Forty pairs of primary HCC and adjacent non-tumor tissues were obtained from patients undergoing surgery at the Department of Hepatopancreatobiliary Surgery, The First Hospital, Jilin University (Changchun, China). The collection and use of patient samples was approved by the Ethics Committee of the First Hospital of Jilin University, and written informed consent was obtained from all patients whose biological samples were used in the study. All samples were immediately snap-frozen in liquid nitrogen and stored at −80°C until use.

The 4 HCC cell lines (SMMC-7721, Hep3B, HepG2 and Huh-7) as well as the normal human hepatocyte cell line HL-7702, were purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). All cell lines were maintained in high glucose Dulbeccos modified Eagles medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (both from Gibco, Gaithersburg, MD, USA) at 37°C in a humidified chamber supplemented with 5% CO₂.

The miR-363 mimic and the appropriate negative control oligonucleotide (miR-NC) were purchased from RiboBio (Guangzhou, China). The E2F3 overexpression vector was amplified and inserted into the pcDNA3.0 vector (Invitrogen Life Technologies, Grand Island, NY, USA) at the BamHI/EcoRI restriction sites. Transfection was performed using Lipofectamine 2000 (Invitrogen) in accordance with the manufacturer's protocol.

Total RNA containing miRNA and mRNA was extracted from tissue samples and cells using TRIzol^® reagent (Invitrogen) according to the manufacturer's instructions. RNA was reverse transcribed to cDNA using Prime Script First Strand cDNA Synthesis kit (Takara, Dalian, China) following the manufacturer's protocol. The cDNAs were subjected to qRT-PCR using SYBR Premix Ex Taq (Takara) under ABI 7900 Fast System (Applied Biosystems, Foster City, CA, USA) to detect miR-363 and E2F3 mRNA. miR-363 and U6 primers were purchased from RiboBio. The primers for E2F3 and GAPDH used in the present study were as previously described (18). U6 and GAPDH were used as internal controls to detect miR-363 and E2F3 mRNA, respectively. Relative expression of miR-363 and E2F5 was determined using the 2^−ΔΔCt method.

Cell proliferation was assessed using Cell Counting Kit-8 (CCK-8) assay (Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer's instructions. In brief, 5×10³ transfected cells/well were seeded in 96-well plates and cultured for 24–72 h. At the indicated time, an amount of 20 µl of CCK-8 solution was added to each well and incubation was carried out for 4 h. Absorbance at 450 nm was measured using an enzyme-linked immunosorbent assay reader (Thermo Labsystems, Helsinki, Finland).

Colony formation ability was determined using plating at 500 transfected cells/well into 6-well plates and culturing for 14 days. Cells were fixed with methanol and stained with 0.1% crystal violet for 15 min. Stained colonies were imaged and counted by a light microscope (Olympus, Tokyo, Japan).

Cell migration was measured using a wound healing assay. In brief, transfected cells were cultured in 6-well plates (5×10⁴ cells/well) and grown to confluency. Subsequently, an artificial homogenous wound was scratched into the monolayer using a sterile plastic micropipette tip, and then the cells were cultured under standard conditions for 24 h. Following several washes, the cultured cells were fixed and observed using an inverted microscope (Olympus).

The invasion assay was performed using Transwell insert chambers (Corning Inc., Corning, NY, USA). For the Transwell invasion assay, 1×10⁵ transfected cells were seeded into each well of the upper chamber of Matrigel-coated inserts in serum-free medium. In the lower chamber, DMEM with 20% FBS was added to serve as a chemoattractant. After incubating for 24 h at 37°C with 5% CO₂, the non-invading cells were gently removed with a cotton swab, whereas cells that had invaded to the lower surface of the filter were fixed in 70% ethanol for 30 min and stained with 2% crystal violet for 10 min. Invaded cells were photographed and quantified by counting them randomly in 5 fields by a light microscope (Olympus).

The 3′UTR of E2F3 containing the potential binding sites of miR-363 (position 272–278) was amplified by PCR using human liver cDNA, and inserting into the psiCHECK2 vector (Promega, Wisconsin, WI, USA) within the XhoI/NotI restriction sites. A mutant 3′UTR of E2F3 was constructed using QuikChange XL Site-Directed Mutagenesis kit (Agilent Technologies, Santa Clara, CA, USA), and cloned into the psiCHECK2 vector. For the luciferase reporter assay, HepG2 cells were seeded in a 96-well plate. When reaching 70% confluency, the cells were cotransfected with miR-363 mimic/miR-NC and wild/mutant-type 3′UTR of E2F3. Luciferase activities were determined 48 h after transfection using the Dual-Luciferase Reporter Assay System (Promega).

Total protein extraction, SDS-PAGE and western blot analyses were performed as previously described (19). The primary antibodies against E2F3, GAPDH, E-cadherin, N-cadherin and vimentin were purchased from Cell Signaling Technology Inc. (Danvers, MA, USA). GAPDH was used as an internal control for all experiments.

For in vivo tumorigenesis assays, 2×10⁶ HepG2 cells carrying either miR-363 or miR-NC were subcutaneously injected into the flanks of female BALB/c nude mice (6–7 weeks of age) (Laboratory Animal Center of Jilin University). Tumor volume was determined by measuring tumor length (L) and width (W) every 7 days using the formula: Tumor volume = 0.5 × (L × W²). At 35 days after injection, animals were sacrificed and subcutaneous tumors were stripped and weighed. All animal handling and research protocols were approved by the Animal Care and Use Ethics Committee of Jilin University.

In all experiments, the data are expressed as the mean ± standard error of the mean (SEM) from at least 3 independent experiments. The differences between groups were analyzed using Student's t-test when only 2 groups were compared, or using a one-way ANOVA when >2 groups were compared. Correlations between miR-363 expression with E2F3 were evaluated by the Pearsons correlation analysis. A value of P<0.05 was considered as an indication of statistical significance. All analyses were performed using SPSS 19.0 software (SPSS, Inc., Chicago, IL, USA).

We examined the expression of miR-363 in 40 paired HCC and corresponding adjacent non-tumor tissues by qRT-PCR. We found that miR-363 was significantly decreased in the HCC tissues (Fig. 1A). To assess the association of miR-363 expression with HCC prognosis, patients were divided into 2 groups using the mean miR-363 expression (0.33) as cut-off: one with relative high miR-363 expressions (>0.33), and another with relative low miR-363 expressions (<0.33). We found that miR-363 expression was significantly associated with TNM stage, tumor differentiation and lymph node metastasis (Table I). We next examined the expression of miR-363 in a series of HCC cell lines. We found that the expression of miR-363 in 4 human HCC cell lines (SMMC-7721, Hep3B, HepG2 and Huh-7) was significantly downregulated compared with that noted in the normal human hepatocyte cell line, HL-7702 (Fig. 1B). These data suggest that miR-363 is involved in the carcinogenesis of HCC.

To assess the functional significance of miR-363 in HCC growth, HepG2 cells with low expression of miR-363 were transfected with the miR-363 mimic (miR-363) or miR-NC, and then cell proliferation and colony formation ability were determined. qRT-PCR results confirmed that miR-363 was upregulated in the HepG2 cells transfected with the miR-363 mimic as compared to cells transfected with miR-NC (Fig. 2A). CCK-8 assay revealed that overexpession of miR-363 significantly suppressed proliferation of the HepG2 cells (Fig. 2B). Consistent with this result, overexpression of miR-363 also significantly suppressed colony formation of the HepG2 cells (Fig. 2C). To further confirm whether miR-363 suppresses tumor growth in vivo, we created tumor xenograft mouse models. Consistently, miR-363 significantly inhibited tumor growth in vivo (Fig. 2D). After 35 days of injection, the mice were sacrificed, and tumor tissues were stripped. We found that the size of the subcutaneous tumors derived from the miR-363-expressing HepG2 cells were markedly smaller than the size of the tumors derived from the miR-NC-transfected cells (Fig. 2E). Taken together, our data support a growth inhibitory activity of miR-363 in HCC in vitro and in vivo.

We next studied the effects of miR-363 on cell migration and invasion in HCC cells using wound healing and Transwell invasion assays, respectively. Notably, we found that miR-363 overexpression significantly inhibited the migration and invasion of HepG2 cells compared to these capacities in the miR-NC group (Fig. 3A and B). Furthermore, epithelial-mesenchymal transition (EMT) has been reported to play a key role in the metastasis of HCC cells (18). To determine whether miR-363 affects molecular changes typical of EMT in HCC cells, the expression of mesenchymal markers, including N-cadherin and vimentin and the epithelial marker, E-cadherin, was determined in the HepG2 cells. As shown in Fig. 3C, overexpression of miR-363 resulted in increased E-cadherin expression, and decreased N-cadherin and vimentin expression. These finding revealed that miR-363 inhibits HCC metastasis by regulating EMT.

To understand the mechanisms by which miR-363 suppresses HCC growth and invasion, we searched for candidate targets of miR-363 using 3 bioinformatic databases (TargetScan, miRanda and PicTar). We identified the 3′UTR of E2F3 that was able to bind to the ‘seed region’ of miR-363 (Fig. 4A). To confirm that miR-363 may bind to the 3′UTR of E2F3, a human E2F3 3′UTR fragment containing the binding sites of miR-363 (Fig. 4A), or the mutant sites were amplified and inserted into the psiCHECK2 vector, and miR-363 mimic or miR-NC were co-transfected into HepG2 cells and cultured for 48 h, and then luciferase activities were measured. As predicted, miR-363 was bound to E2F3 3′UTR, resulting in markedly reduced luciferase activities (Fig. 4B). This effect was specific as miR-363 failed to suppress the luciferase activity when the binding site within E2F3 3′UTR was mutated (Fig. 4B). The results indicated that E2F3 is a downstream target of miR-363. Moreover, qRT-PCR and western blot analysis further demonstrated that overexpression of miR-363 markedly suppressed E2F3 expression at the mRNA and protein levels (Fig. 4C and D) in HepG2 cells. In addition, the E2F3 protein level in tumor tissues isolated from nude mice was determined by western blotting. The expression level of E2F3 protein in the tumor tissues with miR-363 overexpression was downregulated compared with the miR-NC group (Fig. 4D). Taken together, these data indicate that E2F3 is a direct target of miR-363 in HCC cells.

We further studied whether E2F3 is involved in the function of miR-363 in HCC cells. HepG2 cells with high expression of miR-363 were transfected with the E2F3 overexpression plasmid, and the expression of E2F3 was examined by qRT-PCR and western blotting. As shown in Fig. 5A and B, E2F3 expression at the mRNA and protein levels was restored in the HepG2 cells. Then, we carried out CCK-8, colony formation, would healing and Transwell assays to evaluate the effect of E2F3 overexpression on cell proliferation, colony formation, migration and invasion in the HepG2 cells transfected with miR-363 with/without the E2F3 overexpression plasmid. As expected, E2F3 overexpression reversed the suppressive effect on cell proliferation, colony formation, migration and invasion in HepG2 cells induced by miR-363 overexpression (Fig. 5C-F). Therefore, our data clearly demonstrated that miR-363 inhibits HCC cell growth and invasion, at least in part, by targeting E2F3.

We finally examined the E2F3 mRNA expression in 40 pairs of HCC tumors and the corresponding adjacent non-tumor tissues by qRT-PCR. As shown in Fig. 6A, E2F5 mRNA expression was significantly increased in the HCC tissues compared to that noted in the adjacent non-tumor tissues. Pearsons correlation analysis revealed that the expression of miR-363 was inversely correlated with the E2F3 mRNA level in the HCC tissues (r=−0.648; P<0.001; Fig. 6B).