SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Hepatocellular carcinoma (HCC) is a major liver malignancy. We previously demonstrated that deregulation of epigenetic regulators is a common event in human HCC. Suppressor of variegation 3-9 homolog 1 (SUV39H1), the prototype of histone methyltransferase, is the major enzyme responsible for histone H3 lysine 9 trimethylation, which, essentially, is involved in heterochromatin formation, chromosome segregation, and mitotic progression. However, the implication of SUV39H1 in hepatocarcinogenesis remains elusive. In this study, we found that SUV39H1 was frequently up-regulated in human HCCs and was significantly associated with increased Ki67 expression (P < 0.001) and the presence of venous invasion (P = 0.017). To investigate the role of SUV39H1 in HCC development, both gain- and loss-of-function models were established. SUV39H1 overexpression remarkably enhanced HCC cell clonogenicity, whereas knockdown of SUV39H1 substantially suppressed HCC cell proliferation and induced cell senescence. In addition, ectopic expression of SUV39H1 increased the migratory ability of HCC cells, whereas a reduced migration rate was observed in SUV39H1 knockdown cells. The significance of SUV39H1 in HCC was further demonstrated in a nude mice model; SUV39H1 knockdown drastically inhibited in vivo tumorigenicity and abolished pulmonary metastasis of HCC cells. We also identified microRNA-125b (miR-125b) as a post-transcriptional regulator of SUV39H1. Ectopic expression of miR-125b inhibited SUV39H1 3'-untranslated-region–coupled luciferase activity and suppressed endogenous SUV39H1 expression at both messenger RNA and protein levels. We have previously reported frequent down-regulation of miR-125b in HCC. Interestingly, miR-125b level was found to be inversely correlated with SUV39H1 expression (P = 0.001) in clinical specimens. Our observations suggested that miR-125b down-regulation may account for the aberrant SUV39H1 level in HCC. Conclusion: Our study demonstrated that SUV39H1 up-regulation contributed to HCC development and metastasis. The tumor-suppressive miR-125b served as a negative regulator of SUV39H1. (HEPATOLOGY 2013)

Hepatocellular carcinoma (HCC) is a prevalent malignancy worldwide and ranks as the third leading cause of cancer-related death. HCC is highly heterogeneous and develops through complex multistep processes that are accompanied by the acquisition of various molecular abnormalities.1,2 In addition to genetic alterations, such as chromosomal deletion and gene mutation, epigenetic dysregulation has been evidently demonstrated as a key event in liver cancer pathogenesis.

Epigenetic regulation generally refers to the changes in DNA methylation and histone modification pattern that modify gene transcription without affecting DNA sequence.3 Aberrant DNA hypermethylation in HCC has been frequently observed on promoter regions and accounts for the underexpression of tumor-suppressor genes, such as cyclin-dependent kinase inhibitor p16/INK4A,4,5 E-cadherin,6 phosphate and tensin homolog (PTEN),7 deleted in liver cancer 1 (DLC1),8 and tissue factor pathway inhibitor 2 (TFPI-2).9 Apart from DNA methylation, deregulated histone methylation has gained recent recognition as another important epigenetic alteration in carcinogenesis. Histone methylation critically determines chromosomal structure and stability, as well as the accessibility of the transcription factor. Global changes in histone methylation have been associated with tumor recurrence and patient survival in prostate cancer,10,11 non-small-cell lung cancer,12,13 bladder cancer,14 squamous-cell carcinoma of the esophagus,15 and colorectal cancer.16 The frequently observed changes of histone methylation pattern in human cancers may attribute to the deregulation of upstream histone methyltransferases. For instance, our recent study demonstrated that the H3K27 methyltransferase, EZH2, and its associated polycomb repressive complex 2 members (EED, SUZ12, and RBBP7) were substantially up-regulated in human HCC and contributed to HCC metastasis by epigenetic silencing on multiple tumor-suppressive microRNAs (miRNAs). Interestingly, in addition to EZH2, we also observed a common up-regulation of SET-domain containing methyltransferases in primary human HCC, which highlighted the significance of histone methylation in liver carcinogenesis.17

Suppressor of variegation 3-9 homolog 1 (SUV39H1), the mammalian homolog of Drosophila SU(VAR)3-9, is the prototype SET-domain-containing histone methyltransferase. SUV39H1 specifically catalyzes the trimethylation of lysine 9 residue on histone H3 (H3K9me3) and governs global H3K9me3 level. H3K9me3 is a highly conserved repressive histone mark that contributes to heterochromatin formation and therefore indispensable for fundamental cellular processes, including chromosome segregation, mitotic progression, X-chromosome inactivation, and transcriptional silencing. However, the role of SUV39H1 in cancer development remains largely unknown. In this study, we reported a significant up-regulation of SUV39H1 expression in human HCC. Moreover, SUV39H1 level was associated with HCC tumor growth and venous invasion. The oncogenic significance of SUV39H1 on HCC cell proliferation and metastasis was further demonstrated in both in vitro and in vivo experiments. We also demonstrated the negative regulation on SUV39H1 level by microRNA-125b (miR-125b) in HCC. In conclusion, we identified SUV39H1 as an important oncogene in HCC, and aberrant SUV39H1 up-regulation was partly attributed to the underexpression of miR-125b.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Patient Samples and Cell Lines.

Human HCC and the corresponding non-tumorous liver samples were obtained from Chinese patients at Queen Mary Hospital (Pokfulam, Hong Kong). All samples, collected from surgical resection, were snap-frozen in liquid nitrogen and stored at −80°C. Use of human tissues was approved by the institutional review board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster. Human liver cancer cell lines BEL7402, SMMC-7721, MHCC97L, and Huh-7 as well as human immortalized hepatocyte cell line LO2 were used in the present study. BEL7402 and SMMC-7721 were from the Shanghai Institute of Cell Biology (Shanghai, China), MHCC97L was from Fudan University (Dr. Z.Y. Tang, Shanghai, China), and LO2 was from the Shanghai Cancer Institute (Dr. J.R. Gu, Shanghai, China). Huh-7 was from the Hokkaido University School of Medicine (Dr. H. Nakabayashi, Sapporo, Japan).

RNA Extraction and Quantitative Reverse-Transcription Polymerase Chain Reaction.

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, California). One microgram of total RNA was used for complementary DNA synthesis using the GeneAmp RNA PCR Kit (Applied Biosystems, Foster City, CA). SUV39H1 and hypoxanthine-gunaine phosphoribosyltransferase (HPRT) TaqMan probes were ordered from Applied Biosystems. Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) was performed using the 7900HT Fast Real-Time PCR System (Applied Biosystems).

Protein Extraction and Western blotting.

Cells were lysed with 1% NP40 NET buffer for total protein extraction. For subcellular protein fractionation, cells were first lysed with buffer A (10 mM of HEPES [pH 7.9], 10 mM of KCl, 0.1 mM of ethylenediaminetetraacetic acid [EDTA] 0.1 mM of ethylene glycol tetraactic acid [EGTA], 1 mM of dithiothreitol [DTT], and 0.5 mM of phenylmethylsulfonyl fluoride [PMSF]) for 15 minutes, followed by centrifugation at 12,000 rpm for 5 minutes. Cytosolic fraction was collected, and the pellet was lysed with buffer C (20 mM of HEPES [pH 7.9], 0.4 M of NaCl, 1 mM of EDTA, 1 mM of EGTA, 1 mM of DTT, and 1 mM of PMSF) for 15 minutes. Nuclear fraction was collected after centrifugation at 12,000 rpm for 5 minutes. Proteins were resolved in sodium dodecyl sulfate/polyacrylamide gel electrophoresis and blotted onto nitrocellulose membrane. The membrane was incubated with primary antibody (Ab) overnight at 4°C, followed by antimouse or -rabbit immunoglobulin G (GE Healthcare, Waukesha, WI) for 1 hour at room temperature. The ECL detection system (GE Healthcare) was used for protein detection according to the manufacturer's protocol. Anti-SUV39H1 and -topoisomerase I (Topo I) Abs were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-α-tubulin and -β-actin Abs were purchased from Sigma-Aldrich (St. Louis, MO). Anti-H3K9me3 and -panH3 Abs were ordered from Abcam (Cambridge, MA) and Upstate Biotechnology (Lake Placid, NY), respectively.

Clinicopathological Correlation and Statistical Analysis.

Clinicopathological features of HCC patients were analyzed as previously described.18 Statistical analysis was performed using SPSS 19 for Windows (SPSS, Inc., Chicago, IL). Fisher's exact test was used for categorical data, independent t test was used for continuous parametric data, and Mann-Whitney's U test was used for continuous nonparametric data.

Stable Overexpression and Knockdown of SUV39H1 in HCC Cells.

For overexpression study, SUV39H1 expression vector (CMV-[myc]3-SUV39H1) was kindly provided by Prof. Thomas Jenuwein (Max Planck Institute of Immunobiology and Epigenetics, Freiburg, Germany), and the full-length of SUV39H1 coding sequence was further subcloned into pEGFP-C2 expression vector (Clontech Laboratories, Inc., Mountain View, CA). SUV39H1 was transfected into HCC cells using Lipofectamine 2000 (Invitrogen), and stably overexpressing cells were selected using 1 mg/mL of G418. For knockdown study, lentiviral delivery of short hairpin RNAs (shRNAs) targeting SUV39H1 (shSUV-1 and shSUV-2) and nontarget control (NTC) (Sigma-Aldrich) were used. Stably transduced cells were selected by puromycin at 1 µg/mL.

Colony Formation Assay.

HCC cells were seeded at 2 × 105 cells per well onto a six-well plate for transfection or viral transduction. Three days after transfection or viral transduction, 5% of the cells were seeded onto another six-well plate and subjected to G418 or puromycin selection for 2 weeks. Resistant colonies were fixed with 100% methanol and stained with crystal violet.

Cell Proliferation Assay.

HCC cells were seeded at 1 × 104 cells per well into a 12-well plate, and the total cell number of each well was counted three times on days 1, 3, 5, 6, and 7 using the Z1 particle counter (Beckman Coulter, Inc., Brea, CA).

Soft Agar Colony Formation Assay.

The assay was performed in a six-well plate. Fifty HCC cells were resuspended in 2× culture medium and mixed with 1% liquid agar in the ratio of 1:1. The mixture was added on top of the base agar layer (0.5% agar in culture medium). Culture medium was added on top of the solidified agar and was incubated in a CO2 incubator at 37°C for 2 weeks. The total number of colonies formed was counted, and images of colonies were taken under a microscope.

Cell Migration Assay.

Transwell migration assay was performed as previously described.19 Briefly, 1 × 105 cells in serum-free culture medium were added to the upper chamber of the transwell insert (Corning Inc., Corning, NY), whereas the lower chamber was filled with medium containing 10% fetal bovine serum. Cells were allowed to migrate at 37°C for 12 hours. Cells that migrated through the membrane to the lower surface of the transwell were fixed with methanol and stained with crystal violet.

Subcutaneous Xenograft in Nude Mice.

HCC cells (1 × 106) were resuspended in 100 µL of phosphate-buffered saline (PBS) and injected subcutaneously (SC) to the left or right side of 8-week-old male BLB/cAnN nude mice. Tumor size was measured weekly, and tumor volume was calculated with the following formula: 1/2 length × width2. Mice were sacrificed on week 5, and tumors were harvested.

Establishment of Liver Xenograft in Nude Mice.

Firefly-luciferase–labeled MHCC97L cells (MHCC97L-Luc) were established as previously described20 and infected with shSUV39H1 viral particles. MHCC97L-Luc cells (1 × 106) were resuspended in 20 µL of Dulbecco's modified Eagle's medium with high glucose/Matrigel (1:1) and orthotopically injected into the left hepatic lobe of 8-week-old male BLB/cAnN nude mice. For bioluminescent analysis, 100 mg/kg of D-luciferin were injected intraperitoneally into mice 5 minutes before imaging. Mice were sacrificed on week 5, then images of in vivo tumor growth and ex vivo lung and lymph node metastasis were taken using an IVIS 100 Imaging System (Xenogen, Hopkinton, MA). All animal experiments were performed according to the Animals (Control of Experiments) Ordinance (Hong Kong) and the Institute's guidance on animal experimentation.

Senescence-Associated Beta-Galactosidase Activity Assay.

Senescence-associated beta-galactosidase (β-Gal) activity assay was performed according to the previously described protocol.21 Briefly, HCC cells were seeded at 2 × 104 cells per well into a 24-well plate and incubated for 5 days. Cells were washed with PBS twice and fixed in 2% formaldehyde in PBS for 5 minutes. After washing with PBS twice, cells were incubated with staining solution (0.2 M of citric acid/sodium phosphate [pH 6], 0.5 M of K4[Fe(CN)6], 0.5 M of K3[Fe(CN)6], 5 M of NaCl, 1 M of MgCl2, and 1 mg/mL of X-gal) at 37°C for 12-16 hours and washed with PBS twice. Three microscopic images of each well and three wells were counted for each treatment. Percentage of positively stained cell = stained cell/ total number of counted cell × 100%.

Luciferase Reporter Assay.

miR-125b precursor was ordered from Ambion. SUV39H1 3' untranslated region (UTR) sequences, containing wild-type (WT) or mutated miR-125b-binding sites, were cloned into the dual-luciferase miRNA target expression vector, pmirGLO (Promega, Madison, WI). HCC cells (5 × 104) were seeded into each well of a 24-well plate the day before transfection. miR-125b precursor (15 ρmole) was first transfected into BEL7402 cells using X-tremeGene (Roche, Basel, Switzerland). Twenty-four hours later, 0.5 µg of pmirGLO, containing WT or mutated miR-125b-targeted SUV39H1 3' UTR sequence, was transfected into BEL7402 cells using FuGENE 6 (Roche). Firefly and Renilla luciferase activity of transfected cells were determined 48 hours after transfection by using the Dual-Luciferase Assay Kit (Promega), according to the manufacturer's protocol. Renilla luciferase activity was used as the internal control for normalization. Three independent experiments were performed.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

SUV39H1 Was Frequently Up-regulated in Human HCCs and Associated With HCC Proliferation and Venous Invasion.

Epigenetics allows differential gene expression without altering the underlying DNA sequence and is controlled by various epigenetic modifiers. Recently, we analyzed the expression profile of a total of 90 epigenetic regulators in 38 paired human HCC samples.17 We found that the epigenetic regulators′ expression profile could clearly distinguished cancerous tissue from the adjacent non-tumorous liver,17 suggesting that epigenetic alternation is common in HCC development. Interestingly, among the aberrantly expressed epigenetic modifiers, the prototype of SET-domain-containing histone methyltransferase SUV39H1 was one of the most significantly elevated in the primary HCC samples, relative to the non-tumorous liver and normal liver controls (P < 0.001; Fig. 1A). SUV39H1 expression level was also positively associated with proliferation marker Ki67 expression (R = 0.693, P < 0.001; Fig. 1B). This finding suggested that deregulation of SUV39H1 may be implicated in human hepatocarcinogenesis and thus prompted us to further investigate the roles of SUV39H1 in human HCC.

thumbnail image

Figure 1. Up-regulation of SUV39H1 in human HCC. (A) Expression of SUV39H1 mRNA in 4 normal livers and 38 paired primary HCCs was determined by low-density array in an epigenetic profiling study. SUV39H1 expression levels of individual specimen are presented as SUV39H1/HPRT. SUV39H1 was found to be significantly up-regulated, as compared to the corresponding non-tumorous (NT) and normal liver (NL) controls (P < 0.001; Mann-Whitney's U test). (B) SUV39H1 expression was also positively correlated with proliferation marker Ki67 in HCCs and NT livers (R = 0.693, P < 0.001; linear regression). (C) Combining the data from profiling and an expanded paired-primary HCC cohort (N = 67), up-regulation of SUV39H1 was found in 56.2% (59 of 105) of human primary HCCs. Data are presented as the log2 ratio of SUV39H1 mRNA expression in primary HCC, when compared to their corresponding non-tumorous livers. SUV39H1 over- and underexpression cases were defined as log2 (HCC/NT) >1 and <−1, respectively (dotted lines). (D) SUV39H1 overexpression was significantly associated with the presence of venous invasion in human HCC (P = 0.017; Fisher's exact test).

Download figure to PowerPoint

In this study, we first confirmed the up-regulation of SUV39H1 by performing qRT-PCR in an additional 67 paired HCCs and 7 normal liver samples. Combining the data from profiling and validation cohorts, up-regulation of SUV39H1 was frequently found in human primary HCC (59 of 105; 56.2%) (Fig. 1C). Importantly, up-regulation of SUV39H1 was significantly associated with an aggressive HCC pathological feature: the presence of venous invasion in patients' livers (P = 0.017; Fig. 1D; Supporting Tables 1 and 2). These observations highlighted the clinical relevance of SUV39H1 in hepatocarcinogenesis, particularly in the aspect of cancer cell proliferation and metastasis.

Establishment of SUV39H1 Overexpression and Knockdown Models.

To investigate the functional roles of SUV39H1 in HCC, both gain- and loss-of function models were established with references to endogenous SUV39H1 expression levels in HCC cell lines (Supporting Fig. 1). Successful overexpression and knockdown of SUV39H1 was illustrated by western blotting and q-RT-PCR (Fig. 2A, B and Supporting Fig. 2). Consistent with the well-characterized H3K9 trimethylation catalytic function of SUV39H1, we showed that ectopically expressed SUV39H1 was mainly localized in the nucleus and resulted in a substantial increase of global H3K9me3 level (Fig. 2A). In contrast, knockdown of SUV39H1 by lentiviral shRNA significantly decreased H3K9me3 level in HCC cells (Fig. 2B and Supporting Fig. 2C). These experiments demonstrated the successful establishment of SUV39H1 overexpression and knockdown platforms for the later characterization study of SUV39H1 in HCC.

thumbnail image

Figure 2. Establishment of SUV39H1 overexpression and knockdown in vitro models. (A) SUV39H1 overexpression vector was transiently transfected into 293T cells. Protein fractionation of SUV39H1-overexpressing cells showed the predominant nuclear localization of SUV39H1, and SUV39H1 overexpression enhanced total H3K9me3 level. Topo I and α-tubulin were used as the control for nuclear and cytosolic protein fractions, respectively. (B) SUV39H1 knockdown was mediated through lentiviral shRNA. Successful knockdown of SUV39H1 was confirmed at both mRNA and protein levels. Representative data obtained from MHCC97L are shown. Knockdown efficiencies of two independent SUV39H1-targeting shRNA sequences (shSUV-1 and shSUV-2) were 50% and 70%, respectively. Reduced H3K9me3 level was observed in the knockdown cells, as demonstrated in immunoblotting. α-tubulin and pan-H3 were used as loading controls for immunoblotting experiments.

Download figure to PowerPoint

SUV39H1 Contributed to HCC Cell Proliferation In Vitro.

The positive correlation between SUV39H1 and proliferation marker Ki67 expression level suggested the importance of SUV39H1 in HCC cell growth. In line with this observation, we showed that overexpression of SUV39H1 remarkably enhanced HCC cell clonogenicity (Fig. 3A), whereas SUV39H1 knockdown HCC cells reduced colony-forming ability (Fig. 3B), as demonstrated by in vitro clonogenic assay. In addition, knockdown of SUV39H1 significantly decreased cell proliferation and anchorage-independent growth of HCC cells (Fig. 3C, D). Flow cytometry analysis of SUV39H1 knockdown cells showed neither apparent change in cell cycle nor increased cell death; therefore, we excluded the possibility of apoptosis after SUV39H1 knockdown (data not shown). Interestingly, we observed an elevated senescence-associated lysosomal β-Gal activity in SUV39H1 knockdown cells (Fig. 3E), suggesting the potential senescence-protective function of SUV39H1 in cancer progression.

thumbnail image

Figure 3. SUV39H1 contributed to HCC proliferation and clonogenicity in vitro. (A) SUV39H1 overexpression enhanced clonogenicity of HCC cells, whereas (B) SUV39H1 knockdown by shRNA (shSUV-1 and shSUV-2) substantially reduced HCC-cell colony formation, as compared to NTC. Significant reduction on (C) cellular proliferation and (D) anchorage-independent growth was observed in SUV39H1 knockdown HCC cells, as compared to NTC. (E) In the SUV39H1 knockdown HCC cell model, the senescence-associated β-Gal activity was significantly elevated, as compared to NTC, which is consistent with the reduced proliferation after SUV39H1 knockdown. (*P < 0.05; **P < 0.01; ***P < 0.001; t test).

Download figure to PowerPoint

SUV39H1 Contributed to HCC Cell Migration In Vitro.

In addition to cell proliferation, our clinicopathological analysis revealed that SUV39H1 up-regulation in human HCC was significantly associated with the presence of venous invasion, which is a well-established indicator of HCC metastasis. By using SUV39H1 overexpressing and knockdown cell lines, we demonstrated that overexpression of SUV39H1 dramatically enhanced HCC cell migration in transwell migration assay (P < 0.001; Fig. 4A), whereas SUV39H1 knockdown reduced the migratory ability of HCC cells (P < 0.001; Fig. 4B). Consistent findings were obtained from independent stable transfected clones as well as different SUV39H1-targeting shRNA sequences, thus excluding the possibility of clonal bias and off-target effect.

thumbnail image

Figure 4. SUV39H1 contributed to HCC cell migration in vitro. (A) Overexpression of SUV39H1 in immortalized hepatocyte cell line LO2 promoted cell migration. Representative data of two independent vector (GFP #2 and #3) and SUV39H1 overexpressing (SUV39H1 #3 and #4) clones are shown. (B) Knockdown of SUV39H1 reduced the migratory ability of HCC cell line BEL7402 and MHCC97L. Representative pictures of cell migration assay. (***P < 0.001; t test).

Download figure to PowerPoint

SUV39H1 Knockdown Inhibited HCC Tumorigenicity and Metastasis In Vivo.

After exploring the role of SUV39H1 in HCC cell growth and metastasis in vitro, the oncogenic function of SUV39H1 in HCC was further confirmed in vivo by both SC and orthotopic xenograft models. SUV39H1 knockdown and control HCC cells were SC injected into nude mice, and tumor growth was monitored weekly. Consistent with our in vitro data, SUV39H1-knockdown HCC cells showed significantly lower tumorigenicity, as compared to the control (Fig. 5A). Furthermore, we also assessed the effect of SUV39H1 knockdown on HCC tumor growth and metastasis in the native hepatic environment using an orthotopic liver implantation experiment. The shSUV39H1 group showed a significantly reduced tumor size, as compared to the control, as shown by the bioluminescence and gross tumor size of the excised livers (Fig. 5B and Supporting Fig. 3). Most important, SUV39H1 knockdown abolished pulmonary and lymph node metastasis of HCC cells, as shown by the bioluminescence of the excised tissues and histological analysis (Fig. 5C, D). Our in vivo data were consistent with the in vitro data and further strengthened the biological importance of SUV39H1 in liver cancer development.

thumbnail image

Figure 5. SUV39H1 knockdown inhibited HCC tumorigenicity and metastasis in vivo. (A) SUV39H1 knockdown and control BEL7402 cells were injected SC into nude mice. Tumor size was measured weekly, and mice were sacrificed on week 5. SUV39H1 knockdown significantly suppressed HCC tumor growth (*P < 0.05; t test). (B) Orthotopic HCC cell implantation model. Luciferase-labeled MHCC97L cells were injected orthotopically into livers of nude mice. Xenogen images revealed the luciferase activity of liver tumors formed in nude mice. Bioluminescent image (upper panel) and macroscopic pathological examination (lower panel) revealed the reduced tumorigenicity of SUV39H1 knockdown in MHCC97L-Luc cells. Tumors formed in the implanted liver are outlined by dotted lines. (C) Ex vivo bioluminescent imaging (left panel) and histopathological analysis (right panel) showed the effect of SUV39H1 knockdown on suppressing pulmonary metastasis in the orthotopic implantation model. Luciferase signal of lung metastasis was quantified (**P < 0.01; Mann-Whitney's U test). Representative hematoxylin and eosin (H&E) staining revealed the presence of metastasized HCC tumor foci in lungs of mice in the NTC group, but was not found in the shSUV39H1 group (left panel). Arrows point to the tumor foci formed in the lung. Enlarged image of representative fields (magnifications, ×20) are shown (right panel). (D) SUV39H1 knockdown suppressed lymph node metastasis, as evident by ex vivo bioluminescent imaging. Luciferase signal of lymph node metastasis was quantified (*P < 0.05; **P < 0.01; Mann-Whitney's U test). Representative H&E staining revealed the presence of a metastasized HCC cell layer on the lymph node of mice in the NTC group, but was not found in the shSUV39H1 group (left panel). Arrows point to the tumor foci formed in the lymph node. Enlarged image of representative fields (magnifications, ×20) are shown (right panel). The orthotopic HCC cell implantation experiment was been repeated in three independent experiments. For each experiment, 10 mice were used in both the NTC and shSUV39H1 groups.

Download figure to PowerPoint

SUV39H1 Was Negatively Regulated by miR-125b in Human HCC.

Increasing findings demonstrated that dysregulation of miRNA accounts for aberrant gene expression in human cancers. Therefore, we explored the possible relationship between miRNA deregulation and SUV39H1 up-regulation in human HCC. In silico analysis by TargetScan, Pictar, and Miranda miRNA target prediction algorithms revealed that the 3' UTR sequence of SUV39H1 contains putative binding sites for multiple miRNAs (Supporting Fig. 4). Among the predicted miRNAs, miR-125b is the only miRNA that is significantly down-regulated in human HCC (Supporting Fig. 5).22,23 The complementary binding between miR-125b and SUV39H1 3' UTR is kinetically stable and evolutionarily conserved, as illustrated by the diagram of RNA hybrid and sequence alignment among various animal species, respectively (Fig. 6A). Hence, we hypothesized that up-regulation of SUV39H1 in HCC may attribute to the loss of miR-125b. To confirm the binding between miR-125b and SUV39H1 3' UTR, the luciferase reporter assay was performed using the WT or mutated SUV39H1 3'-UTR-coupled luciferase reporter (Fig. 6B). We found that ectopic expression of miR-125b precursor significantly decreased the luciferase signal of WT SUV39H1 3' UTR, as compared to the empty vector control. This suppressive effect was abolished when the putative miR-125b-binding site was mutated (Fig. 6C). Furthermore, miR-125b-overexpressing HCC cells showed a profound reduction of endogenous SUV39H1 expression at both messenger RNA (mRNA) and protein levels (Fig. 6D, E). Consistent results were obtained from BEL7402 and Huh-7 cells, which confirmed the negative regulation of SUV39H1 by miR-125b. Most important, expressions of SUV39H1 and miR-125b were inversely correlated in our clinical HCC and non-tumorous liver samples (R = −0.364, P = 0.001; Fig. 6F). Taken together, our data suggested that SUV39H1 up-regulation is contributed by the underexpression of miR-125b in HCC.

thumbnail image

Figure 6. SUV39H1 was negatively regulated by miR-125b in human HCC. (A) In silico analysis of SUV39H1 3' UTR using RNA hybrid revealed putative binding of miR-125b to SUV39H1 3' UTR. The complementary binding sequence between miR-125b and SUV39H1 3' UTR was found to be highly evolutionarily conserved, as demonstrated by TargetScan bioinformatic algorithm. (B) WT and mutated (MUT) putative miR-125b-binding sequences of SUV39H1 3' UTR were constructed and fused with luciferase reporter. (C) miRNA luciferase reporter assay. WT and mutated SUV39H1 3' UTR were cotransfected with miR-125b precursor into BEL7402 cells. miR-125b significantly repressed luciferase activity of WT/3' UTR, but not in the MUT/3' UTR and empty vector (EV) (***P < 0.001; t test). A fragment of miR-125b complementary sequence (miR-125b sensor) was fused with the luciferase reporter and used as the positive control for the assay. Overexpression of miR-125b precursor significantly attenuated the (D) mRNA and (E) protein levels of endogenous SUV39H1 in Huh-7 cells. (***P < 0.001; t -test). (F) SUV39H1 up-regulation was significantly correlated with miR-125b down-regulation in clinical human HCC and non-tumorous liver tissues (R = −0.364, P = 0.001; linear regression).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Epigenetic dysregulation is one of the most common abnormalities observed in human cancers. Some well-characterized examples of cancer epigenetic changes include DNA hypermethylation on the promoter regions of tumor suppressors and global changes in histone modifications. Histone modifications are known to have profound effects on chromosome stability and gene transcription. Yet, the underlying mechanism of these epigenetic alterations remains largely unknown. Our previous unsupervised clustering analysis revealed two distinct expression patterns of epigenetic regulators between HCC and their corresponding non-tumorous liver, which suggested that deregulation of upstream epigenetic regulators might play a role in cancer epigenetic changes.17 Interestingly, among those differentially expressed epigenetic regulators, a group of SET-domain-containing histone lysine methyltransferases were frequently up-regulated in HCC samples and suggested the significance of histone methylation changes in liver carcinogenesis.

The prototype histone methyltransferase, SUV39H1, responsible for global H3K9 trimethylation, is one of the most significantly up-regulated histone modifiers in HCC. Up-regulation of SUV39H1 mRNA was detected in 56.2% of HCC samples and was significantly associated with increased HCC proliferation and the presence of venous invasion. Consistently, we showed that ectopic expression of SUV39H1 enhanced the colony-forming ability of HCC cells and the migratory ability of the immortalized liver cell line. SUV39H1 knockdown in HCC cells substantially suppressed proliferation and colony formation in both adherent and nonadherent conditions, as well as remarkably reduced HCC cell-migratory ability. The oncogenic property of SUV39H1 was further confirmed in vivo by SC injection and orthotopic implantation models. Knockdown of SUV39H1 dramatically suppressed HCC cell tumorigenicity as well as markedly inhibited pulmonary and lymph node metastasis of HCC cells from orthotopically implanted livers. These findings evidently demonstrated the importance of SUV39H1 in HCC pathogenesis.

In this study, we found that SUV39H1-knockdown HCC cells resembled senescence morphology, along with the enhancement of senescence-associated β-Gal activity. Consistent with our study, knockdown of SUV39H1 substantially inhibited cell growth through telomere shortening and senescence induction in a prostate cancer cell model.24,25 These observations suggested the potential induction of DNA damage response as the consequence of telomere shortening and instability after SUV39H1 knockdown in HCC cells. In colorectal cancer, SUV39H1 mRNA level was significantly elevated and associated with the expression of the DNA methyltransferase, DNA (cytosine-5)-methyltransferase 1 (DNMT1), suggesting a potential collaboration between SUV39H1 and DNMT1 on repressing gene expression.26 Previous reports also showed that SUV39H1 contributed to the transcriptional silencing of tumor-suppressor genes, such as p15 and E-cadherin in acute myeloid leukemia27 and p15 in pancreatic cancer.28 Based on the above-described reports and the function of SUV39H1 on establishing repressive H3K9me3 mark, SUV39H1 up-regulation may be important for telomere maintenance, epigenetic silencing of important tumor-suppressor genes, or senescence evasion during the course of hepatocarcinogenesis.

In addition to histone methylation, miRNA deregulation is also frequently observed in human cancers, including HCC.20,22,23 miRNAs are short noncoding RNAs that negatively regulate the expression of protein-coding genes through translational blockage or promoting degradation of their cognate mRNA targets. Therefore, miRNAs are implicated in many important cellular processes, such as cell-cycle progression, cell differentiation, apoptosis, and cytoskeletal reorganization. Increasing evidences demonstrated the interplay between miRNAs and epigenetic alterations in human cancers. For example, the oncogenic, enhancer of zeste homolog 2 (EZH2), has been found to be overexpressed in various cancer tissues, and EZH2 is targeted by miR-101, miR-124, and miR-214.29-31 Frequent down-regulation of these miRNAs in human cancers thereby accounted for the up-regulation of EZH2. Similar examples have also been reported between the niR-29 family and DNMT3A/B,32 miR-449 and histone deacetylase 1,33 and miR-200c and Bmi-1.34 All these evidences suggested that miRNAs may play a crucial role in modulating epigenetic events. In this study, we explored the possibility of miRNA deregulation as a contributing factor in SUV39H1 expression in human HCC. Interestingly, in silico analysis of SUV39H1 3' UTR suggested the potential regulation of SUV39H1 mRNA by miR-125b. We have previously identified miR-125b as the tumor-suppressor miRNA that is frequently down-regulated in HCC.22 In this study, we experimentally validated the complementary binding between miR-125b and SUV39H1 3' UTR by luciferase reporter assay. Ectopic expression of miR-125b apparently reduced endogenous SUV39H1 mRNA and protein levels in HCC cell lines. In concordance with our findings, a recent study indicated that miR-125b up-regulation may contribute to the increased expression of inflammatory genes in vascular smooth muscle cell (VSMC) of type 2 diabetic db/db mice by targeting SUV39H1.22 Opposite to the VSMCs of db/db mice, miR-125b is frequently down-regulated in human HCC. Interestingly, an inverse correlation was observed between SUV39H1 and miR-125b expression in clinical human HCC samples. Therefore, we speculated that targeting of SUV39H1 by miR-125b may be a conserved event throughout the mammalian cell system, and up-regulation of SUV39H1 in HCC was contributed by the loss of miR-125b.

In conclusion, we provide the first evidence that SUV39H1 is an important oncogene that contributes to HCC tumor growth and metastasis. Besides this, up-regulation of SUV39H1 was, in part, the consequence of tumor-suppressive miRNA-125b underexpression in HCC. This observation further suggested the possible interplay between miRNA and histone methylation during the course of liver carcinogenesis. Our findings have enriched the knowledge of the molecular mechanisms underlying hepatocarcinogenesis and provide potential targets for future therapeutic invention.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The authors thank Ms. Tracy CM Lau from the Faculty Core Facility and Mr. Yeung-Lam Ng from the Department of Pathology, LKS Faculty of Medicine, the University of Hong Kong, for their technical support.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_26083_sm_SuppFig1.tif113KSupporting Information Figure 1.
HEP_26083_sm_SuppFig2.tif389KSupporting Information Figure 2.
HEP_26083_sm_SuppFig3.tif6070KSupporting Information Figure 3.
HEP_26083_sm_SuppFig4.tif182KSupporting Information Figure 4.
HEP_26083_sm_SuppFig5.tif265KSupporting Information Figure 5.
HEP_26083_sm_SuppTab1.doc37KSupporting Information Table 1. Clinicopathological significance of SUV39H1 up-regulation in human HCC.
HEP_26083_sm_SuppTab2.doc29KSupporting Information Table 2. SUV39H1 expression and survival of HCC patients.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.