnology, Inc.; 1:1,000) for 1 h at room temperature. The secondary antibody used was horseradish peroxidase-conjugated rabbit anti-mouse Immunoglobulin G antibody (6120-05; SouthernBiotech, Birmingham, AL, USA; 1:2,000) at room temperature for 45 min. Signals were detected using an enhanced chemiluminescence kit (BH4004, Shanghai Ke Min Biotechnology Co., Ltd., Shanghai, China).

Cells (~5×10⁵) were harvested and washed with phosphate-buffered saline. Cells were stained with Annexin V-fluorescein isothiocyanate (SouthernBiotech) for 5 min at room temperature. A final concentration of 10 µg/ml propidium iodide was added for 30 min at 4°C, away from light, then flow cytometry was performed. Flow cytometry results were analyzed using CytoSpec version 7 (Purdue University Cytometry Laboratories, West Lafayette, IN, USA).

TargetScan version 7.1 (http://www.targetscan.org/vert_71) online software was used to predict the miRNA that could target HLA-G.

Results were analyzed using SPSS 16.0 software (SPSS, Inc., Chicago, IL, USA). Expression data regarding patients were expressed as the median (with maximum-minimum and interquartile range), and all other continuous variables were expressed as the mean ± standard deviation. In vitro experiments were repeated at least three times. Comparisons among groups were analyzed using one-way analysis of variance and the least significant difference post hoc test. All tests were two-tailed and P<0.05 was considered to indicate a statistically significant difference.

To investigate the tumor-specific expression pattern of miR-148a, the expression of miR-148a was assessed in 20 pairs of primary ESCC tissues and matched adjacent normal tissues. As depicted in Fig. 1, expression of miR-148a was significantly reduced to ~30% in the tumor tissue samples when compared with normal adjacent tissues (P<0.01). Therefore, a lower level of miR-148a expression was indicated to occur in ESCC.

Previous studies have indicated that HLA-G expression level may be regulated by miR-148a, and that miR-148b and HLA-G were upregulated in primary ESCC tissues (13,16). Based on these reports, it was hypothesized that miR-148a may contribute to the proliferation of ESCC cells by regulating HLA-G expression. Therefore, the ESCC cell line EC9706 was transfected with miR-148a mimic to identify the potential regulatory role of miR-148a.

Position 376–382 of the HLA-G gene 3′-untranslated region is predicted as the binding site of miR-148a according to the TargetScan, as depicted in Fig. 2A. RT-qPCR confirmed that the level of HLA-G mRNA decreased following transfection with miR-148a mimic (Fig. 2B). Compared with cells transfected with nonsense (NC) or empty (BC) vector, the transcription level of HLA-G was significantly decreased (to a relative expression of ~28%) in cells transfected with miR-148a mimic (P<0.01 vs. BC). Accordingly, western blot analysis demonstrated that the level of HLA-G protein was reduced in cells transfected with miR-148a mimic when compared with NC or BC cells (Fig. 2C). These results confirmed that HLA-G expression was regulated by miR-148a.

To further demonstrate whether miR-148a regulated the growth of cells, the rates of cell apoptosis were compared between the different cell groups. As depicted in Fig. 3, compared with the BC group, transfection with miR-148a significantly increased the rate of apoptosis in EC9706 cells (P<0.01); the apoptotic rate was ~13-fold higher in the cells transfected with miR-148a mimic. Rates of apoptosis did not differ between the NC and BC groups, indicating that the increase in apoptosis was due to transfection with miR-148a mimic. Collectively these results suggest that miR-148a may trigger apoptosis in ESCC cells. This pro-apoptotic function may explain the low expression of miR-148a and overexpression of its target HLA-G in ESCC tissues.