Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jun 15.
Published in final edited form as: Int J Cancer. 2013 Dec 14;134(12):2778–2788. doi: 10.1002/ijc.28610

Genetic and epigenetic changes in fibrosis-associated hepatocarcinogenesis in mice

Grace Chappell 1, Kristy Kutanzi 2, Takeki Uehara 1, Volodymyr Tryndyak 2, Hue-Hua Hong 3, Mark Hoenerhoff 3, Frederick A Beland 2, Ivan Rusyn 1,*, Igor P Pogribny 2,*
PMCID: PMC4209252  NIHMSID: NIHMS544474  PMID: 24242335

Abstract

Hepatocellular carcinoma (HCC) is one of the most prevalent cancers and is rising in incidence worldwide. The molecular mechanisms leading to the development of HCC are complex and include both genetic and epigenetic events. To determine the relative contribution of these alterations in liver tumorigenesis, we evaluated epigenetic modifications at both global and gene specific levels, as well as the mutational profile of genes commonly altered in liver tumors. A mouse model of fibrosis-associated liver cancer that was designed to emulate cirrhotic liver, a prevailing disease state observed in most humans with HCC, was used. Tumor and non-tumor liver samples from B6C3F1 mice treated with N-nitrosodiethylamine (DEN; a single ip injection of 1 mg/kg at 14 days of age) and carbon tetrachloride (CCl4; 0.2 ml/kg, 2 times/week ip starting at 8 weeks of age for 14 weeks), as well as corresponding vehicle control animals, were analyzed for genetic and epigenetic alterations. H-ras, Ctnnb1, and Hnf1α genes were not mutated in tumors in mice treated with DEN+CCl4. In contrast, the increased tumor incidence in mice treated with DEN+CCl4 was associated with marked epigenetic changes in liver tumors and non-tumor liver tissue, including demethylation of genomic DNA and repetitive elements, a decrease in histone 3 lysine 9 trimethylation (H3K9me3), and promoter hypermethylation and functional down-regulation of Riz1, a histone lysine methyltransferase tumor suppressor gene. Additionally, the reduction in H3K9me3 was accompanied by increased expression of long interspersed nucleotide elements (LINE) 1 and short interspersed nucleotide elements (SINE) B2, which is an indication of genomic instability. In summary, our results suggest that epigenetic events, rather than mutations in known cancer-related genes, play a prominent role in increased incidence of liver tumors in this mouse model of fibrosis-associated liver cancer.

Introduction

Hepatocellular carcinoma (HCC) is one of the most prevalent, life-threatening human cancers1. While the overall cancer incidence and death rates are steadily declining, the incidence of HCC continues to increase1,2. The development and progression of HCC is a multistep process characterized by the progressive, sequential evolution of morphologically distinct pre-neoplastic lesions (formed as a result of chronic liver injury, inflammation, hepatocellular degeneration and necrosis, hepatocellular regeneration and small cell dysplasia, followed by the appearance of low- and high-grade dysplastic nodules), which eventually culminates in the formation of HCC3,4. In humans, 70–90% of HCC cases are associated with advanced liver fibrosis or cirrhosis3. HCC most often arises in the presence of chronic liver inflammation and fibrosis/cirrhosis that may result from disturbances in metabolism, toxic insults, or viral infection5.

While the histopathologic features of HCC is well established, the molecular mechanisms of the cancer-promoting effects of the main etiological factors, including cirrhosis, are not well understood4,6. Elucidating the molecular mechanisms underlying the pathogenesis and progression of HCC is critical for prevention of this disease and development of effective therapies7. Investigation of these mechanisms using human HCC samples is desirable; however, few epidemiological studies have established both the causality and molecular underpinnings of the disease. Animal models that resemble human HCC development may provide important additional clues regarding the molecular sequelae of etiological factors linked to HCC8. A commonly used mouse liver cancer model is a single low dose injection of the genotoxic carcinogen N-nitrosodiethylamine (DEN) into 14-day-old male mice9. In addition, repeat dosing of the pro-fibrogenic agent carbon tetrachloride (CCl4) also results in development of HCC10. To model key pathophysiological events of human cirrhosis-associated hepatocarcinogenesis, we used a combination of genotoxic, e.g., DEN, and non-genotoxic, e.g., CCl4, insults to study the mechanisms of the development of fibrosis-promoted HCC In mice11. The incidence of liver adenomas and carcinomas was more than two-fold greater in mice treated with DEN+CCl4 as compared to that in mice treated with each agent alone11.

The development of HCC is driven by the accumulation of genetic and epigenetic alterations12–14, however, a comparative analysis of the relative contribution of these aberrations is typically not performed in individual studies. The goal of the present study was to investigate the mechanistic roles and contribution of genetic and epigenetic events in fibrosis-associated liver carcinogenesis in mice. We hypothesized that distinct differences in such alterations exist in the liver tissue of animals with fibrosis-associated cancer in comparison to liver tissue of vehicle-control animals. Such alterations may explain, in part, the increased incidence of hepatic tumors in fibrotic liver in this mouse model, and provide insight into the molecular characteristics of human HCC.

Materials and Methods

Animals, treatment, and tissue preparation

The in-life portion of this study, mouse treatments, tissue collection protocols, and the incidence of neoplastic liver lesions are detailed in Uehara et al.11. Briefly, male B6C3F1/J mice were allocated randomly to one control and three experimental groups. At two weeks of age, mice from two of the experimental groups were injected i.p. with DEN (1 mg/kg) in sterile phosphate buffered saline (PBS; 15 ml/kg). Mice from the control group and the remaining experimental group were injected with sterile PBS only. At eight weeks of age, mice from the control and the DEN-treated groups were injected i.p. two times per week with sterile olive oil (15 ml/kg). Mice from the remaining two experimental groups were injected i.p. two times per week with CCl4 (0.2 ml/kg) diluted in sterile olive oil for an additional 14 weeks. In summary, the groups were treated with either PBS+olive oil, DEN+olive oil, PBS+CCl4, or DEN+CCl4. All mice were sacrificed at 22 weeks of age.

Screening for mutations in H-ras and Ctnnb1 genes

Total DNA was isolated from frozen liver tissue samples using DNeasy Mini Kits (Qiagen, Valencia, CA) according to the manufacturer’s instructions. Nested PCR analysis was used to examine codon 61 in exon 2 of the v-Ha-ras Harvey rat sarcoma viral oncogene homolog (H-ras) gene and codons 5–57 in exon 2 of the catenin (cadherin associated protein) beta 1 (Ctnnb1) gene. Reactions were performed on tumor and matched adjacent non-tumor liver tissue samples from mice treated with DEN+CCl4, as well as normal livers from control mice. PCR products were purified, cycled with Terminal Ready Reaction Mix-Big Dye (Perkin Elmer, Foster City, CA), and sequenced.

Immunohistochemical staining for β-catenin activation and Hnf1α inactivation

The activity and cellular location of β-catenin (CTNNB1) was evaluated by immunostaining using goat polyclonal CTNNB1 (sc-1496, 1:50; Santa Cruz Biotechnology, Santa Cruz, CA) as described in Hoenerhoff et al14. The levels of glutamine synthetase (GLUL) and liver fatty acid binding protein 1 (FABP1) was assessed by immunostaining as described in Jeannot et al.15.

Global DNA methylation

The methylation status of genomic DNA was evaluated by a methylation sensitive cytosine extension as described in Pogribny et al.16.

McrBC-methylation sensitive quantitative PCR

The methylation status of long interspersed nucleotide elements (LINE) 1 and short interspersed nucleotide elements (SINE) B2 repetitive sequences was determined by a McrBC-methylation sensitive quantitative PCR assay as described in Martens et al.17.

Methylation-specific PCR

The methylation status of CpG sites located within the promoter/first exon region of cyclin-dependent kinase inhibitor 2A (Cdkn2a), O6-methylguanine-DNA methyltransferase (Mgmt), suppressor of cytokine signaling 1 (Socs1), cadherin 1 (Cdh1), and PR domain containing 2, with ZNF domain (Riz1) was determined by methylation-specific PCR (MSP) as previously described18.

RNA extraction and quantitative real-time reverse transcription-PCR

Total RNA was extracted from liver tissue using TRI Reagent (Ambion, Austin, TX) according to the manufacturer’s instructions. Total RNA (5 µg) was reverse transcribed and gene expression was then determined by quantitative reverse-transcription PCR (qRT-PCR) using gene expression assays (Applied Biosystems, Foster City, CA). Each sample was analyzed in duplicate. Reactions were performed in a 96-well plate format. Each plate contained one experimental gene and the housekeeping gene (Gusb1 or Gapdh). The relative amount of each mRNA transcript was determined using the 2−ΔΔCt method19.

Western blotting

The level of trimethylation of histone 3 lysine 9 (H3K9), histone 3 lysine 27 (H3K27), and histone 4 lysine 20 (H4K20) in the livers of the mice was analyzed by Western blot analysis as described in Tryndyak et al.20.

Chromatin immunoprecipitation (ChIP) assay

Formaldehyde cross-linking and ChIP assay with primary antibodies against H3K9me3 (Millipore Corporation) was performed by using a Chromatin Immunoprecipitation Assay Kit (ChIP) (Millipore Corporation). Purified DNA from immunoprecipitates and input DNA were analyzed by quantitative PCR with primers for the mouse LINE1 and SINE B2 repetitive sequences. The results were normalized to the amount of input DNA and presented as fold change for each DNA in liver of mice from experimental groups relative to control mice.

Statistical analyses

Results are presented as mean ± S.D. Data were analyzed by one-way analysis of variance with pair-wise comparisons being made by the Student-Newman-Keuls method. When necessary, the data were natural log transformed before conducting analysis to maintain a more equal variance or normal data distribution. P-values < 0.05 were considered significant.

Results

Incidence of hepatic preneoplastic and neoplastic lesions

The mouse model of liver tumorigenesis in fibrotic liver was detailed in Uehara et al11. Both neoplastic (adenomas and carcinomas) and pre-neoplastic (foci) lesions were found in mice treated with DEN+olive oil, PBS+CCl4, or DEN+CCl4. Specifically, in the DEN+olive oil-treated group, foci were observed in 95% (19/20) of the mice and adenomas in 10% (2/20). In the PBS+CCl4-treated group, the percentage of mice with foci, adenomas and carcinomas was 12.5% (1/8), 12.5% (1/8) and 25% (2/8), respectively. In the DEN+CCl4-treated group, the percentage of mice with foci, adenomas and carcinomas was 31% (8/26), 100% (26/26) and 50% (13/26), respectively. Hepatocellular adenomas displayed proliferation of relatively uniform hepatocytes accompanied with a loss of normal lobular architecture and compression of the surrounding parenchyma, while the histopathological features of hepatocellular carcinomas consisted of a broad trabecular growth pattern of atypical hepatocytes with hemorrhaging and ischemic necrosis in the center of the tumors (Supplementary Figure 1). Avascular and stromal invasions were also occasionally present in the hepatocellular carcinomas.

Mutational profiling of H-ras, Ctnnb1, and Hnf1α in liver tumors in mice treated with DEN and CCl4

H-ras, Hnf1α, and Ctnnb1 genes were examined for the presence of mutations. Codon 61 of the H-ras gene has been identified as a hot spot for point mutations in both spontaneous and chemically-induced mouse hepatic tumors21. Although mutations in HRAS in human HCC are not common, overexpression of members of the RAS oncogene family, including HRAS, has been reported22. Mutations in Ctnnb1 are a commonly observed event in hepatocarcinogenesis, most often found in the mouse exon 2 and in the corresponding human exon 323. Human hepatocellular adenomas often contain inactivating mutations of the HNF1A gene24; these mutations have been suggested to be a result of exposure to genotoxic agents25.

Mutations in codon 61 of H-ras were rare (1/24, 8.3%) in mice treated with DEN+CCl4. One animal (with 13 adenomas and no carcinomas in its liver) harbored H-ras mutations, and the mutations were observed in both tumors and matched adjacent non-tumor liver tissue. Interestingly, the mutation was different in the tumor tissue (A→T transversion at the second base) and the adjacent non-tumor tissue (C→A transversion at the first base). No significant differences in gene expression of H-ras were observed between vehicle control, tumor, and adjacent non-tumor liver tissues of DEN+CCl4-treated mice (Supplementary Figure 2).

Mutations in exon 2 of Ctnnb1 were not observed in either tumor or adjacent non-tumor liver tissue of mice treated with DEN+CCl4. Consistent with this finding, there were no differences in protein expression or cellular localization of CTNNB1 by immunohistochemistry (Figure 1, left panels). Tumors and adjacent normal tissues from mice treated with DEN+olive oil, PBS+CCl4, or DEN+CCl4 were negative for GLUL (Figure 1, middle panels), confirming the absence of an activating mutation in Ctnnb1. Because adenomas were found in all treatment groups, fatty acid binding protein (L-FABP1), a downstream target of Hnf1α, was used as a marker for inactivating mutations in Hnf1α. Liver tissues from animals in all groups were uniformly stained for FABP1, indicating a normal functioning Hnf1α (Figure 1, right panels).

Figure 1.

Figure 1

Open in a new tab

Immunostaining of CTNNB1, GLUL, and FABP1 in formalin-fixed paraffin-embedded sections of liver tumor (T) and adjacent non-tumor (N) liver tissue. Left panel: Liver tumors (pictured in all groups except vehicle control) lack immunoreactivity or nuclear accumulation of CTNNB1. Minimal cytoplasmic accumulation is present in centrilobular hepatocytes in PBS+CCl4-treated mice, primarily in non-tumor fibrotic areas and likely artifact from degeneration. Middle panel: Liver tumors (pictured in all groups except vehicle control) lack immunoreactivity to GLUL. Positive immunoreactivity is restricted to pericentral hepatocytes in non-tumor tissue. Pericentral immunoreactivity is weak in PBS+CCl4-treated mice, likely due to degeneration and loss of these hepatocytes. The positive staining in the PBS+CCl4 representative image is artifact due to necrosis. Right panel: Uniform positive immunoreactivity is present in both tumor (pictured in all groups except vehicle control) and adjacent non-tumor tissue. Original magnification: 40×.

DNA methylation changes in liver tumors in mice treated with DEN and CCl4

It is well-established that both mouse and human liver tumors exhibit substantial alterations in DNA methylation26. Hence, the status of genomic and gene-specific DNA methylation in liver tumors induced by DEN+CCl4 treatment was examined. Figure 2A shows that DNA was markedly hypomethylated both globally and within LINE1 and SINE B2 repetitive elements in tumors as compared to liver tissue from vehicle control mice. It is well-established that the extent of methylation of these repetitive elements strongly corresponds to the level of global DNA methylation27.

Figure 2.

Figure 2

Open in a new tab

DNA methylation and gene expression changes in liver tumors. (A) Genomic, LINE1, and SINE B2 methylation. The results are presented as an average percent change in the degree of DNA hypomethylation in liver tumors of mice treated DEN+CCl4 relative to that in the PBS (vehicle control) group, which were assigned a value of 100%. (B) Extent of Cdkn2a, Mgmt, Socs1, Cdh1 and Riz1 promoter methylation. The Y-axis represents the percentage of samples that had methylation in the promoter region of the described gene. (C) The expression of Cdkn2a, Mgmt, Socs1, Cdh,1 and Riz1 was determined by qRT-PCR as detailed in “Materials and Methods.” The results are presented as an average fold change in the expression of each gene in liver tumors of mice treated DEN+CCl4 relative to expression in liver tissues of the control group, which were assigned a value of 1.

Values shown are mean±S.D., asterisks (*) denote a significant (p<0.05) difference from control mice.

Figure 2B shows that five tumor suppressor genes, Cdkn2a, Mgmt, Socs1, Cdh1, and Riz1, all known to be commonly epigenetically altered in HCC26,28, were heavily methylated in liver tumors in DEN+CCl4-treated mice, while these genes were largely unmethylated in the livers of control mice. One of the main cancer-associated epigenetic features is the silencing of aberrantly methylated genes29. Hence, the expression of these five genes was also evaluated. Figure 2C shows that among the analyzed genes, only the expression of Riz1 and Mgmt was significantly decreased in association with promoter hypermethylation in liver tumors from mice treated with DEN+CCl4. Changes in the expression of Cdkn2a, Socs1, and Cdh1 did not exhibit such an association with the incidence of promoter methylation.

DNA methylation changes in fibrosis-associated mouse hepatocarcinogenesis

To determine the role of epigenetic alterations in the progression of mouse fibrosis-associated liver carcinogenesis, the status of genomic DNA methylation in non-tumor liver tissues was assessed. Figure 3A shows that DNA isolated from non-tumor liver tissue of mice treated either with PBS+CCl4 or with DEN+CCl4 was also markedly hypomethylated. Interestingly, the extent of methylation of global DNA as well as of LINE1 and SINE B2 repetitive DNA sequences in non-tumorous liver tissue from mice treated with PBS+CCl4 or with DEN+CCl4 was similar to that of liver tumors in mice treated with DEN+CCl4 (Figure 3A). In contrast to the results obtained in mice treated with PBS+CCl4 or DEN+CCl4, the level of DNA methylation in the livers of mice treated with DEN+olive oil was not different from that of control mice.

Figure 3.

Figure 3

Open in a new tab

DNA methylation and gene expression changes in non-tumor liver tissues during fibrosis-associated hepatocarcinogenesis in mice. (A) Genomic, LINE1, and SINE B2 methylation. The results are presented as an average percent change in the degree of DNA hypomethylation in non-tumorous liver tissues of mice from experimental groups relative to that in control mice, which was assigned a value of 100%. (B) Extent of Cdkn2a, Mgmt, Socs1, Cdh1 and Riz1 promoter methylation. The Y-axis represents the percentage of samples that had methylation in the promoter region of the described gene. (C) The expression of Cdkn2a, Mgmt, Socs1, Cdh1, and Riz1 genes. The results are presented as an average fold change in the expression of each gene in the livers of mice from experimental groups relative to that in control mice. Values shown are mean±S.D., asterisks (*) denote a significant (p<0.05) difference from control mice.

To further evaluate the role of DNA methylation changes in fibrosis-associated mouse liver carcinogenesis, promoter methylation and expression of Cdkn2a, Mgmt, Socs1, Cdh1, and Riz1 in non-tumor liver tissue were examined. Figure 3B shows that these tumor suppressor genes were methylated in the non-tumorous liver samples in DEN+CCl4-treated mice, while they were largely unmethylated in the livers of control mice. However, the extent of Mgmt, Socs1, and Riz1 methylation in non-tumor tissue was noticeably lower than in liver tumors. Interestingly, in PBS+CCl4-treated mice, Cdkn2a, Socs1, Cdh1, and Riz1 were methylated in ≥ 40% of the non-tumorous liver samples. In DEN+olive oil-treated mice, only Socs1 was methylated in ≥ 40% of adjacent non-tumor liver samples.

Figure 3C shows that among all of the analyzed genes, only the expression of Mgmt and Riz1 was significantly decreased in association with promoter hypermethylation in the non-tumor liver tissues of mice treated with DEN+CCl4. Changes in the expression of Cdkn2a, Socs1, and Cdh1 did not exhibit such an association with the incidence of promoter methylation. The expression of these genes was similar in non-tumor and tumor tissue from the livers of mice treated with DEN+CCl4 (Figures 2C and 3C).

Histone methylation during fibrosis-associated mouse hepatocarcinogenesis

It has been previously reported that a reduction in trimethylation of H3K9, H3K27, and H4K20 affects genomic stability30,31, and is an event observed in various human cancers. Figure 4 shows a large decrease in H3K9 trimethylation in the livers of mice treated with DEN+CCl4. The levels of H3K9 trimethylation in this experimental group were decreased by 58% compared with control mice. In contrast to the H3K9 methylation changes, the levels of H3K27 and H4K20 were decreased only slightly in the livers of DEN+olive oil-, PBS+CCl4-, and DEN+CCl4-treated mice.

Figure 4.

Figure 4

Open in a new tab

H3K9, H3K279, and H4K20 trimethylation in liver tissues during fibrosis-associated liver carcinogenesis in mice. Densitometry analysis of the immunostaining results is shown as percent change in histone modification level in the each experimental group relative to the corresponding values in control mice. Values shown are mean±S.D., asterisks (*) denote a significant (p<0.05) difference from control mice.

Expression of chromatin-modifying genes during fibrosis-associated hepatocarcinogenesis

To determine the underlying mechanisms of these epigenetic aberrations, the expression of chromatin modifying genes involved in the proper maintenance of DNA and histone methylation was assessed by qRT-PCR. The most noticeable changes that occurred during fibrosis-associated carcinogenesis were a distinct up-regulation of the maintenance DNA methyltransferase 1 (Dnmt1) and de novo DNA methyltransferase 3a (Dnmt3a) genes in the livers of mice treated with PBS+CCl4 and DEN+CCl4. Down-regulation of histone lysine (K)-specific demethylase (Kdm4a and Kdm4b) genes was observed in liver tissue of mice treated with DEN+CCl4 (Figure 5).

Figure 5.

Figure 5

Open in a new tab

Expression of chromatin-modifying genes in the liver during fibrosis-associated hepatocarcinogenesis in mice. The expression of Dnmt1, Dnmt3a, Dnmt3b, Suv39h1, Ezh2, Suv4-30h2, Kdm4a, and Kdm4b genes was determined by qRT-PCR. The results are presented as an average fold change in the expression of each gene in the livers of mice from each experimental group relative to that in control mice, which were assigned a value 1. Values shown are mean±S.D., asterisks (*) denote a significant (p<0.05) difference from control mice.

Histone H3K9 trimethylation and expression of LINE1 and SINE B2

To investigate whether or not epigenetic changes observed in this study may be mechanistically related to the progression of fibrosis-associated carcinogenesis, the status of H3K9 methylation at LINE1 and SINE B2 repetitive sequences and their expression were analyzed. Figure 6A shows a substantial decrease of H3K9 trimethylation (H3K9me3) at LINE1 and SINE B2 repetitive sequences in liver tissue of mice treated with DEN+CCl4. The reduction of LINE1 and SINE B2 H3K9me3 was accompanied by a marked increase in the expression of LINE1 and, especially, SINE B2 elements (Figure 6B).

Figure 6.

Figure 6

Open in a new tab

Level of H3K9me3 at LINE1 and SINE B2 repetitive sequences and expression of LINE1 and SINE B2 in the liver during fibrosis-associated liver carcinogenesis in mice. (A) The level of H3K9me3 at LINE1 and SINE B2 determined by a ChIP assay as described in “Materials and Methods.” The data are presented as fold change in liver of mice from experimental groups relative to control mice after normalization to input DNA. (B) Expression of LINE1 and SINE B2 elements. The results are presented as an average fold change in the expression of LINE1 and SINE B2 in the livers of mice from experimental groups relative to that in control mice, which were assigned a value of 1. Values shown are mean±S.D., asterisks (*) denote a significant (p<0.05) difference from control mice.

Expression of progenitor markers in liver tumors

In this mouse model of hepatocarcinogenesis, markers of oncofetal liver transformation and cancer stem cells were found to be significantly associated with tumor incidence; whereas inflammation, fibrogenesis, oxidative stress, proliferation and apoptosis were not indicative of increased tumorigenesis11. Supplementary Figure 3 shows that the expression of alpha-fetoprotein (Afp) was significantly up-regulated in tumor tissues of mice treated with DEN+CCl4 when compared to control animals, a pattern similar to that reported for the non-tumor tissues from the same treatment group11. In contrast, epithelial cell adhesion molecule (Epcam) was down-regulated in tumor tissues when compared to control mice, although the change was not statistically significant. Interestingly, Epcam was significantly up-regulated in non-tumor tissues of the DEN+CCltretaed mice when compared to all other treatment groups in the original study11. These findings corresponded to the hypothesis that this mitogenic signaling molecule is involved in tumor promotion and progression, but is less active in late stage tumors.

Discussion

Hepatocellular carcinogenesis is a complex process that is the consequence of multiple molecular events that lead to the initiation, promotion, and progression of tumor cells3,4,6,7. Recent reviews have emphasized the importance of a distinct set of events that are required for carcinogenesis32,33. Specifically, distinct cellular capabilities that enable tumorigenesis, or “hallmarks of cancer,” are increasingly recognized as essential processes in carcinogenesis32. Mutations in cancer-related genes result in aberrant cellular functions, which can be characterized as specific hallmark capabilities. While genetic alterations have long been known to cause the development and progression of tumors, epigenetic events have emerged more recently as potentially equally important events in the process of hepatocarcinogenesis13,26.

To better understand the role of genetic and epigenetic alterations in liver tumor development, the mutational profile, expression, and epigenetic alterations of several cancer-related genes commonly involved in hepatocarcinogenesis were evaluated using a mouse model of fibrosis-associated hepatocellular carcinoma. Additionally, epigenetic events associated with genomic instability, such as global DNA hypomethylation and histone demethylation, were investigated.

Common mutations do not play a primary role in fibrosis-associated HCC in murine liver tumors

We evaluated the mutational profile of genes commonly implicated in hepatocarcinogenesis of both mice and humans. Alterations in the Wnt/β-catenin pathway are frequently involved in both human and mouse HCC, with CTNNB1 reported to be the most commonly activated oncogene in human HCC23. Activating mutations in CTNNB1 lead to constitutive activation of the CTNNB1 protein and subsequent up-regulation of Wnt signaling,34 the consequence of which is stimulation of cell proliferation and inhibition of apoptosis35. Recent studies have reported contrasting results regarding the association of CTNNB1 mutations and genomic instability in hepatic tumors33,36. However, a general consensus has been established that genomic instability is an enabling characteristic in hepatic and other cancers32. In our study, while epigenetic alterations that are associated with genomic instability were observed, mutations in the hot-spot region (mouse exon 2) of Ctnnb1 were absent. These findings suggest that causative mechanisms of genomic instability may precede Ctnnb1 mutation, and support the theory that genomic instability and Ctnnb1 mutation are distinct mechanisms of hepatocarcingenesis6,35.

Characterizing the frequency and type of mutations in tumor-related genes is informative in the determination of tumor induction and progression, especially in laboratory animals, because the basis of carcinogenic effects may be due to either chemical induction or spontaneous occurrence37. For example, single base substitution mutations at codon 61 of H-ras are one of the most commonly seen mutations in spontaneously occurring liver tumors in B6C3F1 mice21. Mutations at this hotspot are also seen in chemically-induced neoplasms in mice, although there is evidence that the type and frequency of mutations can be chemical- and dose-dependent, with a lower incidence of codon 61 mutations associated with increasing doses of multiple chemicals14, 38, including DEN21. Indeed, the low frequency of H-ras mutations in our study suggests that CCl4 and DEN together preferentially promote cells through mechanisms independent of H-ras mutation.

Epigenetic alterations are an early and important event in fibrosis-associated mouse hepatocarcinogenesis

Epigenetic changes such as DNA hypomethylation at the global level and within repetitive sequences, promoter hypermethylation of tumor suppressor genes, and altered histone lysine methylation are common characteristics in many, if not all, types of cancer29,39,40. In our study, the extent of global DNA hypomethylation was significantly greater in the livers of PBS+CCl4- and DEN+CCl4-treated mice than in DEN+olive oil-treated or control mice. Similarly, demethylation of LINE1 and SINE 2B sequences was found in PBS+CCl4- and DEN+CCl4-treated mice, although the extent of LINE1 demethylation in the livers of PBS+CCl4-treated mice was not statistically significant. Furthermore, we observed a significant decrease in global H3K9me3 in the non-tumor liver tissue in mice treated with DEN+CCl4. Additionally, the degree of DNA methylation and H3K9me3 was significantly decreased in LINE1 and SINE B2 sequences in the livers of DEN+CCl4-treated mice, and the expression of LINE1 and SINE B2 was significantly increased. These interspersed repetitive sequences, which represent nearly half of the human genome and approximately one-third of the mouse genome,41 play an important role in the development of several human cancers42.

Notably, the PBS+CCl4- and DEN+CCl4-treated mice had a higher incidence of liver tumors (37.5% and 100%, respectively) than mice treated with DEN+olive oil (10%), as reported by Uehara, et al. in 2013. While the incidence of liver preneoplastic foci was nearly 100% in DEN+olive oil-treated mice, few of these lesions (2/20) progressed to adenomas. This suggests that the significant increase in tumor incidence in livers of mice treated with DEN+CCl4, as compared to the other treatment groups, may be attributed to a significant hypomethylation of genomic DNA, LINE1 and SINE B2 repetitive sequences, and the decrease in global, LINE1 and SINE B2 H3K9me3. Importantly, these epigenetic alterations appeared to precede genetic aberrations commonly seen in mouse and human HCC.

RIZ1, a member of a histone/protein methyltransferase superfamily, is associated with tumor suppression function43 and is frequently inactivated in HCC44,45. It has been demonstrated that the histone methyltransferase function of RIZ1 is an important constituent of its tumor suppressor activity46. Our results indicate that inhibition of Riz1 by promoter hypermethylation is associated with a substantial decrease in H3K9me3, which may be the result of lost RIZ1 histone methyltransferase function. A significant decrease in Riz1 expression was found in both tumor and non-tumor liver tissues of mice treated with DEN+CCl4, as well as in the livers of mice treated with PBS+CCl4 alone. Because mice in these treatment groups had a higher tumor incidence than DEN+olive oil-treated mice, we can infer that Riz1 plays a major role in tumorigenesis in our study. The level of H3K9me3 was significantly decreased only in mice treated with DEN+CCl4, suggesting a closely link between H3K9me3 and tumor cell promotion and/or progression. This also suggests that mechanisms additional to the loss of Riz1 histone methyltransferase function play a role in H3K9 methylation.

RIZ1 promoter hypermethylation has been observed in both early and advanced stages of human HCC, with a higher incidence in early stage tumors47. This supports the potential tumor-promoting role of Riz1 promoter hypermethylation found in the present study, and also confers the mechanistic human relevance of this mouse HCC model. Additionally, RIZ1 methylation has been associated with a shorter disease-free survival time in human HCC patients45. RIZ1 hypermethylation has been found to be prevalent in various human cancer types, particularly breast and liver, two types in which mutations in the gene have not been found43. These findings indicate that this gene is likely preferably silenced by methylation as opposed to by mutation43. Furthermore, there are no known cis-acting regulators of RIZ1 expression48, which supports the conclusion that methylation is the cause for the down-regulation of the gene in many cancers.

Not all epigenetic events may be involved in carcinogenesis

We found that certain early epigenetic events appear to occur without effect on gene expression or be associated with tumor promotion. For example, although promoter hypermethylation was observed in four (Cdkn2a, Mgmt, Socs1, and Cdh1) of the cancer-related genes examined in this study, the expression of these genes was not down-regulated. This is contrary to the well accepted mechanistic link between promoter hypermethylation and gene silencing27. In fact, Cdkn2a and Cdh1 were up-regulated in the livers of PBS+CCl4- and DEN+CCl4-treated mice a finding consistent with the hypothesis that not all DNA methylation changes, similar to genetic mutations, are equally important in the process of carcinogenesis, and that some DNA methylation changes may be purely passenger events49.

Conclusions

Our results indicate that epigenetic alterations are an essential and early event in hepatocellular carcinogenesis, especially under conditions of liver fibrosis. Based on previous reports, genomic instability is likely a consequence of loss of DNA methylation globally and within repetitive elements and decreased H3K9 methylation, all of which were associated with increased tumor incidence in the present study. Furthermore, loss of Riz1 expression by promoter hypermethylation was associated with increased tumor incidence, and therefore the tumor suppressor function of this gene is likely linked to its inherent histone methyltransferase activity. The epigenetic alterations observed in fibrosis-associated liver tumors in this mouse model indicate important features involved in the development of human liver tumors that arise from fibrosis and cirrhosis, a common progression according to human clinico-pathological evidence. Additionally, these epigenetic alterations may confer a heightened risk for genomic instability, although additional studies are required to confirm such an association. Our results indicate that H-ras, Hnf1α, and Ctnnb1 mutations are not involved in the initiation or promotion of the hepatic tumors in this study, and that epigenetic changes preceded or occurred independently of traditionally observed genetic alterations in mouse HCC. While common HCC-related mutations were not detected in our study, the epigenetic alterations observed may contribute to genomic instability and thus represent one of the proposed hallmarks of cancer32.

Supplementary Material

Supp Fig S1-S3

Novelty and impact statement.

This report demonstrates that epigenetic modifications, rather than mutations in known cancer-related genes, are essential and early events in mouse hepatocellular carcinogenesis that play a prominent role in the increased incidence of liver tumors in fibrosis-associated liver cancer. The epigenetic alterations observed in a mouse model of fibrosis-associated liver tumors may indicate important features of tumor development in human liver tumors that arise from fibrosis and cirrhosis.

Acknowledgements

This work was supported, in part, by the National Institutes of Health grants R01 ES015241, P42 ES005948, and R01 ES023195.

Footnotes

The views expressed in this paper do not necessarily represent those of the U.S. Food and Drug Administration.

References

  • 1.Center MM, Jemal A. International trends in liver cancer incidence rates. Cancer Epidemiol Biomarkers Prev. 2011;20:2362–2368. doi: 10.1158/1055-9965.EPI-11-0643. [DOI] [PubMed] [Google Scholar]
  • 2.American Association for Cancer Research. AACR Cancer Progress Report 2013. Clin Cancer Res. 2013;19:S1–S86. [Google Scholar]
  • 3.Farazi PA, DePinho RA. Hepatocellular carcinoma pathogenesis: from genes to environment. Nat Rev Cancer. 2006;6:674–687. doi: 10.1038/nrc1934. [DOI] [PubMed] [Google Scholar]
  • 4.Aravalli RN, Cressman EN, Steer CJ. Cellular and molecular mechanisms of hepatocellular carcinoma: an update. Arch Toxicol. 2013;87:227–247. doi: 10.1007/s00204-012-0931-2. [DOI] [PubMed] [Google Scholar]
  • 5.Fattovich G, Stroffolini T, Zagni I, Donato F. Hepatocellular carcinoma in cirrhosis: incidence and risk factors. Gastroenterology. 2004;127:S35–S50. doi: 10.1053/j.gastro.2004.09.014. [DOI] [PubMed] [Google Scholar]
  • 6.Zucman-Rossi J. Molecular classification of hepatocellular carcinoma. Dig Liver Dis. 2010;42(Suppl 3):S235–S241. doi: 10.1016/S1590-8658(10)60511-7. [DOI] [PubMed] [Google Scholar]
  • 7.Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet. 2012;379:1245–1255. doi: 10.1016/S0140-6736(11)61347-0. [DOI] [PubMed] [Google Scholar]
  • 8.Heindryckx F, Colle I, Van Vlierberghe H. Experimental mouse models for hepatocellular carcinoma research. Int J Exp Pathol. 2009;90:367–386. doi: 10.1111/j.1365-2613.2009.00656.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Vesselinovitch SD, Mihailovich N. Kinetics of diethylnitrosamine hepatocarcinogenesis in the infant mouse. Cancer Res. 1983;43:4253–4259. [PubMed] [Google Scholar]
  • 10.Fujii T, Fuchs BC, Yamada S, Lauwers GY, Kulu Y, Goodwin JM, Lanuti M, Tanabe KK. Mouse model of carbon tetrachloride induced liver fibrosis: Histopathological changes and expression of CD133 and epidermal growth factor. BMC Gastroenterol. 2010;10:79. doi: 10.1186/1471-230X-10-79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Uehara T, Ainslie GR, Kutanzi K, Pogribny IP, Muskhelishvili L, Izawa T, Yamate J, Kosyk O, Shymonyak S, Bradford BU, Boorman GA, Bataller R, et al. Molecular mechanisms of fibrosis-associated promotion of liver carcinogenesis. Toxicol Sci. 2013;132:53–63. doi: 10.1093/toxsci/kfs342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Nault JC, Zucman-Rossi J. Genetics of hepatobiliary carcinogenesis. Semin Liver Dis. 2011;31:173–187. doi: 10.1055/s-0031-1276646. [DOI] [PubMed] [Google Scholar]
  • 13.Pogribny IP, Rusyn I. Role of epigenetic aberrations in the development and progression of human hepatocellular carcinoma. Cancer Lett. 2012 doi: 10.1016/j.canlet.2012.01.038. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Hoenerhoff MJ, Pandiri AR, Snyder SA, Hong HH, Ton TV, Peddada S, Shockley K, Witt K, Chan P, Rider C, Kooistra L, Nyska A, et al. Hepatocellular Carcinomas in B6C3F1 Mice Treated with Ginkgo biloba Extract for Two Years Differ from Spontaneous Liver Tumors in Cancer Gene Mutations and Genomic Pathways. Toxicol Pathol. 2013;41:826–841. doi: 10.1177/0192623312467520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Jeannot E, Pogribny IP, Beland FA, Rusyn I. Chronic administration of ethanol leads to an increased incidence of hepatocellular adenoma by promoting H-ras-mutated cells. Cancer Lett. 2011;301:161–167. doi: 10.1016/j.canlet.2010.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Pogribny I, Yi P, James SJ. A sensitive new method for rapid detection of abnormal methylation patterns in global DNA and within CpG islands. Biochem Biophys Res Commun. 1999;262:624–628. doi: 10.1006/bbrc.1999.1187. [DOI] [PubMed] [Google Scholar]
  • 17.Martens JH, O'Sullivan RJ, Braunschweig U, Opravil S, Radolf M, Steinlein P, Jenuwein T. The profile of repeat-associated histone lysine methylation states in the mouse epigenome. EMBO J. 2005;24:800–812. doi: 10.1038/sj.emboj.7600545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci USA. 1996;93:9821–9826. doi: 10.1073/pnas.93.18.9821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc. 2008;3:1101–1108. doi: 10.1038/nprot.2008.73. [DOI] [PubMed] [Google Scholar]
  • 20.Tryndyak VP, Muskhelishvili L, Kovalchuk O, Rodriguez-Juarez R, Montgomery B, Churchwell MI, Ross SA, Beland FA, Pogribny IP. Effect of long-term tamoxifen exposure on genotoxic and epigenetic changes in rat liver: implications for tamoxifen-induced hepatocarcinogenesis. Carcinogenesis. 2006;27:1713–1720. doi: 10.1093/carcin/bgl050. [DOI] [PubMed] [Google Scholar]
  • 21.Maronpot RR, Fox T, Malarkey DE, Goldsworthy TL. Mutations in the ras proto-oncogene: clues to etiology and molecular pathogenesis of mouse liver tumors. Toxicology. 1995;101:125–156. doi: 10.1016/0300-483x(95)03112-s. [DOI] [PubMed] [Google Scholar]
  • 22.Tang Z, Qin L, Wang X, Zhou G, Liao Y, Weng Y, Jiang X, Lin Z, Liu K, Ye S. Alterations of oncogenes, tumor suppressor genes and growth factors in hepatocellular carcinoma: with relation to tumor size and invasiveness. Chin Medical J (Engl) 1998;111:313–318. [PubMed] [Google Scholar]
  • 23.de La Coste A, Romagnolo B, Billuart P, Renard CA, Buendia MA, Soubrane O, Fabre M, Chelly J, Beldjord C, Kahn A, Perret C. Somatic mutations of the beta-catenin gene are frequent in mouse and human hepatocellular carcinomas. Proc Natl Acad Sci USA. 1998;95:8847–8851. doi: 10.1073/pnas.95.15.8847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bluteau O, Jeannot E, Bioulac-Sage P, Marques JM, Blanc JF, Bui H, Beaudoin JC, Franco D, Balabaud C, Laurent-Puig P, Zucman-Rossi J. Bi-allelic inactivation of TCF1 in hepatic adenomas. Nat Genet. 2002;32:312–315. doi: 10.1038/ng1001. [DOI] [PubMed] [Google Scholar]
  • 25.Jeannot E, Mellottee L, Bioulac-Sage P, Balabaud C, Scoazec JY, Tran VN, Bacq Y, Michalak S, Buob D, Laurent-Puig P, Rusyn I, Zucman-Rossi J. Spectrum of HNF1A somatic mutations in hepatocellular adenoma differs from that in patients with MODY3 and suggests genotoxic damage. Diabetes. 2010;59:1836–1844. doi: 10.2337/db09-1819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Tischoff I, Tannapfe A. DNA methylation in hepatocellular carcinoma. World J Gastroenterol. 2008;14:1741–1748. doi: 10.3748/wjg.14.1741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Chalitchagorn K, Shuangshoti S, Hourpai N, Kongruttanachok N, Tangkijvanich P, Thong-ngam D, Voravud N, Sriuranpong V, Mutirangura A. Distinctive pattern of LINE-1 methylation level in normal tissues and the association with carcinogenesis. Oncogene. 2004;23:8841–8846. doi: 10.1038/sj.onc.1208137. [DOI] [PubMed] [Google Scholar]
  • 28.Nishida N, Nagasaka T, Nishimura T, Ikai I, Boland CR, Goel A. Aberrant methylation of multiple tumor suppressor genes in aging liver, chronic hepatitis, and hepatocellular carcinoma. Hepatology. 2008;47:908–918. doi: 10.1002/hep.22110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Jones PA, Baylin SB. The epigenomics of cancer. Cell. 2007;128:683–692. doi: 10.1016/j.cell.2007.01.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kouzarides T. Chromatin modifications and their function. Cell. 2007;128:693–705. doi: 10.1016/j.cell.2007.02.005. [DOI] [PubMed] [Google Scholar]
  • 31.Jorgensen S, Schotta G, Sorensen CS. Histone H4 Lysine 20 methylation: key player in epigenetic regulation of genomic integrity. Nucleic Acids Res. 2013;41:2797–2806. doi: 10.1093/nar/gkt012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–674. doi: 10.1016/j.cell.2011.02.013. [DOI] [PubMed] [Google Scholar]
  • 33.Laurent-Puig P, Legoix P, Bluteau O, Belghiti J, Franco D, Binot F, Monges G, Thomas G, Bioulac-Sage P, Zucman-Rossi J. Genetic alterations associated with hepatocellular carcinomas define distinct pathways of hepatocarcinogenesis. Gastroenterology. 2001;120:1763–1773. doi: 10.1053/gast.2001.24798. [DOI] [PubMed] [Google Scholar]
  • 34.Devereux TR, Anna CH, Foley JF, White CM, Sills RC, Barrett JC. Mutation of beta-catenin is an early event in chemically induced mouse hepatocellular carcinogenesis. Oncogene. 1999;18:4726–4733. doi: 10.1038/sj.onc.1202858. [DOI] [PubMed] [Google Scholar]
  • 35.Morin PJ, Sparks AB, Korinek V, Barker N, Clevers H, Vogelstein B, Kinzler KW. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science. 1997;275:1787–1790. doi: 10.1126/science.275.5307.1787. [DOI] [PubMed] [Google Scholar]
  • 36.Aleksic K, Lackner C, Geigl JB, Schwarz M, Auer M, Ulz P, Fischer M, Trajanoski Z, Otte M, Speicher MR. Evolution of genomic instability in diethylnitrosamine-induced hepatocarcinogenesis in mice. Hepatology. 2011;53:895–904. doi: 10.1002/hep.24133. [DOI] [PubMed] [Google Scholar]
  • 37.Sills RC, Boorman GA, Neal JE, Hong HL, Devereux TR. Mutations in ras genes in experimental tumours of rodents. In: McGregor DB, Rice JM, Venitt S, editors. The use of short- and medium-term tests for carcinogens and data on genetic effects in carcinogenic hazard evaluation. Vol. 146. Lyon, France: International Agency for Research on Cancer; 1999. pp. 55–86. [Google Scholar]
  • 38.Devereux TR, White CM, Sills RC, Bucher JR, Maronpot RR, Anderson MW. Low frequency of H-ras mutations in hepatocellular adenomas and carcinomas and in hepatoblastomas from B6C3F1 mice exposed to oxazepam in the diet. Carcinogenesis. 1994;15:1083–1087. doi: 10.1093/carcin/15.5.1083. [DOI] [PubMed] [Google Scholar]
  • 39.Kitkumthorn N, Mutirangura A. Long interspersed nuclear element-1 hypomethylation in cancer: biology and clinical applications. Clin Epigenetics. 2011;2:315–330. doi: 10.1007/s13148-011-0032-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Feinberg AP, Ohlsson R, Henikoff S. The epigenetic progenitor origin of human cancer. Nat Rev Genet. 2006;7:21–33. doi: 10.1038/nrg1748. [DOI] [PubMed] [Google Scholar]
  • 41.Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, An P, Antonarakis SE, Attwood J, et al. Initial sequencing and comparative analysis of the mouse genome. Nature. 2002;420:520–562. doi: 10.1038/nature01262. [DOI] [PubMed] [Google Scholar]
  • 42.Rudin CM, Thompson CB. Transcriptional activation of short interspersed elements by DNA-damaging agents. Genes Chromosomes Cancer. 2001;30:64–71. [PubMed] [Google Scholar]
  • 43.Du Y, Carling T, Fang W, Piao Z, Sheu JC, Huang S. Hypermethylation in human cancers of the RIZ1 tumor suppressor gene, a member of a histone/protein methyltransferase superfamily. Cancer Res. 2001;61:8094–8099. [PubMed] [Google Scholar]
  • 44.Formeister EJ, Tsuchiya M, Fujii H, Shpyleva S, Pogribny IP, Rusyn I. Comparative analysis of promoter methylation and gene expression endpoints between tumorous and non-tumorous tissues from HCV-positive patients with hepatocellular carcinoma. Mutation Res. 2010;692:26–33. doi: 10.1016/j.mrfmmm.2010.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Zhang C, Li H, Wang Y, Liu W, Zhang Q, Zhang T, Zhang X, Han B, Zhou G. Epigenetic inactivation of the tumor suppressor gene RIZ1 in hepatocellular carcinoma involves both DNA methylation and histone modifications. J Hepatol. 2010;53:889–895. doi: 10.1016/j.jhep.2010.05.012. [DOI] [PubMed] [Google Scholar]
  • 46.Kim KC, Geng L, Huang S. Inactivation of a histone methyltransferase by mutations in human cancers. Cancer Res. 2003;63:7619–7623. [PubMed] [Google Scholar]
  • 47.Piao GH, Piao WH, He Y, Zhang HH, Wang GQ, Piao Z. Hyper-methylation of RIZ1 tumor suppressor gene is involved in the early tumorigenesis of hepatocellular carcinoma. Histol Histopathol. 2008;23:1171–1175. doi: 10.14670/HH-23.1171. [DOI] [PubMed] [Google Scholar]
  • 48.Gatti D, Maki A, Chesler EJ, Kirova R, Kosyk O, Lu L, Manly KF, Williams RW, Perkins A, Langston MA, Threadgill DW, Rusyn I. Genome-level analysis of genetic regulation of liver gene expression networks. Hepatology. 2007;46:548–557. doi: 10.1002/hep.21682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Kalari S, Pfeifer GP. Identification of driver and passenger DNA methylation in cancer by epigenomic analysis. Adv Genet. 2010;70:277–308. doi: 10.1016/B978-0-12-380866-0.60010-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Fig S1-S3
NIHMS544474-supplement-Supp_Fig_S1-S3.docx (389KB, docx)

RESOURCES