The effect of dietary polyphenols on the epigenetic regulation of gene expression in MCF7 breast cancer cells

Toxicology Letters 192 (2010) 119–125

Contents lists available at ScienceDirect

Toxicology Letters journal homepage: www.elsevier.com/locate/toxlet

The effect of dietary polyphenols on the epigenetic regulation of gene expression in MCF7 breast cancer cells ´ Jarosław Paluszczak, Violetta Krajka-Kuzniak, Wanda Baer-Dubowska ∗ ´wi˛ecickiego 4, 60-781 Poznan, Poland Department of Pharmaceutical Biochemistry, Poznan University of Medical Sciences, ul. S

a r t i c l e

i n f o

Article history: Received 8 September 2009 Received in revised form 7 October 2009 Accepted 12 October 2009 Available online 17 October 2009 Keywords: Chemoprevention DNA methylation Epigenetics Histone methylation Polyphenols

a b s t r a c t The CpG island methylator phenotype is characterized by DNA hypermethylation in the promoters of several suppressor genes associated with the inactivation of various pathways involved in tumorigenesis. DNA methylation is catalyzed by specific DNA methyltransferases (DNMTs). Dietary phytochemicals particularly catechol-containing polyphenols were shown to inhibit these enzymes and reactivate epigenetically silenced genes. The aim of this study was to evaluate the effect of a wide range of dietary phytochemicals on the activity and expression of DNMTs in human breast cancer MCF7 cell line and their effect on DNA and histone H3 methylation. All phytochemicals inhibited the DNA methyltransferase activity with betanin being the weakest while rosmarinic and ellagic acids were the most potent modulators (up to 88% inhibition). While decitabine led to a partial demethylation and reactivation of the genes, none of the tested phytochemicals affected the methylation pattern or the expression of RASSF1A, GSTP1 or HIN1 in MCF7 cells. The global methylation of histone H3 was not affected by any of the tested phytochemicals or decitabine. The results of our study may suggest that non-nucleoside agents are not likely to be effective epigenetic modulators, in our experimental model at least. However, a long-term exposure to these chemicals in diet might potentially lead to an effect, which can be sufficient for cancer chemoprevention. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction The hypermethylation of DNA is a key epigenetic mechanism for the silencing of many genes, including those for cell cycle regulation, inflammatory and stress response, DNA repair and apoptosis (Esteller, 2008). Hypermethylation of certain genes, particularly tumor suppressor genes, is known to be associated with the inactivation of various pathways involved in tumorigenesis. DNA methylation is catalyzed by specific DNA methyltransferases (DMNTs), which use S-adenosyl-l-methionine (SAM) as the methyl group donor. Multiple DNMTs appear to be present in humans with varying degrees of specificity towards unmethylated and hemi-methylated DNA substrates. DNMT1 shows higher specificity towards hemi-methylated DNA substrates and is responsible for the maintenance of DNA methylation profiles during cell divisions. DNMT3A and DNMT3B catalyze rather de novo DNA methylation

Abbreviations: BA, baicalein; BET, betanin; COMT, catechol-O-methyltransferase; CYA, cyanidin; DAC, decitabine; DNMTs, DNA methyltransferases; EA, ellagic acid; EGCG, (−)-epigallocatechin-3-gallate; GAL, galangin; MYR, myricetin; PCA, protocatechuic acid; PHR, phloretin; RA, rosmarinic acid; RES, resveratrol; SAH, S-adenosyl-l-homocysteine; SIA, sinapic acid; SRA, syringic acid. ∗ Corresponding author. Tel.: +48 61 8546621; fax: +48 61 8546620. E-mail address: [email protected] (W. Baer-Dubowska). 0378-4274/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2009.10.010

and are essential for embryonic development. The function of DNMT2 methyltransferase is not well understood. Interestingly, DNMT1 seems to be more frequently required for aberrant DNA methylation in cancer cells. The reactivation of hypermethylated genes by the inhibition of DNMTs became a promising area in cancer therapy and chemoprevention (Egger et al., 2004; Kopelovich et al., 2003). There is a growing list of DNA methylation inhibitors in addition to 5-azacytidine and 5-aza-2 -deoxycytidine (decitabine, DAC), the first demethylating agents with well-characterized mechanisms of action (Jones and Taylor, 1980). However, the side effects and toxicity are serious concerns. There is a great need for the development of effective and non-toxic inhibitors of DNMTs not only for therapy but also for chemoprevention. It was shown recently that several potential chemopreventive agents, common ingredients of the human diet, are able to inhibit DNMTs, and reactivate genes silenced by aberrant methylation. (−)-Epigallocatechin-3gallate (EGCG) inhibited DNMT and reactivated the suppressor genes RARˇ, p16 (CDKN2A) and O6 -methylguanine methyltransferase in esophageal cancer KYSE 510 cells (Fang et al., 2003). The treatment of human breast cancer MCF-7 and MDA-MB-231 cell lines with either chlorogenic acid or caffeic acid also caused the inhibition of DNA methylation in the promoter region of RARˇ gene. It was suggested that these catechol-containing polyphenols may

120

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125

exert their activity through the increased formation of S-adenosyll-homocysteine (SAH) as a result of their O-methylation, which is mediated by catechol-O-methyltransferase (COMT) (Lee and Zhu, 2006). Further studies brought evidence that some other dietary components, e.g. genistein (Fang et al., 2005; Majid et al., 2009), nordihydroguaiaretic acid (Ciu et al., 2008), lycopene (King-Batoon et al., 2008), parthenolide (Liu et al., 2009) or Annurca apple polyphenols (Fini et al., 2007) may also affect DNA methylation. This activity was often gene-specific and cell-line dependent. Thus EGCG was not effective in reducing DNA methylation in T24 (urinary bladder transitional cell carcinoma), PC3 (prostate adenocarcinoma) and HT29 (colorectal adenocarcinoma) cancer cells and the reactivation of p16 in T24 (Chuang et al., 2005). Beside DNA methylation, gene silencing may be the effect of a shift in the profile of the covalent modifications of histones. The methylation of different amino acid residues in histones correlates with the reduction of the expression of the affected sequences, e.g. gene expression is lost or strongly reduced upon the methylation of histone H3 lysine 9 (H3K9) or lysine 27 (H3K27) (Esteller, 2008). MCF7 cell line, which was derived from breast adenocarcinoma tissue, is frequently used as a model in breast carcinogenesis and chemoprevention studies. It has been shown that MCF7 cells are susceptible to the induction of demethylation of aberrantly silenced genes by two phenolic acids: chlorogenic and caffeic acids (Lee and Zhu, 2006), but they were also resistant to the demethylating effects of genistein and lycopene (King-Batoon et al., 2008). Moreover, various flavonoids were shown to inhibit the activity of DNA methyltransferases (Lee et al., 2005), yet their effect on cancer cell epigenome remains to be elucidated. So, the aim of this study was to evaluate the effect of a wide range of dietary phytochemicals with phenolic acids, flavonoids and the stilbene structure on the activity of DNMT in nuclear extracts and its expression in the human breast cancer MCF7 cell line as well as to assess their effect on both the DNA and histone methylation. The compounds chosen for this study are present in edible plants and have a potential chemopreventive activity. Although showing DNMT inhibition at the screening stage in a cell-free system, none of the tested dietary polyphenols proved to be effective in the induction of either the demethylation of the RASSF1A, GSTP1 or HIN1 tumor suppressor genes or the global histone H3 methylation in MCF7 cells.

Table 1 The sequence of starters used in real-time PCR reactions. Primer

Sequence 

Product size

TBP forward TBP reverse

5 GGCACCACTCCACTGTATC 5 GGGATTATATTCGGCGTTTCG

183 bp

PBGD forward PBGD reverse

5 TCAGATAGCATACAAGAGACC 5 TGGAATGTTACGAGCAGTG

111 bp

DNMT1 forward DNMT1 reverse

5 GACCATCAGGCATTCTACC 5 TTACATTTCCCACACTCAGG

198 bp

GSTP1 forward GSTP1 reverse

5 CAAATACATCTCCCTCATCTACAC 5 TTGCCTCCCTGGTTCTGG

117 bp

compounds were added to the medium and the cells were incubated for 72 h. DMSO concentration did not exceed 0.2%. After 72 h, the cells were washed twice with PBS buffer and fresh medium containing MTT salt (0.5 mg/ml) was added. After a 4-h incubation, the formazan crystals were dissolved in acidic isopropanol and the absorbance was measured at 540 nm and 690 nm. All the experiments were repeated three times, with at least three measurements per experiment. In all the subsequent experiments, non-toxic concentrations of polyphenols (viability above 75% based on the MTT assay) were used, ranging from 10 ␮M to 40 ␮M. DMSO concentration did not exceed 0.2%. 2.3. Poly(ADP-ribose) glycohydrolase (PARG) assay The HT Colorimetric PARG Assay Kit from Trevigen (Gaithersburg, MD, USA) was used for the assessment of the influence of the phytochemicals on the activity of PARG, following manufacturer’s protocol. In this assay, poly(ADP-ribose) polymerase catalyzes the poly(ADP-ribosyl)ation of the activated histones attached to the wells. Subsequently, the product is degraded by the addition of PARG and poly(ADP-ribose) level is measured at 450 nm. The tested compounds were added to the PARG incubation mixture at the concentration of 50 ␮M or 100 ␮M. 2.4. Determination of the activity of DNA methyltransferases

2. Materials and methods

Nuclear extracts containing DNMTs were prepared using the Nuclear Extraction Kit I from Epigentek (Brooklyn, NY, USA) according to the manufacturer’s protocol. The protein content was assessed by the Lowry method (Lowry et al., 1951). The activity of DNA methyltransferases in the nuclear extracts was measured with the Epiquick DNA Methyltransferase Activity/Inhibition Kit (Epigentek) according to the manufacturer’s protocol. The activity of DNMTs in nuclear extracts is assessed, based on the methylation of CG-rich oligonucleotides immobilized in wells. Any methylated cytosines are detected by the ELISA assay. For inhibitor screening purposes, we used the nuclear extract from nonstimulated MCF7 cells as the source of DNMTs. The extract was incubated in the presence of 20 ␮M or 40 ␮M of the tested phytochemicals. Each compound was dissolved in DMSO. All the experiments were repeated twice with three measurements per experiment.

2.1. Chemicals

2.5. DNA methylation analysis

Baicalein (BA), decitabine (DAC), myricetin (MYR), protocatechuic acid (PCA), phloretin (PHR), sinapic acid (SIA), syringic acid (SRA), resveratrol (RES), rosmarinic acid (RA), and ellagic acid (EA) were obtained from Sigma (St. Louis, MO, USA). Betanin (BET), cyanidin (CYA) and galangin (GAL) were obtained from ABCR (Karlsruhe, Germany). All the compounds were dissolved in DMSO (100 mM stock solutions) and stored at −20 ◦ C. Decitabine was dissolved in PBS buffer and stored as a 1 mM solution. TrueStart Hot Start Taq DNA Polymerase from Fermentas (Burlington, Canada) was used in methylation-specific PCR reactions. All the primers were obtained from Oligo.pl (Warsaw, Poland). The primary antibodies against DNMT1 and ␤-actin and secondary antibodies were supplied by Santa Cruz Biotechnology (Santa Cruz, CA, USA). The primary antibodies against anti-histone H3, anti-dimethyl-histone H3 (Lys 9), antitrimethyl-histone H3 (Lys 9), and anti-trimethyl-histone H3 (Lys 27) were supplied by Upstate (Temecula, CA, USA). The rainbow colored protein molecular weight marker was purchased from Amersham Pharmacia Biotechnology (Piscataway, NJ, USA). All the other chemicals were commercial products of the highest purity available.

The methylation status of RASSF1A, GSTP1, HIN-1 was assessed using the methylation-specific polymerase chain reaction (MSP) (Hermann et al., 1996). For the DNA methylation and mRNA transcript analysis, 6 × 104 of MCF7 cells were seeded in 35 mm plates. After 24-h preincubation in DMEM containing 5% FBS, the tested compounds were added to culture and cells were grown for 72 h. DNA was extracted with GenElute Mammalian Genomic DNA Extraction Kit (Sigma) and, subsequently, bisulfite converted using the EZ DNA Conversion Kit from ZymoResearch (Orange, CA, USA). All the experiments were repeated at least twice. The primers and reaction conditions for MSP were chosen based on published data (King-Batoon et al., 2008; Bae et al., 2004; Krop et al., 2004). DNA extracted from the lymphocytes of healthy blood donors was used as a negative control and a completely methylated human DNA from New England Biolabs (Ipswich, MA, USA) as a positive control in the MSP reactions. Amplification products were resolved on 2.5% agarose gels and visualized in UV after ethidium bromide staining.

2.2. Cell culture The well-characterized estrogen-dependent epithelial breast cancer MCF7 cell line was used in all the experiments. The cells were maintained in DMEM, containing 5% FBS and antibiotics/antimycotics at 37 ◦ C in a humidified 5% CO2 atmosphere. The effect of the chemicals on cellular proliferation was assessed with the MTT assay, according to standard protocols. Briefly, 104 MCF7 cells were seeded per well in a 96-well plate. After 24 h of preincubation in DMEM containing 5% FBS, the tested

2.6. Real-time PCR The total RNA was extracted with the Total RNA Extraction Kit (Sigma) according to the manufacturer’s protocol, and subjected to reverse transcription using the RevertAid Kit (Fermentas), followed by quantitative real-time PCR. For real-time analyses, the Maxima SYBR Green Kit (Fermentas) and a BioRad Chromo4 were used. The protocol started with a 5 min enzyme activation at 95 ◦ C, followed by 40 cycles of 95 ◦ C for 15 s, 54 ◦ C for 20 s and 72 ◦ C for 40 s and the final elongation at 72 ◦ C for 5 min. The melting curve analysis was used for product size verification. Experiments were normalized for the expression of TBP and PBGD. The Pfaffl relative method was used for fold-change quantification. Primer sequences are listed in Table 1.

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125

121

2.7. Western blot For the determination of the level of the DNMT1 protein or the covalent modifications of histones, 5 × 105 of MCF7 cells were seeded in 100 mm plates. After 24-h preincubation in DMEM containing 5% FBS, tested compounds were added to the culture and cells were grown for 72 h. Nuclear proteins were extracted using the Nuclear/Cytosol Fractionation Kit from BioVision (Mountain View, CA, USA). Standard acidic extraction with 0.2 M HCl was used for the preparation of histones. The protein content in the samples was determined with the Lowry method (1951). All the experiments were repeated twice. Nuclear extracts or acid-extracted proteins (50 ␮g) were separated on 5% or 15% SDS-PAGE slab gels and proteins were transferred to nitrocellulose membranes (Laemmli, 1970; Towbin et al., 1979). After blocking with 10% skimmed milk, proteins were probed with goat anti-human DNMT1, rabbit anti-human histone H3, rabbit anti-human dimethyl histone H3 (Lys 9), rabbit anti-human trimethyl histone H3 (Lys 9), rabbit anti-human trimethyl histone H3 (Lys 27) and rabbit anti-human ␤-actin antibodies. The ␤-actin protein and histone H3 were used as internal controls. As secondary antibodies in the staining reaction, the alkaline phosphatase-labeled anti-goat IgG and anti-rabbit IgG were used. The amount of immunoreactive product in each lane was determined using Quantity One software (BioRad Laboratories, Hercules, CA, USA). Values were calculated as relative absorbance units (RQ) per mg protein.

2.8. Statistical analysis The statistical analysis was performed by one-way ANOVA. The statistical significance between the experimental groups and their respective controls was assessed by Tukey’s post hoc test, with p < 0.05.

3. Results 3.1. The effect of phytochemicals on cell viability The MTT test was used to evaluate the effect of phytochemicals on the viability of MCF7 cells. Within the concentration range of 2–200 ␮M, most of the phytochemicals reduced the viability of the MCF7 cells in a dose-dependent manner. Rosmarinic acid, baicalein and myricetin were the most toxic, while sinapic and syringic acids showed little toxicity and betanin was not toxic within the tested concentration range (Fig. 1).

Fig. 1. The effect of phytochemicals on the viability of MCF7 cells. Mean values from three experiments ±SEM are shown. The abbreviations (in order of appearance) are as follows: PCA, protocatechuic acid; SIA, sinapic acid; SRA, syringic acid; RA, rosmarinic acid; EA, ellagic acid; MYR, myricetin; BA, baicalein; GAL, galangin; RES, resveratrol; PHR, phloretin; BET, betanin.

Fig. 2. (A) The effect of phytochemicals on PARG activity expressed as percent inhibition in comparison to control measurements (no inhibition). The mean values from three measurements ±SEM are shown. (B) The effect of phytochemicals on the activity of DNA methyltransferases. The mean values ±SEM are shown.

122

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125

betanin, or phloretin at the concentration of 20 ␮M and 40 ␮M. Decitabine (2.5 ␮M and 5.0 ␮M) was used as the reference compound in all the experiments. As expected, decitabine led to a partial demethylation of the genes under study in MCF7 cells (Fig. 3). None of the polyphenols induced the demethylation of any of the analyzed genes. An extended treatment of MCF7 cells for 6 days in the presence of protocatechuic and rosmarinic acids (at two concentrations, 20 ␮M and 40 ␮M, with the compound changed every 48 h) did not lead to any changes in DNA methylation (data not shown). Similarly, a 72h incubation of MCF7 cells in the presence of ellagic acid (20 ␮M or 40 ␮M), phloretin (20 ␮M or 40 ␮M), resveratrol (10 ␮M) and rosmarinic acid (20 ␮M), with compound dosing every 24 h, did not change the methylation profile (data not shown). A subsequent analysis of the GSTP1 transcript showed a significant increase in its level in comparison to the control cells (treated with DMSO) as the result of decitabine treatment in MCF7 cells. A slight increase was also observed as an effect of ellagic acid treatment at the dose of 40 ␮M. Rosmarinic acid and phloretin did not affect the expression of GSTP1 in MCF7 cells (Fig. 3G). 3.4. The effect of phytochemicals on the expression of DNMT1 The expression of DNMT1 was determined at the level of both the transcript and the protein in MCF7 cells. Decitabine reduced the level of both the DNMT1 transcript and protein (Fig. 4). Upon

Fig. 3. The effect of phytochemicals on the methylation of RASSF1A, GSTP1 and HIN-1 in the MCF7 cell line. Panels A, C and E—the MSP results for methylated: RASSF1A, GSTP1 and HIN-1, respectively. Panels B, D, and F—the MSP results for unmethylated: RASSF1A, GSTP1 and HIN-1, respectively. The lanes 1,2 – decitabine 2.5/5.0 ␮M; 3,4 – ellagic acid 20/40 ␮M; 5,6 – protocatechuic acid 20/40 ␮M; 7,8 – sinapic acid 20/40 ␮M; 9,10 – syringic acid 20/40 ␮M; 11,12 – rosmarinic acid 20/40 ␮M; 13,14 – betanin 20/40 ␮M; 15,16 – phloretin 20/40 ␮M. (G) The effect of decitabine, phloretin, rosmarinic acid and ellagic acid on the level of the GSTP1 transcript in MCF7 cells. The values were calculated as a transcript relative change in comparison to control cells treated with DMSO (expression equals 1). The mean values from two experiments (each run in triplicate) ±SEM are shown.

Myricetin, ellagic and rosmarinic acids, in a cell-free assay, efficiently inhibited the activity of PARG (Fig. 2A), the enzyme which might protect against cytotoxic effects of these compounds. 3.2. The effect of polyphenols on DNMT activity in the cell-free system The effect of phytochemicals on the activity of DNMTs was assessed in nuclear extracts from MCF7 cells at the concentration of 20 ␮M and 40 ␮M, which were chosen based on reports on the activity of compounds with similar chemical structures (Lee and Zhu, 2006; Lee et al., 2005). The results are shown in Fig. 2B. All the phytochemicals inhibited the DNA methyltransferase activity, with betanin being the weakest, and rosmarinic acid the most potent modulator (up to 88% inhibition). The stilbene derivative, resveratrol, exhibited lower activity than its proximate metabolite, piceatannol. 3.3. The effect of phytochemicals on gene methylation and expression in cell culture For the evaluation of the ability of the tested phytochemicals to reactivate silenced genes, the methylation status of the promoter regions of RASSF1A, GSTP1 and HIN-1 genes, which are completely methylated in MCF7 cells, was assessed. MCF7 cells were grown for 72 h in the presence of ellagic acid, protocatechuic acid, sinapic acid, syringic acid, rosmarinic acid,

Fig. 4. The effect of phenolic compounds on the expression of DNMT1. (A) The level of mRNA for DNMT1 in MCF7 cells. The values were calculated as a transcript relative change in comparison to the control cells treated with DMSO (expression equals 1). The mean values from two experiments run in triplicate ±SEM are shown. (B) Representative immunoblots from two independent experiments. The expression of ␤-actin was used as an internal control (not shown). The lanes in panel B correspond to the respective bars below in panel C. (C) Data present the level of DNMT1 protein in MCF7 cells as a percentage of the control groups (mean ± SEM), from two separate experiments run in triplicate. The asterisk above the bar denotes statistically significant differences from the control group, *p < 0.05.

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125

123

Fig. 5. The effect of phenolic compounds on histone H3 modifications in MCF-7 cells. (A) The representative immunoblots from two independent experiments. (B) Data present the modifications of histone H3 as a percentage of the control groups (mean ± SEM), from two separate experiments run in triplicate. The bars in panel B correspond to the lanes in panel A.

treatment with the tested polyphenols, neither the DNMT1 transcript nor the DNMT1 protein level was significantly affected. The DNMT1 transcript level only slightly increased after treatments with phloretin and rosmarinic acid. Moreover, rosmarinic acid significantly reduced the DNMT1 protein level in MCF7 cells at both concentrations (20 ␮M and 40 ␮M) by about 30% and 20%, respectively (Fig. 4). 3.5. The effect of phytochemicals on histone H3 methylation In order to check whether the tested compounds may affect the global methylation of histone H3, MCF7 cells were treated with protocatechuic acid, rosmarinic acid, phloretin or resveratrol and grown for 72 h. None of the compounds significantly affected the global level of any of the analyzed covalent histone modifications (Fig. 5). 4. Discussion The essential role for epigenetic defects in carcinogenesis has been established in the last ten years. Aberrant methylation is a mechanism of inactivation of tumor suppressor genes, which are as common as mutations. The important difference is the reversibility of DNA methylation and other epigenetic modifications such as histone acetylation/deacetylation and histone methylation/demethylation (Esteller, 2008). Current data obtained from different experimental models indicate that epigenetic abnormalities may take place at the earliest stages of carcinogenesis, and be responsible for the formation of cancer stem cells (Balch et al., 2007). Tumor-type specificity and an early appearance of aberrant DNA methylation make it an excellent cancer biomarker and the target for both chemoprevention and therapy (Esteller, 2008; Paluszczak and Baer-Dubowska, 2006).

For chemoprevention purposes, the chemical components of edible fruits and vegetables are particularly useful, and several reports have been published in recent years, indicating that phytochemicals may reactivate genes silenced by aberrant methylation. In this regard, it was shown that common phenolic compounds, EGCG, caffeic acid and chlorogenic acid can inhibit DNMTs and restore the expression of certain genes, which are silenced in cancer cells. Thus, it was demonstrated that caffeic acid or chlorogenic acid partially inhibited the methylation of the promoter region of the RARˇ gene in breast cancer cells MCF7 and MDA-MB-231 (Lee and Zhu, 2006; Lee et al., 2005), while EGCG in the doses of 5–50 ␮M caused a reversal of the hypermethylation of p16(INK4A), RARˇ, O6 -methylguanine metyltransferase (MGMT) in esophageal cancer KYSE 510 cells (Fang et al., 2003). Yet, no changes were detected in the methylation of p16 or MAGE-A1 and LINE sequences in bladder carcinoma T24, colorectal adenocarcinoma HT29 or prostate adenocarcinoma PC3 cell lines treated with the same compound (Chuang et al., 2005). In this study, we have evaluated the ability of different plant phenolic acids, flavonoids and stilbene derivatives to inhibit the DNMT activity in nuclear extracts and its expression in MCF7 breast cancer cell line. Viability assays showed that betanin, a chemical present in red beetroot did not exhibit any toxicity within the tested concentration range (2–200 ␮M), while ellagic and rosmarinic acids and myricetin were the most toxic. The latter were also tested for their ability to inhibit PARG activity in a cell-free system. This enzyme, among other functions, may also play a role in cellular protection against the toxic effects of certain chemicals. Its inhibition may lead to the accumulation of the intracellular level of poly(ADP-ribose), which is a known pro-apoptotic factor (Heeres and Hergenrother, 2007). Thus, the cytotoxic effect of myricetin as well as ellagic and rosmarinic acids observed in the MTT test may, at least in part, be

124

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125

attributed to this effect since all of these compounds reduced PARG activity, although this hypothesis will need further verification. Most of the polyphenols tested in this study inhibited the DNMT activity in nuclear extracts from MCF7 cells. While the most potent inhibitors of DNMT were ellagic and rosmarinic acids and the flavonoid, baicalein, the least potent inhibitor, was betanin. However, in contrast to decitabine, neither of these phenolic acids nor phloretin affected the DNMT1 transcript level in MCF7 cells, although rosmarinic acid reduced its protein level. Since the most potent phenolic inhibitors of DNMT in the cellfree system were also the most toxic, it can be assumed that these polyphenols, particularly phenolic acids, may bind not only to the substrate binding site but also to an additional site, that causes the loss of enzyme activity. Such an effect is often observed towards other enzymes, e.g. cytochrome P450 in the case of this class of compounds (Baer-Dubowska et al., 1998). However, for catechol-containing polyphenols, the non-competitive inhibition of DNA methylation catalyzed by DNMT largely due to the increased formation of S-adenosyl-l-homocysteine resulting from the catechol-O-methyltransferase-mediated O-methylation of these chemicals was suggested (Lee and Zhu, 2006). It was based on the fact that chlorogenic and caffeic acids (5 ␮M and 20 ␮M) exerted a very weak direct inhibition of human DNMT1-mediated methylation in the absence of COMT. Such a mechanism cannot be excluded for the polyphenols tested in our study, particularly ellagic and rosmarinic acids, since nuclear extracts might contain COMT (Ulmanen et al., 1997; Weisz et al., 2000). Nevertheless, as dietary green tea polyphenols did not affect the SAH level in mice, this type of indirect inhibition may not occur in vivo. In our study, the analysis of the methylation status of the promoter regions of RASSF1A, GSTP1 and HIN-1 in MCF7 cells, treated with phenolic acids, phloretin and betanin also seems to exclude the possibility of in vivo DNMT inhibition. These genes are hypermethylated in this cell line and none of the compounds tested in this study caused a reversal of this status. The only compound which significantly affected gene methylation in our study was decitabine used as the reference compound. The change of the methylation status of the promoter region of GSTP1 gene caused by this agent resulted in its reactivation. The results of our study may also suggest that an efficient inhibition of DNA methylation may depend not only on the direct interaction with DNMT, but also a diminished expression and/or stability of this enzyme in the cells. Epigenetically silenced genes may also be reactivated by factors modulating the covalent modifications of histones. The effect of the inhibitors of histone deacetylases on the induction of re-expression of genes silenced in cancer cells is now well characterized. These compounds are especially active when used in combination with DNMT inhibitors. The growing understanding of the role of different chemical modifications of histones, allows us to consider them as especially important in the formation of cancer epigenome. Such modifications include the methylation of lysines 9 and 27 in histone H3. Therefore, the potential ability of the tested phytochemicals to affect the global level of H3K9me2, H3K9me3 and H3K27me3 was also assessed in our study. However, no significant changes were detected. These findings do not exclude the likelihood of more discrete changes, which may lead to a local shift in the chromatin structure and affect gene expression, although it has to be stressed that the natural dietary modulators of histone methylation have not been identified so far. Our understanding of the action of currently known epigenetic modifiers is far from complete. The same compounds show different activities in different cell lines. The effect is often gene-specific. Nordihydroguaiaretic acid led to the reactivation of p16 in breast cancer T47D and in colorectal cancer RKO cell lines at concentrations ranging from 10 ␮M to 100 ␮M (Ciu et al., 2008), while it did not affect LINE-1 methylation at 100 ␮M in hepatocellular

carcinoma HepG2 cells (Byun et al., 2008). Genistein (2–20 ␮M) reversed the hypermethylation and reactivated the expression of RARˇ, p16 and MGMT in esophageal squamous cell carcinoma KYSE510 cells (Fang et al., 2005). It was also effective in the induction of demethylation and reactivation of the expression of a novel tumor suppressor gene BTG3 in A498, ACHN and HEK-293 renal cancer cell lines. In this model, genistein not only reversed aberrant methylation through the inhibition of DNMTs and methylated DNA binding proteins, but also induced changes in histone modifications. The activity of histone acetyltransferases was induced, and active histone modifications were introduced to chromatin. Moreover, genistein induced the expression of p16 in prostate carcinoma DuPro cells by a methylation-independent manner, leading to an increased acetylation of histones H3 and H4 (Majid et al., 2009). Genistein and lycopene showed differential effects on DNA methylation in MCF-7, non-cancerous MCF10A breast cells and breast cancer MDA-MB-468 cells. MCF-7 cells were resistant to the demethylation induced by these phytochemicals in contrast to MCF10A and MDA-MB-468 cells. The effect was also gene-specific. Both the compounds led to the reactivation of GSTP1, but not RARˇ2 in MDA-MB-468 cells. Only lycopene, not genistein, led to the demethylation of RARˇ2 and HIN-1 in MCF10A cells (King-Batoon et al., 2008). Taken together, the results of our study seem to confirm the observations of Chuang et al. (2005) that non-nucleoside agents might not likely be effective as epigenetic therapies. However, the weak modulating effects of these compounds, which are common ingredients of food, cannot be excluded. A long-term exposure to these chemicals in diet might potentially lead to an effect which can be sufficient for chemoprevention. In this regard, it was shown that even nucleoside agents like decitabine restore the expression of hypermethylated genes after prolonged exposure (for many generations) of the cells to this compound (Laird et al., 1995). Evidence also exists of the role of inflammation in the induction of aberrant DNA methylation, e.g. through chlorination of cytosine residues (Valinluck and Sowers, 2007). Thus, dietary phytochemicals, which often show anti-inflammatory potential, could indirectly affect the epigenome by the modulation of inflammatory reactions. Conflict of interest None declared. Acknowledgement This research was supported by the Polish Ministry of Science and Higher Education (grant No. N405 048 31/3338). References Bae, Y.K., Brown, A., Garrett, E., Bornman, D., Fackler, M.J., Sukumar, S., Herman, J.G., Gabrielson, E., 2004. Hypermethylation in histologically distinct classes of breast cancer. Clin. Cancer Res. 10, 5998–6005. ´ Baer-Dubowska, W., Szaefer, H., Krajka-Kuzniak, V., 1998. Inhibition of murine hepatic cytochrome P450 activities by natural and synthetic phenolic compounds. Xenobiotica 28, 735–743. Balch, C., Nephew, K.P., Huang, T.H., Bapat, S.A., 2007. Epigenetic “bivalently marked” process of cancer stem cell-driven tumorigenesis. Bioessays 29, 842–845. Byun, H.-M., Choi, S.H., Laird, P.W., Trinh, B., Siddiqui, M.A., Marquez, V.E., Yang, A.S., 2008. 2 -Deoxy-N4-[2-(4-nitrophenyl)ethoxycarbonyl]-5-azacytidine: a novel inhibitor of DNA methyltransferase that requires activation by human carboxylesterase 1. Cancer Lett. 266, 238–248. Chuang, J.C., Yoo, C.B., Kwan, J.M., Li, T.W.H., Liang, G., Yang, A.S., Jones, P.A., 2005. Comparison of biological effects of non-nucleoside DNA methylationinhibitors versus 5-aza-2 -deoxycytidine. Mol. Cancer Ther. 4, 1515–1520. Ciu, Y., Lu, C., Liu, L., Sun, D., Yao, N., Tan, S., Bai, S., Ma, X., 2008. Reactivation of methylation-silenced tumor suppressor gene p16INK4a by nordihydroguaiaretic acid and its implication in G1 cell cycle arrest. Life Sci. 82, 247–255. Egger, G., Liang, G., Aparicio, A., Jones, P.A., 2004. Epigenetics in human disease and prospects for epigenetic therapy. Nature 429, 457–463.

J. Paluszczak et al. / Toxicology Letters 192 (2010) 119–125 Esteller, M., 2008. Epigenetics in cancer. N. Engl. J. Med. 358, 1148–1159. Fang, M.Z., Chen, D., Sun, Y., Jin, Z., Christman, J.K., Yang, C.S., 2005. Reversal of hypermethylation and reactivation of p16INK4a , RAR␤, and MGMT genes by genistein and other isoflavones from soy. Clin. Cancer Res. 11, 7033–7041. Fang, M.Z., Wang, Y., Ai, N., Hou, Z., Sun, Y., Lu, H., Welsh, W., Yang, C.S., 2003. Tea polyphenol (−)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines. Cancer Res. 63, 7563–7570. Fini, L., Selgrad, M., Fogliano, V., Graziani, G., Romano, M., Hotchkiss, E., Daoud, Y.A., De Vol, E.B., Boland, C.R., Ricciardiello, L., 2007. Annurca apple polyphenols have potent demethylating activity and can reactivate silenced tumor suppressor genes in colorectal cancer cells. J. Nutr. 137, 2622–2628. Heeres, J.T., Hergenrother, P.J., 2007. Poly(ADP-ribose) makes a date with death. Curr. Opin. Chem. Biol. 11, 644–653. Hermann, J.G., Graff, J.R., Myohanen, S., Nelkin, B.D., Baylin, S.B., 1996. Methylationspecific PCR: a novel PCR assay for methylation status of CpG islands. Proc. Natl. Acad. Sci. U.S.A. 93, 9821–9826. Jones, P.A., Taylor, S.M., 1980. Cellular differentiation, cytidine analogs and methylation. Cell 20, 85–93. King-Batoon, A., Leszczynska, J.M., Klein, C.B., 2008. Modulation of gene methylation by genistein or lycopene in breast cancer cells. Environ. Mol. Mutagen. 49, 36–45. Kopelovich, L., Crowell, J.A., Fay, J.R., 2003. The epigenome as a target for cancer chemoprevention. J. Natl. Cancer Inst. 95, 1747–1757. Krop, I., Player, A., Tablante, A., Taylor-Parker, M., Lahti-Domenici, J., Fukuoka, J., Batra, S.K., Papadopoulos, N., Richards, W.G., Sugarbaker, D.J., Wright, R.L., Shim, J., Stamey, T.A., Sellers, W.R., Loda, M., Meyerson, M., Hruban, R., Jen, J., Polyak, K., 2004. Frequent HIN-1 promoter methylation and lack of expression in multiple human tumor types. Mol. Cancer Res. 2, 489–494. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Laird, P., Jackson-Grusby, L., Fazeli, A., Dickinson, S.L., Jung, W.E., Li, E., Weiberg, R.A., Jaenisch, R., 1995. Suppression of intestinal neoplasia by DNA hypomethylation. Cell 81, 197–205.

125

Lee, W.J., Shim, J.-Y., Zhu, B.T., 2005. Mechanisms for the inhibition of DNA methyltransferases by tea catechins and bioflavonoids. Mol. Pharmacol. 68, 1018– 1030. Lee, W.J., Zhu, B.T., 2006. Inhibition of DNA methylation by caffeic acid and chlorogenic acid, two common catechol-containing coffee polyphenols. Carcinogenesis 27, 269–277. Liu, Z., Liu, S., Xie, Z., Pavlovicz, R.E., Wu, J., Chen, P., Aimiuwu, J., Pang, J., Bhasin, D., Neviani, P., Fuchs, J.R., Plass, C., Li, P.K., Li, C., Huang, T.H.M., Wu, L.C., Rush, L., Wang, H., Perroti, D., Marcucci, G., Chan, K.K., 2009. Modulation of DNA methylation by a sesquiterpene lactone parthenolide. J. Pharmacol. Exp. Ther. 329, 505–514. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 230–265. Majid, S., Dar, A.A., Ahmad, A., Hirata, H., Kawakami, K., Shahryari, V., Saini, S., Tanaka, Y., Dahiya, A.V., Dahiya, R., 2009. BTG3 tumor suppressor gene promoter demethylation, histone modification and cell cycle arrest by genistein in renal cancer. Carcinogenesis 30, 662–670. Paluszczak, J., Baer-Dubowska, W., 2006. Epigenetic diagnostics of cancer: the application of DNA methylation markers. J. Appl. Genet. 47, 365–375. Towbin, H., Staehelin, T., Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350–4354. Ulmanen, I., Peranen, J., Tenhunen, J., Tilgmann, C., Karhunen, T., Panula, P., Bernasconi, L., Aubry, J.-P., Lundstrom, K., 1997. Expression and intracellular localization of catechol-O-methyltransferase in transfected mammalian cells. Eur. J. Biochem. 243, 452–459. Valinluck, V., Sowers, L.C., 2007. Inflammation-mediated cytosine damage: a mechanistic link between inflammation and the epigenetic alterations in human cancers. Cancer Res. 67, 5583–5586. Weisz, J., Fritz-Wolz, G., Gestl, S., Clawson, G.A., Creveling, C.R., Liehr, J.G., Dabbs, D., 2000. Nuclear localization of catechol-O-methyltransferase in neoplastic and nonneoplastic mammary epithelial cells. Am. J. Pathol. 156, 1841–1848.