The current study was approved by the Ethics Committee of Central Hospital of Linyi (Linyi, China). A total of 89 primary HCC tissues and adjacent non-tumor tissues were collected from patients at the Central Hospital of Linyi between September 2010 and June 2012, and informed consent was obtained from all participants. No patients received radiation therapy or chemotherapy prior to surgical resection. Tissues were immediately snap-frozen in liquid nitrogen following surgical resection, and stored in liquid nitrogen prior to use. The clinical information of patients involved in this study is summarized in Table I.

Human normal liver epithelial cell line THLE-3 and four liver cancer cell lines including HepG2, Hep3B HCCLM3 and SMCC7721 were purchased from Cell Bank of Chinese Academy of Sciences (Shanghai, China). Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Inc.) in a 37°C humidified atmosphere with 5% CO₂. For cell transfection, HepG2 cells were transfected with scramble miR mimic (miR-NC) (CmiR0001-MR04), miR-137 mimic (HmiR0011-MR04), negative control (NC) inhibitor (CmiR-AN0001-SN), miR-137 inhibitor (HmiR-AN0175-SN-10) (all purchased from Guangzhou Fulengen Co., Ltd., Guangzhou, China) or co-transfected with miR-137 mimic and pcDNA3.1-enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) expression plasmid (Yearthbio, Changsha, China) using Lipofectamine 2000 (Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions.

Total RNA was extracted from 1×10⁷ cells (THLE-3 cells and the liver cancer cell lines) using TRIzol Reagent (Thermo Fisher Scientific, Inc.). Total RNA was then converted into cDNA using PrimeScript First Strand cDNA Synthesis kit (Takara Bio, Inc., Otsu, Japan). For miR expression, RT-qPCR was performed using miRNA Q-PCR Detection kit (GeneCopoeia, Inc., Rockville, MD, USA) according to the manufacturer's instructions on an ABI 7500 thermocycler (Thermo Fisher Scientific, Inc.). U6 was used as a reference gene internal control for miR expression. U6 (HmiRQP9001) and miR-137 (HmiRQP0175) primers were purchased from Guangzhou Fulengen Co., Ltd. For mRNA expression detection, RT-qPCR was performed using SYBR-Green I Real-Time PCR kit (Biomics Biotechnologies Co., Ltd., Nantong, China). The primer sequences for EZH2 were as follows: Forward, 5′-AATCAGAGTACATGCGACTGAGA-3′ and reverse, 5′-GCTGTATCCTTCGCTGTTTCC-3′. The primer sequences for GAPDH were as follows: Forward, 5′-CTGGGCTACACTGAGCACC-3′ and reverse, 5′-AAGTGGTCGTTGAGGGCAATG-3′. The PCR reaction system was 0.33 µl cDNA solution, 10 µl SYBR-Green I mix, 2 µl primers and 7.67 µl H₂O were mixed to obtain a final reaction volume of 20 µl. The reaction condition were 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 15 sec and annealing/elongation step at 60°C for 30 sec. The relative expression was analyzed by the 2^−ΔΔCq method (16).

Cells (1×10⁷) were lysed with ice-cold radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology, Haimen, China). Proteins were collected, quantified using a bicinchoninic acid assay and separated (50 µg/lane) by 10% SDS-PAGE (Beyotime Institute of Biotechnology). Proteins were transferred onto polyvinylidene difluoride (PVDF) membrane (Thermo Fisher Scientific, Inc.). The PVDF membrane was incubated with phosphate buffered saline (PBS; Thermo Fisher Scientific, Inc.) containing 5% non-fat milk overnight at 4°C, and then incubated with the polyclonal mouse anti-human EZH2 antibody (1:100; cat. no. ab168764; Abcam, Cambridge, MA, USA) and polyclonal mouse anti-human GAPDH antibody (1:50; cat. no. ab8245; Abcam) at room temperature for 3 h. After washing with PBS-Tween three times, the membrane was incubated with goat polyclonal anti-mouse HRP-conjugated secondary antibody (1:5,000; cat. no. ab97040; Abcam) at room temperature for 1 h. The enhanced chemiluminescence system (Thermo Fisher Scientific, Inc.) was used to detect the immunoreactive bands. The protein expression was measured using Image Pro Plus software (Media Cybernetics, Inc., Rockville, MD, USA). GAPDH was used as the internal reference.

HepG2 cell suspension (5×10⁴ cells/well) was plated in a 96-well plate, and cultured for 12, 24, 48 or 96 h. Subsequently, MTT (10 µl, 5 mg/ml) was added, and cells were then incubated at 37°C for 4 h. The supernatant was removed, and 100 µl dimethyl sulfoxide was added into each well. The absorbance at 570 nm was determined using the Model 680 Microplate Reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

HepG2 cells were cultured to confluence, and a wound was created with a plastic scriber. Then, cells were washed and incubated in serum-free DMEM at 37°C for 24 h. Subsequently, the medium was replaced with DMEM with 10% FBS, and then cultured at 37°C for 24 h. Cells were observed under a Nikon microscope (Nikon Corporation, Tokyo, Japan).

A 24-well Transwell chamber (Chemicon; Merck KGaA, Darmstadt, Germany) with a layer of Matrigel was used to perform Transwell assays. A HepG2 cell suspension (5×10⁴ cells) was added in the upper chamber, and DMEM containing 10% FBS was added into the lower chamber. After incubation for 24 h, non-invading cells on the interior of the inserts were removed using a cotton-tipped swab. Cells on the lower surface of the membrane was stained with 0.5% gentian violet at room temperature for 10 min, and then rinsed by water, and air-dried. Invading cells were counted under a Nikon microscope.

Targetscan 7.0 (www.targetscan.org) was used to predicate the putative targets of miR-137, according to the manufacturer's instructions. According to the manufacturer's instructions, the QuickChange Site-Directed Mutagenesis kit (Stratagene; Agilent Technologies, Inc., Santa Clara, CA, USA) was used to construct the mutant type (MT) EZH2 3′ untranslated region (3′UTR) lacking complementarity with the miR-137 seed sequence. The wild-type (WT) or MT of EZH2 3′UTR was cloned into the downstream of the firefly luciferase-coding region of pMIR-GLO™ Luciferase vector (Promega Corporation, Madison, WI, USA). The cloning procedure was performed by Yearthbio. HepG2 cells were co-transfected using Lipofectamine 2000 with the WT- or MT-EZH2-3′UTR luciferase reporter plasmid, and miR-NC or miR-137 mimic, respectively. The luciferase activity was detected after transfection for 48 h using the Dual Luciferase Reporter Assay system (Promega Corporation), according to the manufacturer's instruction.

Data are expressed as the mean ± standard deviation of three independent experiments. SPSS 19.0 (IBM Corp., Armonk, NY, USA) was used to perform statistical analysis. The association between miR-137 expression and clinical characteristics in HCC were analyzed using χ² test. The Kaplan-Meier estimator was used for survival analysis. The difference between two groups was analyzed using Student's t-test. The difference among more than two groups was analyzed using analysis of variance and Tukey post hoc test. Pearson correlation analysis was used to analyze the correlation between miR-137 and EZH2 mRNA levels in HCC tissues. P<0.05 was considered to indicate a statistically significant difference.

To study the regulatory mechanism of miR-137 in HCC, its expression levels were initially examined in HCC tissues and adjacent non-tumor tissues. RT-qPCR data demonstrated that miR-137 was significantly downregulated in HCC tissues compared with adjacent non-tumor tissues (Fig. 1A). Additionally, the expression of miR-137 was further lower in HCC at high stage (III–IV) compared with that in HCC at low stage (I–II; Fig. 1B). These findings suggest that the decreased expression of miR-137 may contribute to the malignant progression of HCC. To further confirm these findings, the association between the miR-137 expression and clinical characteristics in HCC was examined. Using the mean value of miR-137 expression as a cutoff, the HCC patients were divided into a high miR-137 expression group and low miR-137 expression group. Low expression of miR-137 was significantly associated with vein invasion, lymph node metastasis and advanced clinical stage in HCC, but not associated with age, sex, tumor number, histologic grade, or HPV infection (Table I). Taken these findings together, we suggest that downregulation of miR-137 is involved in HCC progression. Subsequently, the association between miR-137 expression and overall survival time of patients with HCC was analyzed. As presented in Fig. 1C, the patients with HCC with low expression of miR-137 had shorter survival time compared with those with high miR-137 levels. These findings suggest that downregulation of miR-137 is associated with poor prognosis in HCC.

To further investigate the underlying mechanism, in vitro experiments using liver cancer cell lines were performed. RT-qPCR data showed that the miR-137 levels were significantly reduced in HepG2, Hep3B, LM3 and SMCC7721 cell lines compared with THLE-3 normal liver epithelial cells (Fig. 2A). HepG2 cells were then transfected with miR-137 mimic or miR-NC as the control group. Following transfection, the miR-137 levels were examined using RT-qPCR. As presented in Fig. 2B, the miR-137 levels were significantly increased in HepG2 cells transfected with miR-137 mimic compared with the miR-NC group. Further studies demonstrated that overexpression of miR-137 caused a significant reduction in cell proliferation, migration and invasion compared with the miR-NC group (Fig. 2C-E). These data suggest that miR-137 has a suppressive role in HCC growth and metastasis.

Targetscan software was used to predict that EZH2 was a putative target of miR-137 (Fig. 3A). To confirm their targeting relationship, the WT-EZH2-3′UTR or MT-EZH2-3′UTR luciferase reporter plasmids were generated (Fig. 3B). HepG2 cells were then co-transfected with WT- or MT-EZH2-3′UTR luciferase reporter plasmid, and miR-NC or miR-137 mimic. As demonstrated in Fig. 3C, the luciferase activity was significantly reduced in cells co-transfected with miR-137 mimic and WT-EZH2-3′UTR luciferase reporter plasmid when compared with the control group; the same effect was not observed in cells co-transfected with miR-137 mimic and MT-EZH2-3′UTR. Accordingly, these results indicated that miR-137 is a direct target gene of miR-137 in HepG2 cells.

Effects of miR-137 on the protein expression of EZH2 were also determined. The protein levels of EZH2 were significantly reduced in the miR-137 group compared to miR-NC group, indicating that overexpression of miR-137 downregulates EZH2 expression (Fig. 4A). To further confirm these findings, HepG2 cells were transfected with miR-137 inhibitor, or NC inhibitor as the control group. RT-qPCR data demonstrated that the miR-137 levels were downregulated in miR-137 inhibitor group compared to NC inhibitor group (Fig. 4B). Additionally, the miR-137 inhibitor significantly increased the protein expression of EZH2 in HepG2 cells (Fig. 4C). Accordingly, these results indicated that miR-137 negatively regulates the protein expression of the target gene, EZH2, in HepG2 cells.

To further confirm the targeting relationship between miR-137 and EZH2, the expression levels of EZH2 in HCC tissues were examined. As presented in Fig. 5A, RT-qPCR data demonstrated that the mRNA levels of EZH2 were significantly upregulated in HCC tissues compared to adjacent non-tumor tissues. Furthermore, its expression levels were higher in HCC at high stage (III–IV) when compared with those in HCC at low stage (I–II; Fig. 5B). Further investigation demonstrated an inverse correlation between the miR-137 and EZH2 mRNA levels in HCC tissues (Fig. 5C). Therefore, these data further support that downregulation of miR-137 causes the upregulation of its target gene EZH2 in HCC. In addition, the mRNA and protein levels of EZH2 were also upregulated in the liver cell lines compared with THLE-3 cells (Fig. 5D-E).

Subsequently, the role of EZH2 in the miR-137-mediated malignant phenotypes of HCC cells was investigated. HepG2 cells were transfected with miR-137 mimic, or co-transfected with miR-137 mimic and pcDNA3.1-EZH2 ORF plasmid. RT-qPCR and western blot data indicated that the mRNA and protein levels of EZH2 were significantly upregulated in cells co-transfected with miR-137 mimic and pcDNA3.1-EZH2 ORF plasmid, when compared with those in the miR-137 group (Fig. 6A and B). Cell proliferation, migration and invasion were also examined. As indicated in Fig. 6C-E, the proliferation, invasion and migration of HepG2 cells were significantly increased in the miR-137 + EZH2 group compared with the miR-137 group. Therefore, overexpression of EZH2 may reverse some of the inhibitory effects of miR-137 on the malignant phenotypes of HepG2 cells. These findings suggested that miR-137 inhibits cell proliferation, migration and invasion, at least partially, by directly targeting EZH2.