Immortalized normal liver cells (L02), human HCC cell lines (SK-HEP-1, HepG2, MHCC97H, MHCC97L, PLC, HuH7, SMMC-7721 and Bel-7402) and human embryonic kidney cells (HEK293T) were maintained in the recommended culture condition and incubated at 37ºC in a humidified environment containing 5% CO₂. Paired HCC and adjacent non-tumor liver tissues were collected from patients undergoing liver transplantation or partial hepatectomy at The First Affiliated Hospital, School of Medicine, Zhejiang University (Hangzhou, Zhejiang, China). Written informed consent was obtained from each patient. A total of 120 patients had a clear histologic diagnosis of HCC with complete clinicopathological data, and all patients were followed up for survival analysis. None of the patients received radiotherapy or chemotherapy before surgery. All sample data were obtained from clinical and pathologic records and are summarized in Table I.

miR-200a mimic and the control RNA duplex (referred to as NC) were used for the transient gain-of-function study. The small interfering RNA (siRNA) targeting the human CDK6 transcript was designated as siCDK6. The NC for both the miRNA mimic and siRNA was non-homologous to any human genome sequence. All the RNA oligoribonucleotides (Table II) were purchased from Genepharma (Shanghai, China). All oligoribonucleotides used in the present study are shown in Table II.

Total RNA from cell lines and clinical samples was isolated using the mirVana miRNA Isolation kit (Ambion). qRT-PCR was performed to detect the expression of miR-200a in various cell lines and clinical samples and the expression of CDK6 at the mRNA level. RNA was reverse transcribed using One Step PrimeScript miRNA cDNA Synthesis kit (Takara, Japan). The cDNA was then quantified by real-time RT-PCR using SYBR Premix Ex Taq (Takara). All PCR reactions were performed using the ABI7500 system (Applied Biosystems, Foster City, CA, USA). U6 or glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control. All primers used are listed in Table II.

Western blotting was performed to detect the gene expression at the protein level. Protein was extracted from transfected SK-HEP-1 cells using modified RIPA buffer in the presence of proteinase inhibitor cocktail. Equivalent quantities (30–50 μg) of protein were separated on 10% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% non-fat milk and then incubated overnight at 4ºC with the appropriate primary antibody at the dilutions specified by the manufacturer. The membranes were then washed three times in 10 ml TBST and incubated with the corresponding horseradish peroxidase (HRP)-conjugated secondary antibody at a dilution of 1:2,000 for 1 h. The bound secondary antibody was detected using an Enhanced chemiluminescence (ECL) system (Pierce Biotechnology, Inc., Rockford, IL, USA). Primary antibodies were as follows: anti-CDK6, anti-cyclin E2, anti-Rb, anti-phospho-Ser780-Rb, anti-cdc2, (Cell Signaling Technology), anti-phospho-Ser795-Rb and anti-β-actin (Epitomics).

All transfections were performed using Lipofectamine^® 2000 (Invitrogen) according to the manufacturer’s instructions. Briefly, SK-HEP-1, Huh7 or HEK293T cells were transfected with DNA, miRNA mimic, siRNA or NC. All RNA transfections were performed at a final concentration of 50 nM. Cells were collected for assay 48 h after transfection.

SK-HEP-1 and Huh7 cells seeded at a density of 4,000/well into 96-well plates were transfected with miR-200a mimic or NC. After incubating the cells for the specified time (1, 2, 3 or 4 days), a cell proliferation assay was performed using Cell Counting Kit-8 (CCK-8) (Dojindo) according to the manufacturer’s instructions. The solution absorbance was measured spectrophotometrically at 450 nm with the MRX II absorbance reader (Dynex Technologies, Chantilly, VA, USA). The experiments were performed in triplicate.

Aliquots of viable transfected SK-HEP-1 and Huh7 cells (1,000/well) were placed in 6-well plates 24 h after transfection and maintained in complete medium for 2 weeks. Colony formation was evaluated by staining the cells with 0.1% crystal violet. The rate of colony formation was calculated using the following equation: Colony formation rate = (number of colonies/number of seeded cells) × 100%. The experiments were performed in triplicate.

Forty-eight hours after transfection, SK-HEP-1 and Huh7 cells were harvested, washed with PBS and fixed in 70% ethanol at 4ºC overnight. Staining for DNA content was performed using a DNA Prep Stain (Beckman Coulter, Fullerton, CA, USA). Cell populations in the G0/G1, S and G2/M phases were determined using the BD™ LSRII flow cytometry system with FACSDiva software (both from BD Bioscience, Franklin Lakes, USA). Data were analyzed using ModFit LT software. The experiments were performed in triplicate.

To predict the target genes and the 3′ UTR binding sites bound by the seed region of miR-200a, the TargetScan (15) (http://www.targetscan.org/) and PicTar (16) (http://pictar.mdc-berlin.de/) databases were used. CDK6 was chosen as the target candidate according to both TargetScan and PicTar databases.

The 3′ UTR of CDK6 was amplified by PCR. The amplified sequence was subcloned and ligated into the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Madison, WI, USA). The recombinant reporter vector was identified by sequencing and termed the wild-type (Wt). To create the miR-200a binding site mutants, the binding region of the seed sequence (5′-AGTGTT-3′) was mutated to the sequence 5′-TCACAA-3′, using the QuikChange Lightning Site-Directed Mutagenesis kit (Stratagene) according to the manufacturer’s protocol. The recombinant vector was confirmed by sequencing and termed the mutant type (Mut). 293T cells plated in a 24-well plate were co-transfected with 50 nM of either the miR-200a mimic or NC and 100 ng of pmirGLO vector comprising Wt or Mut 3′ UTR of CDK6 using Lipofectamine^® 2000. Forty-eight hours after transfection, the relative luciferase activity was measured by the Dual-Luciferase Reporter Assay System (Promega). All transfection experiments were performed in triplicate.

Data are presented as the means ± SD. Comparisons were made by using the Student’s t-test, the χ² test and the log-rank test. Overall survival rates were calculated actuarially according to the Kaplan-Meier method. Variables with a value of P<0.05 in univariate analysis were used in a subsequent multivariate analysis based on the Cox proportional hazards model. A value of P<0.05 was considered to indicate a statistically significant result.

To further confirm the expression pattern of miR-200a in HCC, we detected the expression of miR-200a in 120 paired HCC and adjacent non-cancerous liver tissues by real-time qRT-PCR. Compared with the adjacent non-tumor tissues, significant downregulation of miR-200a was observed in the HCC tissues (Fig. 1A). Furthermore, the expression of miR-200a was noticeably reduced in 6 of 8 (75.0%) HCC cell lines examined when compared with the expression in the normal liver cell line (L02) (Fig. 1B). These results suggest that reduced miR-200a expression is a frequent event in human HCC and may be involved in tumorigenesis.

In the present study, patients with expression values less than the average level of miR-200a in HCC tissues (1.87, normalized to U6) were assigned to the low expression group (n=64) whereas patients with values above the average were assigned to the high expression group (n=56). Our data showed a significant association between miR-200a expression and the AFP level, histologic grade, TNM stage and PVTT (Table I). Furthermore, the Kaplan-Meier method revealed that a lower miR-200a level was associated with shorter overall survival (P<0.01) (Fig. 2). Multivariate analysis further confirmed that a reduced miR-200a level was an independent and significant prognostic factor for survival (HR, 2.621; CI, 1.116–4.642; P=0.004; Table III). Collectively, these data suggest that deregulation of miR-200a may contribute to the development and progression of HCC.

The frequent downregulation of miR-200a in HCC cell lines and tissues suggests that miR-200a may serve as a potential TSG in HCC. To test the potential tumor-suppressor role of miR-200a in HCC, the effect of ectopic miR-200a on cell growth was evaluated using the CCK-8 cell viability assay. The expression of miR-200a was markedly increased in the SK-HEP-1 and Huh7 cells transfected with 50 nM miR-200a mimic when compared with NC, normalized to U6 (Fig. 3A). Our data revealed that miR-200a overexpression significantly inhibited cell proliferation in SK-HEP-1 and Huh7 cells compared to NC. At day 1, the OD values of the miR-200a mimic and NC control cells were not significantly different. However, from day 3 onwards, the OD values of the miR-200a-treated SK-HEP-1 and Huh7 cells were significantly lower than the value of the control cells (Fig. 3B). These data indicated that miR-200a inhibited cell proliferation in vitro.

To further investigate the potential role of miR-200a in tumorigenesis, a colony formation assay was performed in SK-HEP-1 and Huh7 cells. Cells were transfected with the miR-200a mimic or NC duplex, and then allowed to grow to a very low density. Notably, SK-HEP-1 and Huh7 cells transfected with 50 nM miR-200a mimic displayed much fewer and smaller colonies compared with the NC duplex-transfected cells (Fig. 3C), suggesting a growth-inhibitory role of miR-200a.

To explore the mechanisms underlying the suppression of tumor growth by miR-200a, the effect of miR-200a on cell cycle progression was investigated by flow cytometry. The assay showed that overexpression of miR-200a resulted in an accumulation of a G1 population in the SK-HEP-1 (Fig. 4A) and Huh7 cells (Fig. 4B). These results indicated that miR-200a inhibited HCC cell proliferation by inducing cell cycle arrest at the G1 phase.

It is generally believed that miRNAs exert their function through negatively regulating the expression of their downstream target genes. To further elucidate the molecular mechanism by which miR-200a inhibits G1/S transition, CDK6 was predicted as a potential target of miR-200a by TargetScan and PicTar. A single putative miR-200a-binding site was mapped in the 3′ UTR of CDK6 (Fig. 5A). To validate whether CDK6 is a direct target of miR-200a, a human CDK6 3′ UTR fragment containing Wt or the mutant (Mut) miR-200a binding sequence (Fig. 5A) was cloned downstream of the firefly luciferase reporter gene in pmirGLO. In HEK293 cells cotransfected with the reporter plasmids and the miR-200a mimic or NC duplex, the luciferase activity of the reporter that contained Wt 3′ UTR was significantly suppressed by miR-200a mimic, but the luciferase activity of the mutant reporter was unaffected (Fig. 5B), indicating that miR-200a may suppress gene expression through miR-200a binding sequence at the 3′ UTR of CDK6. Additionally, transfection of the miR-200a mimic decreased CDK6 expression in SK-HEP-1 cells at both the protein and mRNA levels (Fig. 5C), suggesting that CDK6 expression is inhibited by miR-200a at the post-transcriptional level. Collectively, the results showed that miR-200a regulated the expression of endogenous CDK6 by directly targeting the 3′ UTR of CDK6 mRNA and that CDK6 is a novel target of miR-200a.

To explore the role of CDK6 in the miR-200a-regulated G1/S transition, we investigated whether knockdown of CDK6 may phenocopy the effect of miR-200a overexpression. SK-HEP-1 cells were transfected with siRNA duplex targeting CDK6, which resulted in a significant reduction in the protein levels (Fig. 5D). The silencing of CDK6 led to G1-phase arrest (Fig. 4C), phenocopying the outcome of miR-200a overexpression.

CDK6 is one of the crucial molecules that initiate the phosphorylation of retinoblastoma (Rb), which results in the release of E2F and subsequently the transactivation of genes required for S-phase entry. Therefore, we investigated whether miR-200a could attenuate these events. Ectopic expression of miR-200a caused a reduction in phosphorylated Rb protein, however little effect on the total Rb protein was noted (Fig. 6). Moreover, the effect of miR-200a on endogenous genes such as cdc2 and cyclin E2, downstream genes of E2F3, was further investigated. Overexpression of miR-200a induced significant downregulation of cdc2 and cyclin E2 in SK-HEP-1 cells (Fig. 6).

These data indicate that miR-200a may induce cell cycle arrest at G1 phase by directly inhibiting CDK6 through the Rb-E2F signaling pathway.