miR-214-5p targets KLF5 and suppresses proliferation of human hepatocellular carcinoma cells

2.1 HCC tissue collection

This study was approved by the Ethics Review Committee of The Affiliated Hospital of Qingdao University (West Coast District) (Qingdao, China), and written consent form was obtained from all the patients. A total of 78 patients who underwent surgery at Qingdao West Coast New Area Central Hospital between March 2010 and January 2012 were recruited. All patients enrolled did not receive any anticancer treatment before surgery. The tumor tissues and matched normal tissues were immediately frozen in liquid nitrogen after surgical resection and preserved at −80°C until further use. Clinical stage for HCC was evaluated based on the seventh edition of the UICC/AJCC TNM Staging System. The clinicopathological features were collected and summarized in Table 1.

2.2 Follow-up survey

After surgery, all patients were followed up and the overall survival time (OS) and disease-free survival time were calculated. OS was defined as the time between the date of surgery and the patient's death from HCC. Disease-free survival was defined as the time between the date of surgery and recurrence.

2.3 Cell culture and transfection

The HCC cell lines (Hep3B and Huh7) and the normal liver cell line (HL-7702) were obtained from Shanghai Cell Bank of Chinese Academy of Sciences (Shanghai, China). All cell lines were maintained in Dulbecco modified Eagle medium (DMEM, Thermo Fisher Scientific, Inc., Waltham, MA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc.) in a humidified incubator at 37°C containing 5% of CO₂.

Synthetic mimic (5′-UGCCUGUCUACACUUGCUGUGC-3′) or inhibitor (5′-GCACAGCAAGUGUAGACAGGCA-3′) of miR-214-5p, and negative control (NC) miRNA (5′-GUGAGAAGUACCACCGAGACAG-3′) were purchased from Ribobio (Guangzhou, China). Synthetic small interfering RNA (siRNA) targeting KLF5 (5′-GCUCCAGAGGUGAACAAUA-3′) and NC siRNA (5′-AGCGACUCAAGAUUACGAG-3′) were also purchased from Ribobio (Guangzhou). For transfection, the cells were cultured in a 6-well plate at a density of 3 × 10⁵ cells/well, and growth to 70% to 80% confluence. Lipofectamine 2000 (Thermo Fisher Scientific, Inc.) was used for synthetic siRNAs or miRNAs transfection according to the manufacture's recommendations. Briefly, the miRNAs or siRNAs and Lipofectamine 2000 were added into Eppendorf tubes containing 50 µL DMEM. Following, the mixture was added to well after incubating for 20 minutes at room temperature. Following transfection for 48 hours, the cells were collected to detect the expressions of miR-214-5p or KLF5. Transfection efficiency was confirmed by quantitative real-time polymerase chain reaction (RT-qPCR) or Western blot.

2.4 Cell proliferation assay

The cell proliferation rate was assessed by a Cell Counting Kit 8 assay (CCK-8, Beyotime, Haimen, Jiangsu, China) following the supplier's instructions. Briefly, the cells were plated in a 96-well plate at a density of 2 × 10³ cells/well. Following, 10 μL CCK-8 solution was added to each well at indicated time points (0 hours, 48 hours, and 72 hours) and incubated for another 4 hours at 37°C. The optical absorbance of each well was measured at 450 nm using ELISA plate reader (Bio-tek, Winooski, VT). The experiments were repeated in triplicate.

2.5 Cell cycle analysis

For cell cycle analysis, the cells were treated with trypsin and fixed using 80% ethanol at 4°C for overnight. The cells were then washed with cold phosphate-buffered saline (PBS) and stained with propidium iodide/RNase A mixture (Beyotime) at darkness for 30 minutes. Following, cell cycle analysis was conducted using FACSCalibur™ flow cytometer (BD Biosciences, Franklin Lakes, NJ). A total of 20 000 cells were acquired and analyzed for DNA content. The results were analyzed by ModFit 3.2 software (Verity Software House Company, Topsham, ME). Histograms represent the percentage of cells in each phase of the cell cycle (G0/G1, S, and G2/M). The experiments were repeated in triplicate.

2.6 Cell migration analysis

Wound-healing assay was used to evaluate the migratory ability of the cultured cells. Scratch was created using a sterile micropipette tip in the cell surface. The detached cells were removed by washing the cell monolayer using PBS. The initial (0 hours) and the residual gap length of 24 hours after wounding were calculated from photomicrographs. The experiments were repeated in triplicate.

2.7 RNA extraction and RT-qPCR

Total RNA was extracted from all the fresh tissues with an RNAeasy tissue MiniKit (Qiagen Inc., Nordrhein-Westfalen, Germany) according to the manufacture's recommendations. The concentration of extracted RNA was measured using Nanodrop-ND-1000 (Thermo Fisher Scientific Inc.). For miRNA expression, reverse transcription was performed using a One Step PrimeScript RT Reagent Kit (Takara Bio., Inc., Dalian, China), followed by RT-qPCR with Takara SYBR Master Mix (Takara Bio., Inc.) at an ABI 7200 Real-Time PCR System (Applied Biosystems, Life Technologies GmbH, Darmstadt, Germany). PCR conditions included an initial holding period at 95°C for 10 minutes and for 1 cycle, then 95°C for 15 seconds and 60°C for 30 seconds for 40 cycles. The primer sequences used in this study were as follows: miR-214-5p forward, 5′-ACACTCCAGCTGGGTGCCTGTCTACACTTG-3′, and miR-214-5p reverse, 5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGGCACAGCA-3′; U6 snRNA forward, 5′-GCGCGTCGTGAAGCGTTC-3′, and U6 snRNA reverse, 5′-GTGCAGGGTCCGAGGT-3′. Differences in gene expression were calculated using the 2^−ΔΔCq method18 and expressed as a fold-change. Each experiment was repeated in triplicate.

2.8 Protein extraction and Western blot

All the fresh tissues or cell lines were lysed with lysis buffer (150 mm NaCl, 50 mm Tris-HCl, 1% NP-40, pH 7.5; Beyotime Biotechnology). Protein samples were quantified using a Bradford Protein Concentration Determination Kit (Beyotime Biotechnology) according to the supplier's instructions. Protein samples were separated at a 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis and subsequently transfected onto nitrocellulose membranes (CWBiotech, Beijing, China). Following blocked with 10% non-fat milk in PBS, membranes were incubated with primary antibodies at 4°C overnight (anti-KLF5: ab24331, 1:2000, Abcam, Cambridge, MA or anti-GAPDH: ab181602, 1:1000, Abcam) independently. Subsequently, the membranes were washed with TBST for three times and then incubated with secondary antibody (sc-2491, 1:1500, Santa Cruz Biotechnology, Inc., Dallas, TX) for 3 hours at 25°C. The signals were detected using a BeyoECL Plus Kit (Beyotime). Protein expression levels were normalized to total GAPDH expression.

2.9 Immunohistochemistry (IHC)

HCC tumor tissues and matched normal tissues were previously formalin-fixed and paraffin-embedded were sliced into 4 μm sections and underwent deparaffination and then rehydration. Antigen retrieval, suppression of endogenous peroxidase activity, and 10% skim milk blocking were performed before primary antibody incubation. An anti-KLF5 (ab24331, Abcam) antibody was used as a primary antibody and incubated with the sections overnight at 4°C. The section was subsequently incubated with peroxidase conjugated secondary antibody (ab6721, Abcam) for 90 minutes, and DAB solution was used for signal development for 5 minutes. The sections were counterstained with hematoxylin followed by dehydrating and mounting.

2.10 Luciferase reporter assay

Both the wild-type (WT) and mutant (Mut) 3′-UTR of KLF5 containing the putative binding sequence of miR-214-5p were inserted into the pmiRGLO dual-luciferase expression vector (Promega, Madison, WI). The renilla luciferase expression vector was used as an internal control. The cells were transfected with miR-214-5p mimic or NC miRNA and luciferase reporter constructs (WT or Mut) using lipofectamine 2000 (Thermo Fisher Scientific, Inc.). 48 hours post transfection, the luciferase activity was measured using the dual-luciferase reporter assay system (Promega).

2.11 Statistical analysis

Data were presented as mean ± SD. Graphpad Prism version 5.0 software (GraphPad, San Diego, CA) was used for data analysis. The corrections between miR-214-5p or KLF5 expression and clinicopathological features were assessed by χ² test. Student t test was conducted to analyze the difference between two groups. One-way analysis of variance and Tukey test were performed to compare the difference among three or above groups. Kaplan-Meier curve and log-rank test were used for survival analysis. Cox proportional hazards regression model was used for multivariate survival analysis to assess prognostic factors that were significant in the univariate analysis. P-value <0.05 was considered to be statistically significant.

3.1 The general characteristics of patients in this study

The clinical characteristics of the 78 patients with HCC were summarized in Table 1. The average age of patients with HCC was 58.6 ± 9.9 years (range, 33-80 years). The number of male patients was 41, while that of female was 37. The clinicopathological features used in this study includes age, gender, tumor size, and tumor stage.

3.2 Expression of miR-214-5p in HCC tissues and cell lines

The expression of miR-214-5p in 78 pairs HCC tissues and HCC cell lines was examined by RT-qPCR. We found miR-214-5p downregulation was occurred in 60.3% (47/78) of HCC tissues when compared with matched normal tissues (>2 folds decrease, Figure 1A). Therefore, these patients were classified into low miR-214-5p expression group. The rests of the enrolled patients were classified into high miR-214-5p expression group accordingly. Moreover, we found the expression level of miR-214-5p was also significantly downregulated in HCC cell lines than in normal liver cell line (P < 0.05, Figure 1B).

3.3 Effect of miR-214-5p expression on cell proliferation, cell cycle, and cell migration

Following, we investigated the biological role of miR-214-5p since it was downregulated in HCC. The miR-214-5p mimic, inhibitor, and NC miRNA were transfected into HCC cell lines (Hep3B and Huh7) to manipulate miR-214-5p expression. RT-qPCR analysis showed that miR-214-5p expression was increased by miR-214-5p mimic but decreased by miR-214-5p inhibitor transfection (P < 0.05, Figure 1C). CCK-8 analysis demonstrated that miR-214-5p overexpression inhibited the cell proliferation of HCC cell lines, whereas miR-214-5p downregulation promoted cell proliferation (Figure 1D). Furthermore, we measured the cell cycle distribution by flow cytometry. The results indicated miR-214-5p mimic transfection increased the HCC cells at G0/G1 phase but decreased the cells at the S phase compared with NC miRNA transfected counterpart (Figure 1E). Importantly, we examined the cell migration using the wound-healing assay and we revealed that miR-214-5p mimic reduced the cell migration of HCC cell lines compared with the NC miRNA transfected counterpart (Figure 1F). These results indicate that miR-214-5p may inhibit cell proliferation, arrest cell cycle at G0/G1 phase, and inhibit cell migration.

3.4 KLF5 was a direct target of miR-214-5p in HCC

To understand the underlying mechanism of miR-214-5p in HCC, we searched for the potential targets of miR-214-5p using TargetScan. The results implied that KLF5 might be a possible target of miR-214-5p since it contains a putative binding sequence for miR-214-5p in its 3′-UTR (Figure 2A). To confirm this prediction, the dual-luciferase assay was conducted. As presented in Figure 2B, miR-214-5p mimic suppressed the relative luciferase activity of cells containing WT 3′-UTR of KLF5 construct but did not alter that of cells containing Mut 3′-UTR of KLF5 construct. To further validate KLF5 was a direct target of miR-214-5p, we measured KLF5 expression in miR-214-5p mimic or NC miRNA transfected cells. As expected, the protein expression of KLF5 was decreased by miR-214-5p overexpression (Figure 2C), which implied KLF5 was a direct target of miR-214-5p in HCC. Moreover, we investigated the expression KLF5 in HCC tissues and we found KLF5 expression was upregulated in 53.8% (42/78) of HCC tissues when compared with matched normal tissues (Figure 2D). Also, we examined the expression of KLF5 and miR-214-5p in HCC tissues and found they were inversely correlated (Figure 2E). Furthermore, the hyperexpression status of KLF5 in HCC tumor tissues was confirmed by IHC (Figure 2F).

3.5 Involvement of KLF5 in HCC cell proliferation, cell cycle progression, and cell migration processes mediated by miR-214-5p

To further investigate whether KLF5 was a mediator for the tumor suppressive role of miR-214-5p in HCC, we performed loss-of-function studies. As shown in Figure 3A, the transfection of siRNA targeting KLF5 decreased the protein expression of KLF5. Meanwhile, we found knockdown of KLF5 resulted in cell proliferation inhibition and cell cycle arrested in G0/G1 phase (Figure 3B,C). Moreover, we found the knockdown of KLF5 could abolish the effect of miR-214-5p downregulation on cell proliferation and cell cycle progression (Figure 3B,C). Notably, we also found the si-KLF5 inhibited cell migration and partially reversed the effect of miR-214-5p inhibitor on cell migration (Figure 3D).

3.6 Clinical significance of miR-214-5p and KLF5 expression in HCC

To further investigate the clinical significance of miR-214-5p and KLF5 expression in HCC, we investigated the correlations between miR-214-5p or KLF5 expression with the clinicopathological variables. As summarized in Table 1, the results indicated low miR-214-5p expression was correlated with tumor size (P = 0.028) and tumor stage (P = 0.014). Meanwhile, we also found the high KLF5 expression was correlated with the tumor size (P = 0.033) and tumor stage (P = 0.023). However, the correlations of miR-214-5p or KLF5 expression with gender and age were not statistically significant (all P > 0.05, Table 1).

Furthermore, the patients’ 5-year OS time were collected and analyzed by Kaplan-Meier curve and log-rank test. We found patients with low miR-214-5p expression (P < 0.05, Figure 4A) or high KLF5 expression (P < 0.05, Figure 4B) tend to had a poorer 5-year OS time and disease-free survival time. Univariate analysis showed that variables including miR-214-5p expression (P = 0.033), KLF5 expression (P = 0.021), tumor size (P = 0.032), and tumor stage (P = 0.031) had significantly prognostic influences on OS and disease-free survival time (Tables 2 and 3). However, variables including gender and age did not have significantly influence on OS and disease-free survival time (all P > 0.05, Tables 2 and 3). Moreover, multivariate analysis using Cox proportional hazard model to analyze the variables that were significant in the univariate analyses revealed that miR-214-5p expression (P = 0.026) and KLF5 expression (P = 0.019) were independent prognostic factors for OS, along with tumor size (P = 0.036) and tumor stage (P = 0.016, Table 2). Furthermore, we found miR-214-5p expression (P = 0.021) and KLF5 expression (P = 0.029) were also independent prognostic factors for disease-free survival, along with tumor size (P = 0.024) and tumor stage (P = 0.022, Table 3).