Role of DNA damage and alterations in cytosine DNA methylation in rat liver carcinogenesis induced by a methyl-deficient diet

Mutation Research 669 (2009) 56–62

Contents lists available at ScienceDirect

Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis journal homepage: www.elsevier.com/locate/molmut Community address: www.elsevier.com/locate/mutres

Role of DNA damage and alterations in cytosine DNA methylation in rat liver carcinogenesis induced by a methyl-deficient diet夽 Igor P. Pogribny a,∗ , Svitlana I. Shpyleva a,b , Levan Muskhelishvili c , Tetyana V. Bagnyukova a , S. Jill James d , Frederick A. Beland a a

Division of Biochemical Toxicology, National Center for Toxicological Research, Jefferson, AR 72079, USA Department of Mechanisms of Anticancer Therapy, R.E. Kavetsky Institute of Experimental Pathology, Oncology and Radiobiology, Kyiv 03022, Ukraine Toxicologic Pathology Associates, National Center for Toxicological Research, Jefferson, AR 72079, USA d Department of Pediatrics, University of Arkansas for Medical Sciences, Little Rock, AR 72205, USA b c

a r t i c l e

i n f o

Article history: Received 3 March 2009 Received in revised form 27 April 2009 Accepted 4 May 2009 Available online 12 May 2009 Keywords: Oxidative DNA damage DNA methylation Liver carcinogenesis Methyl-deficiency

a b s t r a c t Currently, cancer is recognized as a disease provoked by both genetic and epigenetic events. However, the significance of early genetic and epigenetic alterations with respect to carcinogenic process in general and to liver carcinogenesis in particular remains unexplored. A lack of knowledge regarding how specific alterations during early preneoplasia may be mechanistically related to tumor formation creates a major gap in understanding the role of these genetic and epigenetic abnormalities in carcinogenesis. In the present study we investigated the contribution of DNA damage and epigenetic alterations to liver carcinogenesis induced by a methyl-deficient diet. Feeding Fisher 344 rats a methyl-deficient diet for 9 weeks resulted in DNA damage and aberrant DNA methylation. This was evidenced by an early upregulation of the base excision DNA repair genes, accumulation of 8-oxodeoxyguanosine and 3 OH-end strand breaks in DNA, pronounced global loss of DNA methylation, and hypermethylation of CpG islands in the livers of methyl-deficient rats. These abnormalities were completely restored in the livers of rats exposed to methyl-deficiency for 9 weeks after removal of the methyl-deficient diet and re-feeding a methyl-sufficient diet. However, when rats were fed a methyl-deficient diet for 18 week and then given a methyl-sufficient diet, only DNA lesions were repaired. The methyl-sufficient diet failed to restore completely the altered DNA methylation status and prevent the progression of liver carcinogenesis. These results suggest that stable alterations in DNA methylation are a factor that promotes the progression of liver carcinogenesis. Additionally, the results indicate that epigenetic changes may be more reliable markers than DNA lesions of the carcinogenic process and carcinogen exposure. Published by Elsevier B.V.

1. Introduction Classically, cancer has been viewed as a set of diseases initiated and driven by permanent heritable genetic aberrations in critical cancer genes [1,2] caused by endogenous and environmental agents [3–5]. Therefore, the general approach for elucidating key events during cancer development has been focused on identification of genetic lesions associated with carcinogenesis [6–8]. However, even within the classical genotoxic carcinogenesis models, the presence of genetic lesions per se is not sufficient for tumor formation, which results from much broader alterations in cellular homeostasis, mainly from the inability of cells to maintain and

夽 Note: The views expressed in this paper do not necessarily represent those of the U.S. Food and Drug Administration. ∗ Corresponding author. Tel.: +1 870 543 7096; fax: +1 870 543 7720. E-mail address: [email protected] (I.P. Pogribny). 0027-5107/$ – see front matter Published by Elsevier B.V. doi:10.1016/j.mrfmmm.2009.05.003

control accurately the expression of genetic information. Additionally, genetic alterations alone cannot explain the extremely diverse phenotypic changes observed in preneoplastic and malignant cells. This has led to the suggestion that the transition from promotion, a reversible stage of carcinogenesis, to progression, an irreversible stage, may involve and be driven primarily by epigenetic abnormalities [9]. Currently, it is evident that the induction of these epigenetic abnormalities is a major event in the pathogenesis of cancer [10–12]. In our previous studies, using several models of rodent liver carcinogenesis, we demonstrated that non-genotoxic (methyldeficient diet and the peroxisome proliferator WY-14,643) and genotoxic (tamoxifen and 2-acetylaminofluorene) carcinogens induce genetic and epigenetic changes in livers [13–16]. However, the significance of early carcinogen-induced genetic and epigenetic alterations with respect to carcinogenic process in general and to liver carcinogenesis in particular remains unexplored [5]. Specifically, a major gap in understanding the role of these genetic and

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

epigenetic abnormalities in carcinogenesis is a lack of knowledge regarding how specific alterations during early preneoplasia may be mechanistically related to tumor formation. One of the most extensively studied models of liver cancer is methyl-deficient liver carcinogenesis in rats. This model is unique because dietary omission of sources of methyl groups rather than xenobiotic addition leads to tumor formation [17,18]. In addition, the sequence of pathological and molecular events is remarkably similar to the induction of human hepatocellular carcinomas by viral hepatitis B and C infections, alcohol exposure, and metabolic liver diseases [19]. It is believed that underlying mechanisms of liver carcinogenesis induced by a methyl-deficient diet are associated with early oxidative damage to DNA [17,19] and alterations in the cellular epigenome [14,20,21]. However, the mechanistic basis for these alterations in liver carcinogenesis and, especially, whether or not these alterations can be considered as neoplastic are largely unknown. In light of these considerations, we conducted experiments to define the contribution of DNA damage and epigenetic alterations to liver carcinogenesis. 2. Materials and methods

57

The status of glutathione-S-transferase placental form (GSTP) expression in the livers was determined after feeding rats the methyl-deficient diet and after refeeding the methyl-sufficient diet. Formalin-fixed paraffin-embedded liver sections were deparaffinized and re-hydrated. Endogenous peroxidase was inhibited by incubating with freshly prepared 3% hydrogen peroxide containing 0.1% sodium azide for 10 min at room temperature. Nonspecific staining was blocked with normal goat 10% normal goat serum (Sigma, St. Louis, MO) for 20 min at room temperature. The sections were then incubated with rabbit anti-human GSTP antibody (DAKO, Carpinteria, CA) at the dilution of 1:100 (10 ␮g/ml) for 1 h at room temperature. After incubation with primary antibody, tissue sections were incubated with biotinylated goat anti-rabbit IgG (ExtrAvidin Kit, Sigma) at a dilution of 1:30 for 30 min at room temperature, and then with streptavidin-conjugated horseradish peroxidase (ExtrAvidin Kit, Sigma) at the dilution of 1:30 for 30 min at room temperature. Staining was developed with diaminobenzidine (Sigma) for 5 min at room temperature, and sections were counterstained with hematoxylin and mounted with Permount (Fisher Scientific, Pittsburgh, PA). All sections were examined by light microscopy (BX40 Olympus, Tokyo, Japan). 2.7. Image analysis GSTP-stained rat liver sections were scanned and digital images were obtained with Aperio Scanscope System (Aperio Technologies, Vista, CA). The proportion of the area that stained positive for GST-P was calculated in the digital images with Positive Pixel Count Algorithm (Aperio Technologies), particularly designed for evaluation of a proportion of immunohistochemically stained area (for details see www.aperio.com).

2.1. Animals, diets, and tissue preparations Male weaning Fisher 344 (F344) rats were obtained from the National Center for Toxicological Research (NCTR) breeding facility, housed 2 per cage in a temperaturecontrolled (24 ◦ C) room, with a 12 h light/dark cycle, and given ad libitum access to water and NIH-31 pelleted diet (Purina Mills, Richmond, IN). At 4 weeks of age, the rats (body weight 50 g) were allocated randomly to receive either a methyl-deficient diet containing low methionine (0.18%) and lacking choline and folic acid (118753 Lombardi choline deficient diet without added folic acid; Dyets, Inc., Bethlechem, PA), or a methyl-sufficient diet (118754 Lombardi choline supplemented diet; Dyets, Inc.) that was supplemented with 2 mg/kg folic acid. After being fed the methyldeficient diet for either 9 or 18 weeks, the rats were given the methyl-sufficient diet. Four rats from each of the methyl-deficient diet groups and four rats from the control group were sacrificed at 9, 18, 36, or 60 weeks after diet initiation. The livers were excised and a portion of the medial lobe was fixed in 10% neutral buffered formalin for 48 h, embedded in paraffin, sectioned at 4 ␮m, and mounted on glass slide for immunohistochemical evaluation. The remaining liver was frozen immediately in liquid nitrogen and stored at −80 ◦ C for subsequent analyses. All animal experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee at the NCTR.

2.8. Statistical analysis Results are presented as mean ± S.D. Statistical analyses were conducted by twoway ANOVA, using treatment and weeks as fixed factors. Pair-wise comparisons were conducted by the Student–Newman–Keuls test. P-values < 0.05 were considered significant.

3. Results 3.1. Effect of the methyl-deficient diet on damage of hepatic DNA

2.4. Determination of global DNA methylation status

Feeding a methyl-deficient diet resulted in progressive accumulation of 8-oxodG (Fig. 1A) and 3 OH-end strand breaks (Fig. 1B) in hepatic DNA with difference being significant after 18 weeks of deficiency. It has been suggested that expression of BER genes is a sensitive in vivo biomarker of persistent oxidative DNA damage [30,31]. In view of this, we assessed the expression of these DNA repair genes in the livers of rats fed the methyl-deficient diet for 9 and 18 weeks. Fig. 1C shows a significant increase in expression of apurinic/apyrimidinic endonuclease 1 (Ape), poly(ADP-ribose) polymerase 1 (Parp), and DNA polymerase beta (Polˇ) genes in the livers of methyl-deficient rats. Specifically, after 9 and 18 weeks, expression of Parp and Ape was approximately 2 times greater than in the age-matched control groups, while Polˇ 1.5 times greater compared with their age-matched controls. The smaller increase in the expression of Polˇ, encoding a rate-determining enzyme in BER [32,33], compared to the expression of Parp and Ape, is consistent with the presence of DNA strand breaks found in this (Fig. 1B) and previous studies using the similar model of liver carcinogenesis [18,34]. Additionally, feeding the methyl-deficient diet enhanced expression of O6 -methylguanine-DNA methyl transferase (Mgmt) gene that encodes an enzyme involved in the direct repair of alkylated guanine residues [35]. Similar findings have been reported previously [19].

The extent of global DNA methylation was evaluated with a radiolabeled [3 H]dCTP extension assay as described in Pogribny et al. [24] and with a SssI methylacceptance assay [25].

3.2. Effect of the methyl-deficient diet on the status of DNA methylation

2.2. Determination of DNA damage Genomic DNA was isolated from liver tissues by standard digestion with proteinase K, followed by phenol-chloroform extraction and dialysis against 10 mM Tris–HCl, 1 mM EDTA buffer, pH 7.6, instead of ethanol precipitation [22]. The levels of 8-oxodeoxyguanosine (8-oxodG) were assessed by an OxiSelect Oxidative DNA damage ELISA Kit from Cell Biolabs, Inc. (San Diego, CA) according to the manufacturer’s protocol. The 3’OH-end DNA breaks were quantified by the modified random oligonucleotide primed synthesis (ROPS) assay as described previously [23]. 2.3. Base excision DNA repair gene expression Total RNA (n = 3 from each diet group) was extracted from liver tissues using TRI Reagent (Ambion, Austin, TX) according to the manufacturer’s instructions. The expression of base excision DNA repair (BER) genes was determined an RNase protection assay using rat multi-nucleotide RNA probe template sets (BD Biosciences, San Jose, CA) and a RiboQuant multi-probe RNase protection assay kit (BD Biosciences). Protected fragments were separated on polyacrylamide gels and the intensity of protected fragments was analyzed using a Cyclone Storage Phosphor Screen (Packard Instrument, Meriden, CT).

2.5. Determination of gene-specific methylation by methylation-specific PCR The methylation status of the promoter CpG island of four tumor suppressor genes, Rassf1a, p16INK4A , Socs1, and Cdh1, was determined by methylation-specific PCR (MSP) analysis [26]. The primer sequences and MSP conditions were reported previously in detail [27–29].Immunohistochemistry

Feeding the methyl-deficient diet resulted in a rapid hypomethylation of hepatic DNA (Table 1). After 9 weeks on the diet, the incorporation of [3 H]dCTP into HpaII-digested DNA, which is directly proportional to the number of unmethylated CCGG sites, was 1.6 times greater than in the age-matched control

58

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

Fig. 1. DNA damage and DNA repair in rat livers induced by a methyl-deficient diet. (A) Accumulation of 8-oxodG in hepatic DNA in rats fed a methyl-deficient diet. Genomic DNA was isolated from the livers of control rats and rats fed a methyl-deficient die for 9 or 18 weeks and the presence of 8-oxodG was determined as described in Section 2. Data presented as mean ± S.D. (n = 4). (B) Presence of 3 OH-end DNA strand breaks in the livers of rats fed a methyl-deficient dietas detected by ROPS assay. Data are presented as percent change relative to age-matched control rats. (C) Expression of DNA base excision repair genes in the livers of control and methyl-deficient rats. Total RNA was isolated from the livers of rats fed a methyl-deficient diet and age-matched control rats (n = 3). An RNase protection assay was performed as described in Section 2. The relative expression of each gene was normalized to the expression of the L32 housekeeping gene. *Significantly different from control (p < 0.05).

rats. After 18 weeks of deficiency the extent of DNA hypomethylation increased slightly and was 1.8 times greater compared to the corresponding control rats (Table 1). These changes in global DNA methylation were confirmed independently by using an SssI methyl-acceptance assay (Table 1). We also determined the regional effect of methyl-deficiency on the extent of cytosine methylation at CpG islands of functional genes. For this purpose, hepatic DNA from control and methyldeficient mice after 9 and 18 weeks of deficiency was digested with NotI methylation-sensitive restriction endonuclease, the recognition sequence of which occurs almost exclusively in CpG islands of functional genes [36]. In the livers of rats fed the methyl-deficient diet, the extent of cytosine methylation at CpG islands did not change over the 9 week period, after 18 weeks of methyl deficiency, there was significant hypermethylation as indicated by decreased [32 P]dGTP incorporation into NotI-digested DNA (Table 1). To confirm the induction of promoter hypermethylation of functional

genes, we analyzed the methylation status of several critical genes, including Rassf1a, p16INK4A , Socs1, and Cdh1, genes frequently hypermethylated during hepatocarcinogenesis [37]. Table 1 shows that feeding the methyl-deficient diet resulted in hypermethylation of the Rassf1a gene as detected by MSP assay. This finding is consistent with earlier reports demonstrating high frequency of the Rassf1a gene methylation at early stages of hepatocarcinogenesis [38]. The methylation status of the p16INK4A , Socs1, and Cdh1 genes, three other genes often hypermethylated in hepatocellar carcinoma, did not change at these early preneoplastic hepatocarcinogenesis stages (data not shown). 3.3. Status of DNA damage and DNA methylation in the livers of methyl-deficient rats after re-feeding a methyl-sufficient diet In order to evaluate the significance of early DNA damage and DNA methylation abnormalities in hepatocarcinogenesis, we

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

59

Table 1 Status of cytosine DNA methylation in the livers of control rats and rats fed a methyl-deficient diet. Value

Group

9 weeks

18 weeks

Global DNA methylationa , [3 H]dCTP incorporation, cpm/␮gDNA

Control Methyl-deficient diet Control Methyl-deficient diet Control Methyl-deficient diet Control Methyl-deficient diet

3129 ± 416 5068 ± 492* 6220 ± 840 9516 ± 1022* 8651 ± 1210 8078 ± 1450 Unmethylated (n = 4) Unmethylated (n = 2) Methylated (n = 2)

3406 ± 536 6232 ± 738* 5986 ± 764 11320 ± 1780* 7930 ± 1344 4318 ± 786* Unmethylated (n = 4) Methylated (n = 4)

Global DNA methylationb , [3 H-methyl] incorporation, cpm/␮gDNA CpG islands methylationc , [32 P]dGTP incorporation, cpm/␮gDNA RASSF1A methylationd

a The DNA methylation status in the livers of control and methyl-deficient rats was assessed using a [3 H]dCTP methylation sensitive extension assay after digestion of genomic DNA with methylation-sensitive restriction endonuclease HpaII. The extent of [3 H]dCTP incorporation is directly proportional to the number of unmethylated CCGG sites. b The DNA methylation status was assessed by an SssI methyl-acceptance assay. c The status of CpG islands methylation in the livers of control and methyl-deficient rats was assessed using a methylation sensitive extension assay with [32 P]dGTP after digestion of genomic DNA with methylation sensitive restriction endonuclease NotI, which cleaves unmethylated GCGGCCGC sequences to produce overhanging 5 cytosine residues that can be used for the subsequent labeling with [32 P]dGTP. Data are presented as mean ± S.D. for each group (n = 4) and asterisks (*) indicate significant difference from age-matched control rats. d The status of the RASSF1A promoter methylation in the livers of control and methyl-deficient rats was assessed using a MSP assay.

Fig. 2. Levels of 8-oxodG and 3’OH-end strand breaks in hepatic DNA of methyl-deficient rats re-fed a methyl-sufficient diet. (A) Diagram of the feeding protocol. Male F344 rats were maintained on a methyl-deficient diet for 9 and 18 weeks followed by re-feeding a methyl-sufficient diet. (B, C) Levels of 8-oxodG and 3 OH-end strand breaks in DNA in the livers of F344 maintained on a methyl-deficient diet for 9 weeks (B) or 18 weeks (C) followed by re-feeding a methyl-sufficient diet. Four rats per diet group and control group were sacrificed at 9 and 18 weeks of methyl-deficiency and at specific time intervals of re-feeding, and at the end of the experiment (at 60 weeks of the study). The results presented as percent change in the levels of 8-oxodG and 3 OH-end strand breaks in hepatic DNA relative to age-matched control rats. *Significantly different from control (p < 0.05).

60

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

Rassf1a gene remained in the hypermethylated state despite feeding the methyl-sufficient diet to rats exposed to methyl deficiency for 18 weeks (Fig. 3C). 3.4. Effect of re-feeding the methyl-supplemented diet on evolution of preneoplastic GSTP-positive foci Immunohistochemical staining of liver sections after 9 weeks on the methyl-deficient diet revealed only single GSTP-positive hepatocytes or isolated minifoci. After 18 weeks of methyl-deficiency, the livers consisted of small GSTP-positive foci evenly distributed throughout the entire section of the liver (Table 2). In rats maintained on the methyl-deficient diet for 9 weeks and then given a methyl-sufficient diet, only isolated GSTP-positive minifoci were detected in the livers (Table 2). In contrast, in rats exposed to the methyl-deficient diet for 18 weeks, size of GSTP-positive foci and the proportion of area of the livers that stained positive for GSTP substantially were increased despite the re-feeding the methyl-adequate diet (Table 2, Fig. 4). No GSTP-positive foci or single-positive hepatocytes were detected in the livers of agematched control rats. 4. Discussion

Fig. 3. DNA methylation changes in the livers of methyl-deficient rats re-fed a methyl-sufficient diet. (A, B) The status of global DNA methylation in the livers of F344 maintained on a methyl-deficient diet for 9 weeks (A) or 18 weeks (B) followed by re-feeding a methyl-sufficient diet. Four rats per diet group and control group were sacrificed at 9 and 18 weeks of methyl-deficiency and at specific time intervals of re-feeding, and at the end of the experiment (at 60 weeks of the study). The results presented as percent change relative to age-matched control rats. * Significantly different from control (p < 0.05). (C) The status of the Rassf1a gene methylation in the livers of rats exposed to methyl deficiency for 18 weeks and after re-feeding a methyl-sufficient diet until the end of the experiment at 60 weeks.

determined the evolution of these alterations in the livers of methyl-deficient rats returned to a methyl-sufficient diet (Fig. 2). When rats were returned to a methyl-adequate diet after being fed a methyl-deficient diet for 18 weeks, there was a diminution of the DNA damaging effects caused by a methyl-deficient diet. This was evidenced by reduction of the 8-oxodG levels and 3 OH-end strand breaks in the hepatic DNA of methyl-deficient rats to normal values (Fig. 2C) and normalization of expression of BER genes (data not shown). Feeding the methyl-sufficient diet to rats exposed to methyldeficiency for 9 weeks resulted in the normalization of the DNA methylation status (Fig. 3A). In contrast, in animals maintained on the methyl-deficient diet for 18 weeks, the DNA remained hypomethylated to the end of the experiment despite feeding the methyl-sufficient diet (Fig. 3B). Likewise, the promoter of the

Liver carcinogenesis induced by methyl-deficiency is characterized by two distinctive stages: transition from normal liver to a state of chronic liver injury (reversible stage), followed by a second transition from a state of chronic injury to formation of liver tumors (irreversible stage) [19,39]. The results of our previous studies, as well as results of studies conducted by other investigators, indicated that damage to DNA and changes in the cellular epigenome, including loss of DNA methylation and alterations of histone modifications, are the earliest changes observed in livers during hepatocarcinogenesis induced by a methyl-deficient diet [17–21]. These molecular changes are attributed to the induction of first transitional stage of liver carcinogenesis; however, whether or not the changes may be considered as neoplastic is not clear. In this report we demonstrate that feeding rats a methyldeficient diet results in induction of DNA damage and alterations in DNA methylation in the livers. This was evidenced by an early upregulation of Parp, Ape, and Polˇ genes involved in the BER pathway, accumulation of 8-oxodG and 3 OH-end strand breaks in DNA, pronounced global loss of DNA methylation, and hypermethylation of CpG islands in the livers of methyl-deficient rats. Similar effects of a methyl-deficient diet have been reported previously [19,39,40]. However, in the present study by using re-feeding phase, in which methyl-deficient rats were returned to a methyl-sufficient diet, we were able to monitor the contribution of these changes to carcinogenic process. It is well-known that for any carcinogen-induced alteration found at preneoplastic stages of carcinogenesis to be considered as neoplastic, it needs to be stable and persist after the factor that induced the changes has been removed. To establish a mechanistic link between any early molecular change and the carcinogenic

Table 2 GSTP-positive lesions in the livers of methyl-deficient rats before and after re-feeding a methyl-sufficient diet. Group

Control Treated

Area of GSTP-positive lesions (% from total area)a 9 weeks methyl-deficient diet

9 weeks methyl-deficient diet + 51 weeks methyl-sufficient diet

18 weeks methyl-deficient diet

18 weeks methyl-deficient diet + 42 weeks methyl-sufficient diet

ND 0.72 ± 0.10

ND 0.59 ± 0.15

ND 1.78 ± 0.20

ND 3.04 ± 0.22*

a The proportion of the area stained positive for GSTP was evaluated with Positive Pixel Count Algrorithm (Aperio Technologies). Data are presented as mean ± S.D. for each group (n = 4). ND, not detected. * Significantly different from 18 weeks methyl-deficient rats.

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

61

Fig. 4. Representative immunohistochemical demonstration of GSTP-positive foci in the livers of methyl-deficient rats re-fed a methyl-sufficient diet.

process, a well-defined and sensitive phenotypic marker of carcinogenesis is required. It is well-established that formation of foci of altered hepatocytes, characterized by up-regulation of the GSTP in rat liver, is a sensitive phenotypic indicator of hepatocarcinogenesis [41,42]. The results of several studies have demonstrated the involvement of oxidative stress in generation of GSTP-positive hepatocytes during the initiation stage of hepatocarcinogenesis [18,19,43,44]. However, the formation of initiated GSTP-positive cells by themselves is not sufficient to be indicative of carcinogenic process and does not lead directly to neoplasia. Specifically, GSTPpositive hepatocytes can occur spontaneously and only around 1% of these initiated cells expand clonally into altered hepatic foci [9]. Only the formation and progression of GSTP-positive foci are considered to be a sufficient indicator of liver carcinogenesis [9,41,42]. The results of our study demonstrate that alterations in DNA methylation but not DNA damage contribute to the formation and progression of GSTP-positive foci during liver carcinogenesis as evidenced by persistence of DNA methylation changes after removal of the methyl-deficient diet. The exposure of rats to the carcinogenic a methyl-deficient diet for 9 weeks caused temporal changes in DNA damage and DNA methylation, which returned to normal state after removal of the methyl-deficient diet and feeding of the methylsufficient diet. In contrast, feeding the methyl-sufficient diet to rats maintained on the methyl-deficient diet for 18 weeks restored the integrity of DNA only by repairing DNA lesions, but failed to restore completely the altered DNA methylation status that is associated with the progression of liver carcinogenesis. While these results do not contradict with the previous findings showing the importance of DNA damage and epigenetic alterations as early initial alterations in tumorigenic process [18,19,45–48], our study signifies the role of stable alterations in DNA methylation as factor that contributes to the progression of liver carcinogenesis. The appearance of DNA methylation abnormalities during the first transitional stage of hepatocarcinogenesis and their persistence during the second irreversible stage, even in the absence of the methyl-deficient diet, a factor that induced those changes, suggest that epigenetic abnormalities steer progression of histopathological changes toward tumor formation. These results are consistent with similar findings, showing that DNA methylation changes, rather than DNA damage are potential facilitator of carcinogenic process [49]. Additionally, these results establish that stable epigenetic changes are more reliable marker of methyl-deficient carcinogenesis than the presence of DNA lesions. Conflict of interest None. References [1] D. Hanahan, R.A. Weinberg, The hallmarks of cancer, Cell 100 (2000) 57–70.

[2] B. Vogelstein, K.W. Kinzler, Cancer genes and the pathways they control, Nat. Med. 10 (2004) 789–799. [3] L.A. Loeb, Endogenous carcinogenesis: molecular oncology into the twenty-first century—presidential address, Cancer Res. 49 (1989) 5489–5496. [4] A. Luch, Nature and nurture—lessons from chemical carcinogenesis, Nat. Rev. Cancer 5 (2005) 113–125. [5] L.A. Loeb, C.C. Harris, Advances in chemical carcinogenesis: a historical review and prospective, Cancer Res. 68 (2008) 6863–6872. [6] M.C. Poirier, Chemical-induced DNA damage and human cancer risk, Nat. Rev. Cancer 4 (2004) 630–637. [7] D.H. Phillips, DNA adducts as markers of exposure and risk, Mutation Res. 577 (2005) 284–292. [8] A. Rundle, Carcinogen–DNA adducts as biomarker for cancer risk, Mutation Res. 600 (2006) 23–36. [9] H.C. Pitot, Adventures in hepatocarcinogenesis, Annu. Rev. Pathol. 2 (2007) 1–29. [10] L.A. Jaffe, Epigenetic theories of cancer initiation, Adv. Cancer Res. 30 (2003) 209–230. [11] A.P. Feinberg, The epigenetics of cancer etiology, Semin. Cancer Biol. 14 (2004) 427–432. [12] P.A. Jones, S.B. Baylin, The epigenomics of cancer, Cell 128 (2007) 683–692. [13] V.P. Tryndyak, L. Muskhelishvili, O. Kovalchuk, R. Rodriguez-Juarez, B. Montgomery, M.I. Churchwell, S.A. Ross, F.A. Beland, I.P. Pogribny, Effect of long-term tamoxifen exposure on genotoxic and epigenetic changes in rat liver: implications for tamoxifen-induced hepatocarcinogenesis, Carcinogenesis 27 (2006) 1713–1720. [14] I.P. Pogribny, I. Rusyn, F.A. Beland, Epigenetic aspects of genotoxic and nongenotoxic hepatocarcinogenesis: studies in rodents, Environ. Mol. Mutagen. 49 (2008) 9–15. [15] T.V. Bagnyukova, V.P. Tryndyak, B. Montgomery, M.I. Churchwell, A.R. Karpf, S.R. James, L. Muskhelishvili, F.A. Beland, I.P. Pogribny, Genetic and epigenetic changes in rat preneoplastic liver tissue induced by 2-acetylaminofluorene, Carcinogenesis 29 (2008) 638–646. [16] I.P. Pogribny, V.P. Tryndyak, A. Boureiko, S. Melnyk, T.V. Bagnyukova, B. Montgomery, I. Rusyn, Mechanisms of peroxisome proliferator-induced DNA hypomethylation in rat liver, Mutation Res. 644 (2008) 17–23. [17] D. Nakae, Endogenous liver carcinogenesis in the rat, Pathol. Int. 49 (1999) 1028–1042. [18] S.J. James, I.P. Pogribny, M. Pogribna, B.J. Miller, S. Jernigan, S. Melnyk, Mechanisms of DNA damage, DNA hypomethylation, and tumor progression in the folate/methyl-deficient rat model of hepatocarcinogenesis, J. Nutr. 133 (2003) 3740S–3747S. [19] C.L. Powell, O. Kosyk, B.U. Bradford, J.S. Parker, E.K. Lobenhofer, A. Denda, F. Uematsu, D. Nakae, I. Rusyn, Temporal correlation of pathology and DNA damage with gene expression in a choline-deficient model of rat liver injury, Hepatology 42 (2005) 1137–1147. [20] E. Wainfan, M. Dizik, M. Stender, J.K. Christman, Rapid appearance of hypomethylated DNA in livers of rats fed cancer-promoting, methyl-deficient diets, Cancer Res. 49 (1989) 4094–4097. [21] E. Wainfan, L.A. Poirier, Methyl groups in carcinogenesis: effects on DNA methylation and gene expression, Cancer Res. 52 (1992) 2071s–2077s. [22] W.M. Strauss, Preparation of genomic DNA from mammalian tissue, in: F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Struhl (Eds.), Current Protocols in Molecular Biology, Wiley–Interscience, New York, 1989, pp. 2.2.1–2.2.3. [23] A.G. Basnakian, S.J. James, Quantitation of 3 OH DNA breaks by random oligonicleotide-primed synthesis (ROPS) assay, DNA Cell Biol. 15 (1996) 255–262. [24] I. Pogribny, P. Yi, S.J. James, A sensitive new method for rapid detection of abnormal methylation patterns in global DNA and within CpG islands, Biochem. Biophys. Res. Commun. 262 (1999) 624–628. [25] M. Balaghi, C. Wagner, DNA methylation in folate deficiency: use of CpG methylase, Biochem. Biophys. Res. Commun. 193 (1993) 1184–1190. [26] J.G. Herman, J.R. Graff, S. Myöhänen, B.D. Nelkin, S.B. Baylin, Methylationspecific PCR: a novel PCR assay for methylation status of CpG islands, Proc. Natl. Acad. Sci. USA 93 (1996) 9821–9826.

62

I.P. Pogribny et al. / Mutation Research 669 (2009) 56–62

[27] K. Asada, R. Asada, H. Yoshiji, H. Fukui, R.A. Floyd, Y. Kotake, DNA cytosine methylation profile in various cancer-related genes is altered in cultured rat hepatocyte cell lines as compared with primary hepatocytes, Oncol. Rep. 15 (2006) 1241–1248. [28] D.S. Swafford, S.K. Middleton, W.A. Palmisano, K.J. Nikula, J. Tesfaigzi, S.B. Baylin, J.G. Herman, S.A. Belinsky, Frequent aberrant methylation of p16INK4a in primary rat lung tumors, Mol. Cell. Biol. 17 (1997) 1366–1374. [29] T. Niwa, S. Yamashita, T. Tsukamoto, T. Kuramoto, T. Nomoto, K. Wakazono, H. Fujita, T. Matsushima, M. Tatematsu, T. Sugimura, T. Ushijima, Whole-genome analyses of loss of heterozygosity and methylation analysis of four tumour suppressor genes in N-methyl-N -nitro-N-nitrosoguanidine-induced rat stomach carcinomas, Cancer Sci. 96 (2005) 409–413. [30] C.L. Powell, J.A. Swenberg, I. Rusyn, Expression of base excision DNA repair genes as a biomarker of oxidative DNA damage, Cancer Lett. 229 (2005) 1–11. [31] C.L. Dianov, J.L. Parsons, Co-ordination of DNA single strand break repair, DNA Repair (Amst.) 6 (2007) 454–460. [32] D.C. Cabelof, J.J. Raffoul, J. Nakamura, D. Kapoor, H. Abdalla, A.R. Heydari, Imbalanced base excision repair in response to folate deficiency is accelerated by polymerase ␤ haploinsufficiency, J. Biol. Chem. 279 (2004) 36504–36513. [33] J.K. Horton, M. Watson, D.F. Stefanick, D.T. Shaughnessy, J.A. Taylor, S.H. Wilson, XRCC1 and DNA polymerase ␤ in cellular protection against cytotoxic DNA single-strand breaks, Cell Res. 18 (2008) 48–63. [34] I.P. Pogribny, L. Muskhelishvili, B.J. Miller, S.J. James, Presence and consequence of uracil in preneoplastic DNA from folate/methyl-deficient rats, Carcinogenesis 18 (1997) 2071–2076. [35] G.N. Major, J.D. Collier, Repair of DNA lesion O6-methylguanine in hepatocellular carcinogenesis, J. Hepatobilliary Pancreat. Surg. 6 (1998) 355–366. [36] E.R. Zabarovsky, R. Gizatullin, R.M. Podowski, V.V. Zabarovska, L. Xie, O.V. Muravenko, S. Kozyrev, L. Petrenko, N. Skobeleva, J. Li, A. Protopopov, V. Kashuba, I. Ernberg, G. Winberg, C. Wahlestedt, NotI clones in the analysis of human genome, Nucleic Acids Res. 28 (2000) 1635–1639. [37] J. Martin, J.F. Dufour, Tumor suppressor and hepatocellular carcinoma, World J. Gastroenterol. 14 (2008) 1720–1733. [38] Y.J. Zhang, H. Ahsan, Y. Chen, R.M. Lunn, L.Y. Wang, S.Y. Chen, P.H. Lee, C.J. Chen, R.M. Santella, High frequency of promoter hypermethylation of RASSF1A and

[39]

[40]

[41] [42]

[43]

[44]

[45] [46]

[47] [48] [49]

p16 and its relationship to aflatoxin b1DNA adduct levels in human hepatocellular carcinoma, Mol. Carcinogenesis 35 (2002) 85–92. I.P. Pogribny, S.A. Ross, C. Wise, M. Pogribna, E.A. Jones, V.P. Tryndyak, S.J. James, Y.P. Dragan, L.A. Poirier, Irreversible global DNA hypomethylation as a key step in hepatocarcinogenesis induced by dietary methyl deficiency, Mutation Res. 593 (2006) 80–87. K. Shimizu, M. Onishi, E. Sugata, Y. Sokuza, C. Mori, T. Nishikawa, K. Honoki, T. Tsujiuchi, Disturbance of DNA methylation patterns in the early phase of hepatocarcinogenesis induced by a choline-deficient L-amino acid-defined diet in rats, Cancer Sci. 98 (2007) 1318–1322. P. Bannash, Preneoplastic lesions as end points in carcinogenicity testing. I. Hepatic preneoplasia, Carcinogenesis 7 (1986) 689–695. P. Bannasch, T. Haertel, Q. Su, Significance of hepatic preneoplasia in risk identification and early detection of neoplasia, Toxicol. Pathol. 31 (2003) 134–139. D. Nakae, Y. Kotake, H. Kishida, K.L. Hensley, A. Denda, Y. Kobayashi, W. Kitayama, T. Tsujiuchi, H. Sang, C.A. Stewart, T. Tabatabaie, R.A. Floyd, Y. Konishi, Inhibition by phenyl N-tert-butyl nitrone of early phase carcinogenesis in the livers of rats fed a choline-deficient L-amino acid-defined diet, Cancer Res. 58 (1998) 4548–4551. J. Nishimura, Y. Dewa, T. Okamura, M. Muguruma, M. Jin, Y. Saegusa, T. Umemura, K. Mitsumori, Possible involvement of oxidative stress in fenofibrate-induced hepatocarcinogenesis, Arch. Toxicol. 82 (2008) 641–654. B. Halliwell, Oxidative stress and cancer: have we moved forward? Biochem. J. 401 (2007) 1–11. K. Salnikov, A. Zhitkovich, Genetic and epigenetic mechanisms in metal carcinogenesis and cocarcinogenesis: nickel, arsenic, and chromium, Chem. Res. Toxicol. 21 (2008) 28–44. J. Liu, M.P. Waalkes, Liver is a target of arsenic carcinogenesis, Toxicol. Sci. 105 (2008) 24–32. R. Franco, O. Schoneveld, A.G. Georgakilas, M.I. Panayiotidis, Oxidative stress, DNA methylation and carcinogenesis, Cancer Lett. 266 (2008) 6–11. D. Huang, Y. Zhang, Y. Qi, C. Chen, W. Ji, Global DNA hypomethylation, rather than reactive oxygen species (ROS), a potential facilitator of cadmiumstimulated K562 cell proliferation, Toxicol. Lett. 179 (2008) 43–47.