SEARCH

SEARCH BY CITATION

Keywords:

  • hepatocellular carcinoma;
  • microRNA;
  • miR-155;
  • p53;
  • p21waf1/cip1

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

BACKGROUND:

Recent research has suggested that the oncomir microRNA 155 (miR-155) is up-regulated in hepatocellular carcinoma (HCC). In this study, the authors investigated the tumorigenic mechanism of this oncomir in the development of HCC.

METHODS:

Quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) was conducted to analyze the expressions of miR-155 and its potential target genes in paired tumor tissues and adjacent tumor-free tissues and in disease-free liver tissue samples. The in silico predicted target genes of miR-155 were assessed by dual-luciferase reporting assay, real-time RT-PCR, and Western blot analyses. U6 promoters that drive miR-155 precursor overexpression and miR-155 tough decoy knock-down constructs were used to study its affects on cell proliferation in vitro and on tumor formation in nude mice.

RESULTS:

Quantitative RT-PCR demonstrated a gradual ascension of miR-155 expression in cirrhotic liver tissues and in HCC tumor tissues compared with low expression levels in normal liver tissues. Ectopic expression of miR-155 in HepG2 cells enhanced its tumorigenesis, whereas depletion of the endogenous miR-155 reversed these tumorigenic properties. Ectopic expression of sex-determining region Y box 6 (SOX6) was able to reverse the growth-promoting property of miR-155. Concordantly, the results demonstrated for the first time that SOX6 is a direct target of miR-155. Further analysis revealed that SOX6 reduced cell growth by up-regulating p21waf1/cip1 expression in a p53-dependent manner. In addition, a decline in p21waf1/cip1 expression caused by miR-155 could be reversed by SOX6 expression.

CONCLUSIONS:

The current data indicated that SOX6 is a novel target of miR-155 and that miR-155 enhances liver cell tumorigenesis at least in part through the novel miR-155/SOX6/p21waf1/cip1 axis. These findings suggest that miR-155 may be a potential target for HCC treatment. Cancer 2012. © 2011 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

Hepatocellular carcinoma (HCC) is 1 of the most common malignancies worldwide with a 5-year survival rate of only approximately 5% if left untreated.1 The prevalence of HCC varies geographically: Asia and sub-Saharan Africa have the highest incidence2 and China alone accounts for >55% of newly diagnosed HCC cases each year. The known major risk factors for HCC are hepatitis B virus (HBV) and hepatitis C virus (HCV) infection, alcohol abuse, exposure to xenbiotics,3 and specific genetic disorders.4

MicroRNAs (miRNAs) are a class of short (19-22 nucleotides), noncoding RNA sequences that negatively regulate their target genes. It is recognized that these miRNAs are involved in the regulation of a variety of biologic processes, including cellular proliferation, differentiation, and apoptosis. More recently, accumulating evidence has suggested that aberrant expression of miRNAs may contribute significantly to the development of several human malignancies.5

The focus of the current study is miRNA 155 (miR-155), which is located on chromosome 21 and is transcribed from the B-cell integration cluster.6 In addition to its involvement in proinflammatory responses, recent evidence indicates that this miRNA has a significant role in the process of carcinogenesis of several solid tumors, including colon cancer,7 breast cancer,8 and pancreatic ductal adenocarcinoma.9 Recently, Tili et al reported that miR-155 can act as a mutator and, together with an miR-155-linked inflammatory environment, may be a potential mechanism connecting inflammation and cancer.10 Wang et al used a mouse model of chemical carcinogenesis and also demonstrated that a gradual increase in miR-155 expression occurred during the progressive hepatocyte transformation that they documented.11 This observation strongly suggests that aberrant miR-155 expression may play some role during the development of HCC. In silico screening for potential miR-155 targets identified the transcriptional regulator sex-determining region Y box 6 (SOX6) as a candidate, prompting an investigation of the association between SOX6 expression and this oncomir in the development of HCC.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

Patient Specimens and Cell Lines

Thirty pairs of matched primary human HCC samples and adjacent tumor-free tissue samples were obtained from patients who underwent surgical resection between 2009 and 2010, and 8 normal liver tissue samples were obtained from healthy donors. These specimens were provided by the Affiliated Oncology Hospital of Zhengzhou University. All patients who were enrolled were Han Chinese. Both tumor samples and noncancerous samples were confirmed histologically, and all tumors originated from patients who had a background of chronic hepatitis B infection and cirrhosis without any other hepatotropic virus infections. Human liver cell lines (HepG2, HBxF344, Huh-7, PLC-PRF-5, SNU-449, SK-Hep-1) were stored in our laboratory and maintained in Dulbecco modified Eagle medium supplemented with 10% fetal bovine serum (Gibco, Carlsbad, Calif).

This study was approved by the Ethics Committee of Peking University Health Science Center. Informed consent was obtained from each participant.

Construct of Plasmid Recombinants

Human SOX6 combinational DNA (without the 3′ untranslated region [UTR]) was a kind gift from Professor Kenji Ohe of Fujita Health University.12 The fragment was reconstructed into the cytomegalovirus immediate-early promoter (pCMV) and viral hemagglutinin (HA) gene insert (pCMV-HA) backbone plasmid.

The p21-luciferase (p21-Luc) plasmid was kindly provided by Professor Weiguo Zhu from the Department of Biochemistry, Peking University Health Science Center. The p21-Luc plasmid was used as a reporter plasmid that contained wild-type p21 promoter upstream of the luciferase gene.

Tough-decoy (TuD) RNA is a new strategy that can efficiently suppress a specific endogenous microRNA.13 The miR-155 TuD cassette was designed as suggested13 and was synthesized according to the GenScript Corporation cassette protocol (Piscataway, NJ). The U6-miR-155-TuD was produced based on the pRNA-U6.1/Neo-small interfering Firefly luciferase (siFluc) vector, and the siFluc fragment was replaced with the miR-155-TuD cassette to create the U6-driven miR-155 antagonist. To construct the miR-155 expression vector, the human miR-155 gene and its 5′-flanking and 3′-flanking regions (100 base pairs and 105 base pairs, respectively) was amplified from HEK-293 genomic DNA and cloned into the TA vector. It was retrieved and cloned into the pRNA-U6.1/ Neo-siFluc vector to create the U6-driven miR-155.

Generation of Stable Hepatocellular Carcinoma Cell Lines That Overexpress/Knock-Down MicroRNA 155

Stable cell lines were generated by transfection with U6-miR-155-TuD or U6-miR-155-precursor plasmids into HepG2 cells, and U6-siFluc also was transfected as a control. Cells were split 1:3 48 hours after transfection; then, G418 (600 μg/mL) was added to select the transfected clones. At least 3 separate clones were picked up for each type of cell. Taqman polymerase chain reaction (PCR) analysis was performed to detect whether miR-155 expression was successfully inhibited or increased.

Luciferase Reporter Assay

Cells were seeded into 12-well plates. Twenty-four hours later, the cells were transfected, and the luciferase activity in each well was quantified 24 hours after using a dual-luciferase reporter kit (catalog no. E1910; Promega, Madison, Wis) according to the manufacturer's instructions.

RNA Extraction and Quantitative Reverse Transcriptase-Polymerase Chain Reaction

Total RNA was prepared using TRIzol reagents (Invitrogen, Carlsbad, Calif) in accordance with the manufacturer's instructions. Real-time PCR was carried out using SYBR Green on a Lightcycler 480II Real-Time PCR Detection System (Roche, Indianapolis, Ind). Each sample was analyzed in triplicate.

The expression of mature miRNAs was detected using Taqman microRNA assays that were specific for human serum albumin miR-155 (hsa-mir-155) (lot 0910082-A; Applied Biosystems, Foster City, Calif) on 30 matched samples of HCC and adjacent tissues and on 8 samples of completely normal liver tissues. Each sample was analyzed in triplicate using the Taqman Universal PCR Master Mix II (lot 0907001; Applied Biosystems). Taqman microRNA assays for U6 RNA (lot 0911354-C; Applied Biosystems) were used to normalize the relative abundance of miRNA.

Immunoblot Analysis

Proteins were extracted using a radioimmunoprecipitation assay buffer. Protein samples (30 μg each) were loaded onto 12% sodium dodecyl sulfate-polyacrylamide gels, electrophoresed, and transferred onto nitrocellulose membranes (Amersham Biosciences, Uppsala, Sweden). Briefly, membranes were blocked with 5% dried milk in phosphate-buffered saline (PBS) buffer for 2 hours followed by incubation with primary antibody for 2 hours; after 3 washes with PBS containing 0.1% Tween-20, the membranes were incubated with secondary antibodies conjugated to LI-COR IRDye for 1 hour at room temperature, and the antibodies were detected using the Odyssey Imager (LI-COR Biosciences, Lincoln, Neb). For coimmunoprecipitation experiments, HepG2 cells transiently transfected with HA-SOX6, together with either empty vector or p53-expressing plasmid, were lysed in radioimmunoprecipitation assay buffer. One milligram of cell lysate was incubated for 2 hours with 30 μL of protein A-Sepharose (Amersham Biosciences) precoated with 2 μg of anti-HA antibody. The pellets were washed 6 times with radioimmunoprecipitation assay buffer before adding 5 times loading buffer. Samples were then treated as described above.

Cell Proliferation Assay

Cell proliferation was monitored using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) method. Cells (1000 per well) were allowed to grow in 96-well plates. MTT solution was added into the cell culture at a final concentration of 5 μg/mL and was allowed to remain in culture for 4 to 6 hours before measurement. Cell proliferation was documented every 24 hours by measuring the absorbance in a microplate reader (Bio-Rad, Hercules, Calif).

Anchorage-Independent Colony-Formation Assay in Soft Agar

Colony-formation assays were performed using a soft agar kit (GMS10024 vA; GenMed Scientifics, Inc., New Castle, Del) in 6-well plates according to the manufacturer's instructions. In total, 1 × 104 each of stable miR-155 overexpressing cells, stable miR-155 knock-down cells, and mock-transfected control cells were seeded, respectively, in top agar over the base agar. The plates were incubated at 37°C for more than 2 weeks until the visible colony was formed, the colonies were photographed under a microscope with a digital camera and also were counted under the same microscope after dividing the plates into grids.

Tumorigenicity Assays in Nude Mice

The in vivo tumorigenesis ability of miR-155 was investigated in a tumor xenograft experiment. In this experiment, 5 × 106 miR-155 overexpressing HepG2 cells (stable cell line) or mock-transfected control HepG2 cells were suspended in 100 μL PBS and then injected subcutaneously into either side of the posterior flank of a 6-week-old male nude mouse. Sixteen mice were used. Tumor formation in nude mice was monitored over a 4-week period. The tumor volume was calculated by the formula V = 0.5 × L × W2 (in millimeters).14 Where, L is the length of the tumor measured in nude mouse; W is the width of the tumor measured in nude mouse.

Statistical Analysis

For statistical analyses, 2-tailed Student t tests were conducted using GraphPad Prism software (version 5.0a; GraphPad Software, Inc., San Diego, Calif).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

Up-Regulation of MicroRNA 155 Expression in Hepatitis B Virus-Related Primary Human Hepatocellular Carcinoma

An increase in miR-155 expression during the early stages of hepatocarcinogenesis induced by a chemical diet in C57BL/6 mice has been reported, and the up-regulation of miR-155 also has been observed in primary human HCC.11 Both to confirm this latter observation and to extend it to include HBV-related HCC, a Taqman-based assay was used. Compared with disease-free liver tissues, increased miR-155 expression levels were observed both in HCC tumors and in adjacent tumor-free tissues. Although there was only a marginal increase in the level of miR-155 expression in tumor tissues versus the level in adjacent, cirrhotic, tumor-free tissues, compared with the level in disease-free liver tissues, the increased miR-155 expression was statistically significant (P ≤ .05). In addition, in 12 of the 30 HCC samples analyzed, there were significantly higher levels of miR-155 expression in tumor tissues compared with the levels in matched, adjacent, paratumor tissues (Fig. 1). Overall, the results confirmed that miR-155 expression is up-regulated in primary human HCC. The comparative analysis of matched tumor and paratumor samples also suggested that the up-regulation of miR-155 expression occurs at an early stage in the process of hepatocarcinogenesis. Expression levels of miR-155 also were measured in the cell lines HepG2, HBxF344, Huh-7, PLC-PRF-5, SNU-449, and SK-Hep-1 (Fig. 2A). We also measured changes in miR-155 expression levels in HepG2, Huh-7, and HBxF344 cells that were transfected with U6-miR-155 precursor or U6-miR-155-TuD (Fig. 2B-D).

thumbnail image

Figure 1. MicroRNA 155 (miR-155) is up-regulated in primary hepatocellular carcinoma (HCC). The expression of miR-155 was measured by using Taqman reverse transcriptase-polymerase chain reaction analysis in primary human HCCs, paratumor tissues, and normal liver tissues. Each sample was analyzed in triplicate. The results are presented as box-and-whisker plots. The horizontal line in each box represents the median value of miR-155 to U6 miRNA. Boxes represent the 50th and 75th percentile ranges of scores, and whiskers represent the highest and lowest values.

Download figure to PowerPoint

thumbnail image

Figure 2. MicroRNA 155 (miR-155) levels were measured by using the Taqman quantitative polymerase chain reaction method (A) in the Huh-7, PLC-PRF-5, SNU-449, HepG2, HBxF344, and SK-Hep-1 cell lines and (B-D) in the cell lines (B) HepG2, (C) HBxF344, and (D) Huh-7 24 hours or 48 hours after transient transfection with miR-155 precursor, miR-155 “tough decoy,” or vector control. Data were normalized to U6 miRNA, and the results shown are the mean ± standard deviation of triplicate experiments.

Download figure to PowerPoint

MicroRNA 155 Expression Promotes Tumorigenic Properties in Hepatocellular Carcinoma-Derived Cell Lines

Two plasmid constructs were used to examine the tumorigenic potential of miR-155 in the HepG2 cell line, which was derived originally from HCC tissue. The first plasmid carried miR-155 under the control of the U6 promoter and was transfected into HepG2 cells, and a line of cells that stably overexpressed miR-155 was isolated. The second plasmid expressed an miR-155-TuD RNA again under the control of the U6 promoter. The TuD RNA technology developed by Haraguchi et al13 was designed to efficiently suppress the endogenous expression of specific miRNAs and, in this case, was used to allow the isolation from transfected HepG2 cells of a stable cell line in which endogenous expression of miR-155 had been suppressed. The overexpression of miR-155 and the suppression of endogenous miR-155 expression were confirmed in both cell lines using a Taqman RT-PCR assay (data not shown).

The cell proliferation capacity of these 2 cell lines was measured using an MTT assay. The results revealed that growth of the cell line that overexpressed ectopic miR-155 was significantly increased to approximately 175% of that observed in control HepG2 cells (Fig. 3Ai). In contract, a small suppression of cell proliferation was observed in the miR-155 knock-down cell line. To further substantiate the proliferative activity of miR-155 2 hepatocyte-derived cell lines, Huh-7 and HBxF344 cells were transiently transfected with the 2 miR-155 plasmids. In this experiment, the growth rates of the miR-155 overexpressing cells were enhanced compared with the rates in mock-transfected controls, whereas knock-down of endogenous miR-155 expression resulted in growth inhibition (Fig. 3Aii).

A second in vitro assay of tumorigenic potential was conducted to examine the ability of the stably transfected cell lines to form colonies in agar. This revealed that cells stably transfected with the miR-155-TuD plasmid had decreased colony-forming efficiency relative to mock-transfected controls, whereas the cells that overexpressed ectopic miR-155 had greater colony-forming efficiency (Fig. 3B).

thumbnail image

Figure 3. Ectopic expression of microRNA 155 (miR-155) can enhance the tumorigenic properties of hepatocellular carcinoma (HCC) cell lines. (Ai) The growth of HepG2 cells with stable expression of miR-155, knock down of miR-155, or vector control was measured using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay according to the number of days (d) in culture. Each cell type was analyzed in 6 experiments. (Aii) An MTT assay was used to measure the expression of miR-155 in HBxF344 and Huh-7 cells after transient overexpression or knock down of miR-155. Cells were transfected with miR-155 precursor, miR-155 “tough decoy,” and normal control (NC) plasmid; and, 48 hours after transfection, the MTT assay was performed to measure their growth abilities. (B). Identical numbers of HepG2 cells with overexpression or knock down of miR-155 and a vector control were used for soft agar assay for colony formation. After, 2 weeks, the colonies were scanned and counted. (C). The in vivo tumorigenesis ability of miR-155 was investigated in a tumor xenograft experiment. HepG2 cells or mock-transfected control HepG2 cells were injected subcutaneously into either side of the posterior flank of the same male nude mouse. Tumor formation in nude mice was monitored over a 4-week period. The tumor volume was calculated by the formula V = 0.5 × L×W2. The volumes of tumors are illustrated in mice that were injected with different cell types at different time points. In total, 16 mice were used in this experiment, Curves represent the mean ± standard deviation.

Download figure to PowerPoint

To address tumorigenic potential in vivo, a tumor formation assay was carried out in nude mice. For this experiment, cells that stably overexpressed ectopic miR-155 were injected into the posterior flank of a nude mouse, and an equal number of mock-transfected cells were injected into the other posterior flank of the same animal. Tumor formation was then monitored daily; and, after 7 days, tumors appeared in 11 of 16 animals in the flank that was injected with miR-155-overexpressing cells. By contrast, tumors were observed in only 7 of 16 animals injected with the mock-transfected HepG2 cells. In addition, significant differences in average tumors size were observed on day 21 and at the end of observation postinjection, (tumor volume on day 28, 1.744 mm3 vs 1.194 mm3; P < .05) (Fig. 3C). This result was obtained with several miR-155 overexpressing clones (data not shown), demonstrating the reproducibility of the phenotype.

Identification of SOX6 as a Direct Target of MicroRNA 155

It is generally accepted that miRNAs function by regulating expression levels of their target gene(s). Therefore, to identify the specific target of miR-155, the TargetScan and PicTar algorithms were used. This in silico analysis identified hundreds of genes that harbored the conserved miR-155 recognition sites in their 3′ untranslated regions (UTRs), but attention was focused on identified genes that have been linked with the regulation of cell growth. In total, 11 such genes (cyclin D1 [CCND1], E2F transcription factor 2 [E2F2], F-box protein 11 [FBXO 11], F-box protein 33 [FBXO 33], kinase suppressor of ras 1 [KSR1], p21 protein-activated kinase 7 [PAK7], phosphoinositide-3-kinase regulatory subunit 1 [PIK3R1], SOX1, SOX6, SOX10, and SOX11) were enrolled as potential targets of miR-155 and subjected to further experimental analysis. For these experiments, the putative 3′-UTR target site of each gene was inserted downstream of a Firefly luciferase reporter gene to generate a series of pGL3-Luc-3′-UTR plasmids. These plasmids were then transfected into HepG2 cells together with a renilla luciferase vector (pRL-TK) to allow the monitoring of transfection efficiency. The cells also were cotransfected with the miR-155 expression plasmid described above. Several these 3′-UTR constructs (E2F2, FBXO11, KSR1, and SOX6) had reduced levels of Firefly luciferase in dual-luciferase assays (data not shown).

Western blot and quantitative reverse transcriptase (qRT)-PCR analyses of the expression level of each candidate gene in the presence of ectopically expressed miR-155 were carried out for the genes that had reduced Firefly luciferase in the dual-luciferase assay. This revealed that the only gene directly regulated by miR-155 was SOX6, which had a 25% reduction in messenger RNA (mRNA) levels in the qRT-PCR assay and had even more significant suppression in the levels of protein detected on Western blot analysis (Fig. 4, Fig. 5Ai,Aii). These results indicated that miR-155 suppressed SOX6 expression not only by blocking translational processes but also by inducing SOX6 mRNA cleavage directly, and the former was dominant. To independently confirm the identification of SOX6 as an miR-155 target, the impact of depleting miR-155 expression in a second liver-derived cell line (HBxF344) was examined. This revealed that SOX6 mRNA levels increased by 55% compared with mock-transfected controls when miR-155 was depleted using TuD RNA (Fig. 5B).

thumbnail image

Figure 4. (A) The predicted cognate site of microRNA 155 (miR-155) in the 3′ untranslated region (UTR) of sex-determining region Y box 6 (SOX6) and human serum albumin miR-155 (hsa-miR-155) is illustrated. (B) In an analysis of luciferase activity, HepG2 cells were cotransfected with U6-miR-155 (miR-155 expression) or control vector (no mir-155 expression), Firefly luciferase reporter containing the 3′-UTR of SOX6 or the control 3′-UTR without the miR-155 complementary site, and renilla construct as an internal control. Luciferase activity was assayed 24 hours after transfection. Firefly luciferase values normalized for renilla luciferase are presented. The data represent the mean ± standard error of 9 independent experiments performed in triplicate.

Download figure to PowerPoint

thumbnail image

Figure 5. Sex-determining region Y box 6 (SOX6) is a direct target of microRNA 155 (miR-155). (Ai) SOX6 mRNA levels in HepG2 cell transfected with U6-miR-155 or mock-transfected normal control (NC) were analyzed by real-time reverse transcriptase-polymerase chain reaction (RT-PCR). The data were normalized to hydroxymethylbilane synthase (HMBS). (Aii) Western blot analysis was used to analyzed SOX6 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) levels in HepG2 cell extracts that were transfected with control or U6-miR-155 (+ indicates positive; −, negative). (B) SOX6 messenger RNA (mRNA) levels in HBxF344 cells transfected with U6-miR-155 Tough Decoy or mock-transfected controls were analyzed by real-time RT-PCR. (C) The expression of mRNA in primary hepatocellular carcinomas and in matched adjacent paratumor tissues was measured by real-time RT-PCR. HMBS was used as internal reference. Horizontal lines in each box represent the median value of SOX6 mRNA.

Download figure to PowerPoint

Having established SOX6 as a target for miR-155, its expression levels in primary HCC tissues were assayed. There was a statistically significant (P = .01) decrease in SOX6 mRNA levels in HCC tissues compared with the levels in matched, adjacent, tumor-free tissues (Fig. 5C), as expected. It is also noteworthy that, when all samples analyzed were grouped according to their levels of miR-155 expression, there was a correlation between SOX 6 expression and miR-155 levels: Higher SOX6 levels were observed when miR-155 expression levels were reduced, irrespective of the tumor status of the tissue. However, the number of samples in each grouping was not sufficient to give this result statistical significance (data not shown).

To further substantiate the proposed interaction of miR-155 and SOX6 in HCC progression, we transiently transfected and expressed HA-tagged SOX6 (pCMV-HA-SOX6, without the 3′UTR in it which could not be suppressed by miR-155) in a cell line that expressed ectopic miR-155 to examine its effect on cellular proliferation. The results indicated that additional SOX6 expression was able to attenuate the induction of cell proliferation by miR-155 (Fig. 6A). In addition, ectopic expression of SOX6 alone was able to suppress the proliferation of HepG2 cells (Fig. 6B). The expression level of SOX6 was sharply increased when ectopic pCMV-HA-SOX6 was induced in HepG2 cells (Fig. 6C), as expected.

thumbnail image

Figure 6. Growth acceleration capability can be reversed by ectopic expression of sex-determining region Y box 6 (SOX6). (A) Cells with stable overexpression of microRNA 155 (miR-155) were transfected with SOX6 or control vector, and HepG2 cells were stably transfected with U6-siFluc control plasmid (Mock). Cells were seeded in 96-well plates, and cell growth was monitored every 24 hours using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. (B) HepG2 cells were transfected with SOX6 or with control vector in 96-well plates, and cell growth was monitored every 24 hours using the MTT assay. (C) HepG2 cells were transfected with cytomegalovirus promoter (pCMV)-SOX6 or with control vector. Forty-8 hours later, Western blot analysis was performed using cell lysis (+ indicates positive; −, negative).

Download figure to PowerPoint

p53-Dependent Activation of p21 by SOX6 in Hepatocellular Carcinoma-Derived Cell Lines

To investigate the mechanism underpinning SOX6 involvement in cell growth and proliferation, several the key factors (p21, p27, cyclin D1, and cyclin A) involved in controlling cell proliferation were studied. Western blot analysis revealed a significant increase in p21 levels in HepG2 cells that expressed ectopic SOX6, and this increase had a dose-dependent correlation with the level of SOX6 expression (Fig. 7Ai,Aii). This interplay of p21 and SOX6 expression levels was then confirmed using qRT-PCR analysis (Fig. 7B).

thumbnail image

Figure 7. Sex-determining region Y box 6 (SOX6) can activate p21waf1/cip1 (p21) expression in hepatocellular carcinoma cell lines in a p53-dependent manner. (Ai) Western blot analyses of p21, p27, cyclin D1 (CCND1), and cyclin A (CCNA) protein levels are illustrated in HepG2 cells that overexpressed SOX6. pCMV indicates cytomegalovirus promoter; HA, hemagglutinin; −, negative; +, positive. (Aii) SOX6-induced p21 expression was dose dependent. (B) Levels of p21 messenger RNA were measured by real-time reverse transcriptase-polymerase chain reaction (RT-PCR) in HepG2 cells that were transfected with SOX6 or with mock-transfected control. C-terminal binding protein (CTBP) was used as an endogenous control. Data shown are the mean ± standard error of 9 independent experiments done in triplicate. (C) SOX6-induced p21 up-regulation was p53-dependent. HepG2 cells were cotransfected with E6, SOX6, or control vector. Forty-eight hours after transfection, cell lysates were subjected to Western blot analysis. HepG2 cells were cotransfected with or without SOX6 and control vector together with different doses of E6 plasmid, and cell extracts were subjected to Western blot analysis. (D) A dual-luciferase assay was conducted to analyze the regulation of SOX6 by p21 promoter. HepG2 cells or Huh-7 cells were cotransfected with p21-luciferase and either pCMV-HA-SOX6 or mock-transfected control. Renilla luciferase vector was used as an internal control. Luciferase activity was assayed 24 hours after transfection. Firefly luciferase values normalized for renilla luciferase are presented. The data represent the mean ± standard error of 9 independent experiments done in triplicate. (E) HepG2 cells that harbored wild-type p53 were transfected with the indicated constructs and cell lysates and were subjected to coimmunoprecipitation (co-IP) with the indicated antibodies. (F) MicroRNA 155 (miR-155) down-regulates p21 by inhibiting SOX6. HepG2 cells were transfected with U6-miR-155 precursor or with U6-miR-155 precursor plus pCMV-SOX6. The cells that were transfected with control vector were used as a mock-transfected control. Western blot analysis was performed 48 hours after transfection.

Download figure to PowerPoint

The possible involvement of p53 in this interplay was examined first by analyzing the impact of ectopic expression of the p53 inhibitor E6 on the SOX6 induction of p21. It is widely accepted that E6 has a function of inhibiting the activity of p53.15 Western blot analysis revealed that ectopic E6 expression caused a reduction in the SOX6-mediated induction of p21 in HepG2 cells (Fig. 7C). To further substantiate the involvement of p53, a luciferase reporter assay was used in which reporter expression was driven by a wild-type p21 promoter. In the p53 wild-type HepG2 cell line, ectopic SOX6 expression was able to cause a marked activation of luciferase reporter expression (Fig. 7D). By contrast, in Huh-7 cells, which have a mutated p53 gene, ectopic SOX6 expression caused only a slight increase in p21 promoter transcription activity in a reporter assay experiment (Fig. 7D). Another coimmunoprecipitation experiment demonstrated that SOX6 can bind with p53 (Fig. 7E). The physical binding between SOX6 and p53 may provide a clue to understanding the potential molecular mechanism through which SOX6 play its growth-inhibition role in HCC cells.

Having established the interaction between SOX6 and p21-induced cell proliferation, it was important to investigate the involvement of miR-155 in this interaction. For this investigation, Western blot analysis of p21 levels in HepG2 cells that ectopically expressed miR-155 was used. The results revealed that p21 was markedly decreased and that this decrease could be reduced by the concomitant ectopic expression of SOX6 (Fig. 7F).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

The development of HCC is clearly a multistep process involving multiple factors. To date, efforts to understand the principles underpinning this malignancy have focused on molecular changes occurring in cell signals and on genomic changes. In the current study, we focused on a different aspect of HCC pathogenesis in seeking to define the involvement of miR-155 in the process. It was reported first that this miRNA had a close relation with immunity and immune system-linked inflammatory disease.16 More recently, it was implicated in the pathogenesis of several malignancies, including pancreatic cancer11 and lymphocytic leukemia.17 In addition, Wang et al reported a significant increase in miR-155 levels in primary HCC tissues in humans.11 However, to our knowledge, the detailed mechanism(s) surrounding any role that miR-155 may play in HCC development has not been reported previously.

In the current study, elevated miR-155 expression was defined first in HBV infection-related HCC tumor tissues. Compared with the levels in disease-free liver tissues, miR-155 expression levels were increased in both HCC tumor tissues and in adjacent, tumor-free tissues, although only the increase in tumor tissues was statistically significant. The observation of elevated miR-155 expression levels in adjacent, cirrhotic, tumor-free tissues implicated an early aberration of miR-155 in the preneoplastic stage of HBV-related HCC. To some extent, this results was consistent with the data reported by Yoon et al,18 who observed that miR-155 expression was higher in non-neoplastic tissues compared with normal liver tissues. We observed only a marginal increase in miR-155 expression in the malignant HCC tissues compared with the adjacent tumor-free tissues. Wang et al demonstrated that 16 of 20 primary HCCs had increased miR-155 expression, whereas our data indicated only a slight increase, as mentioned above. A potential reason for this difference may relate to patient history, because all of the patients in our study had a previous history of cirrhosis in their whole liver, whereas the samples reported by Wang et al did not have a cirrhotic background, and the data reported by Sun et al18 indicated that the arising of miR-155 in hepatocarcinogenesis was at an early stage and occurred before cirrhosis.

The full range of cirrhosis pathologies are widely regarded as representative of the early stages in the development of HCC; therefore, the data on aberrant miR-155 expression in samples from patients with chronic HBV are consistent with it also being indicative of an early stage in HCC development in this patient cohort as originally suggested by Wang et al based on their mouse model studies.11 This suggestion that miR-155 functions as an oncomir in the liver was supported further by the demonstration that overexpression and knock down of miR-155 expression in several HCC-derived cell lines (HepG2, HBxF344, and Huh-7) affected tumorigenic properties, such as the rate of cell proliferation, anchorage-independent growth, and tumor-forming ability, in nude mice. However, we should be cautious when interpreting these observations, because the level of ectopically expressed miR-155 may be higher than the physiologic level. We noted that, except for HBxF344 cells, the HCC cell lines tested here had relatively low miR-155 expression levels compared with human HCC tissues. This indicates that in vitro-cultured HCC cell lines may be different in some manner from primary human HCC tissues. It has been suggested that inflammatory cytokines like interferon γ, tumor necrosis factor α, and interleukin 1β may enhance miR-155 expression,19 and the in vitro-cultured HCC cells lack such stimulation provided by the tumor microenvironment.

It is has been established that miRNAs function primarily through negatively regulating the expression of their specific target gene(s). Consequently, it is important to identify the specific target of miR-155 both to allow an assessment of the relevance of oncomir function and to unravel the detailed molecular mechanism of miR-155 action. In silico screening has identified hundreds of potential miR-155 targets, but less than 3 dozen have been validated to date.19 The current study focused on 11 potential targets identified by in silico screening that are known to function in cell cycle progression, transcription, cell survival, and differentiation. The expression of several of these potential target genes was affected by miR-155 overexpression in a dual-luciferase reporter system. However, we confirmed that, in HepG2 cells, only SOX6 was reduced at both the mRNA level and the protein level, leading to its proposal in this study as the primary target for miR-155 action.

It was reported previously that the attenuation of SOX6 expression stimulated the proliferation of insulinoma INS-1E and of NIH-3T3 cells, whereas overexpression of the gene resulted in the inhibition of cell growth.19 The same study also indicated that SOX6 can suppress cyclin D1 activity by interacting with β-catenin and histone deacetylase 1.20 However, using a TOPFlash (T-cell factor reporter plasmid) assay,21 we were unable to demonstrate that SOX6 overexpression resulted in the suppression of wnt signaling activity.

More recent evidence in support of a role for SOX6 in tumorigenesis has come from a study of esophageal cancer in which it was suggested that SOX6 can inhibit cell proliferation by affecting several genes involved in regulating cell cycle events, including cyclin A, cyclin D1, cyclin-dependent kinase 4, p21, and p27. However, in the current study, our initial experiments were unable to demonstrate that miR-155 had any effect on cyclin D1 levels. Furthermore, in our subsequent experiments involving the ectopic overexpression of SOX6, there was no significant impact on the expression of cyclin D1 or on other cycle-progression modulators, such as cyclin A or p27, whereas the levels of p21 were significantly up-regulated. These results suggest that, in this hepatic system, SOX6 plays its growth-inhibition role through the up-regulation of p21, which is quite different from its reported modulation of cyclin D1 in pancreatic tumors.

The final series of experiments reported here were designed to determine whether the overexpression of miR-155 could directly affect the expression of p21. In HepG2 cells, which express moderate levels of SOX6, transient ectopic expression of miR-155 was able to decrease endogenous levels of p21. In addition, cotransfection of these cells with a vector that produced ectopic SOX6 overexpression was able to alleviate the depression of endogenous p21 levels caused by miR-155 expression. These results allowed us to conclude that miR-155 plays a tumorigenic role at least in part through the inhibition of SOX6 expression, which, in turn, leads to a p53- dependent inactivation of the p21 growth regulator. Nevertheless, we noted the effect of ectopic miR-155 overexpression or endogenous miR-155 knock down on proliferation of FBxF344 cells and Huh-7 cells, both of which harbor p53 mutations. This result suggested that pathways other than p53/p21waf1/cip1 are relevant to miR-155 oncogenicity.

In conclusion, the current study has revealed a novel mechanism involving the regulation of SOX6 and p21 by miR-155, and we postulate that this mechanism is a key step in the multistep process leading to the development of HCC. We hope that further investigation of this process will help improve our understanding of this globally important malignancy and possibly lead to the development of novel therapeutic strategies targeted at miR-155.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

We thank Professor Kenji Ohe at Fujita Health University for his kind help in providing the SOX6 construct.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES

This work was supported by the National S&T Major Project for Infectious Diseases Control (2012ZX10002-007 to X. M. Chen) and by the Natural Science Foundation (30771099 to F. Lu). This study was also supported in part by Professor Guangping Gao at the University of Massachusetts Medical School.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. FUNDING SOURCES
  9. REFERENCES