Document zR4gO24nLew4GNORa1rxBpp6

Toxicology and Applied Pharmacology xxx (2010) xxxxxx YTAAP-11871; No. of pages: 11; 4C: 3, 6 Contents lists available at ScienceDirect Toxicology and Applied Pharmacology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / y t a a p 1 Epigenetic influences of low-dose bisphenol A in primary human breast 2 epithelial cells 3 Yu-I Weng a, Pei-Yin Hsu a, Sandya Liyanarachchi a, Joseph Liu a, Daniel E. Deatherage a, Yi-Wen Huang a, 4 Tao Zuo a, Benjamin Rodriguez a, Ching-Hung Lin b, Ann-Lii Cheng b, Tim H.-M. Huang a, 5 a Human Cancer Genetics Program, The Ohio State University, Columbus, OH 43210, USA 6 b Department of Internal Medicine and Oncology, National Taiwan University Hospital, Taipei, Taiwan 7 article info 8 9 Article history: 10 Received 24 May 2010 11 Revised 16 July 2010 12 Accepted 16 July 2010 13 Available online xxxx 1456 17 Keywords: 18 Bisphenol A 19 Estrogen 20 DNA methylation 21 Epigenetics 22 Breast cancer 44 43 abstract Substantial evidence indicates that exposure to bisphenol A (BPA) during early development may increase 23 breast cancer risk later in life. The changes may persist into puberty and adulthood, suggesting an epigenetic 24 process being imposed in differentiated breast epithelial cells. The molecular mechanisms by which early 25 memory of BPA exposure is imprinted in breast progenitor cells and then passed onto their epithelial 26 progeny are not well understood. The aim of this study was to examine epigenetic changes in breast 27 epithelial cells treated with low-dose BPA. We also investigated the effect of BPA on the ER signaling 28 pathway and global gene expression profiles. Compared to control cells, nuclear internalization of ER was 29 observed in epithelial cells preexposed to BPA. We identified 170 genes with similar expression changes in 30 response to BPA. Functional analysis confirms that gene suppression was mediated in part through an ER- 31 dependent pathway. As a result of exposure to BPA or other estrogen-like chemicals, the expression of 32 lysosomal-associated membrane protein 3 (LAMP3) became epigenetically silenced in breast epithelial cells. 33 Furthermore, increased DNA methylation in the LAMP3 CpG island was this repressive mark preferentially 34 occurred in ER-positive breast tumors. These results suggest that the in vitro system developed in our 35 laboratory is a valuable tool for exposure studies of BPA and other xenoestrogens in human cells. Individual 36 and geographical differences may contribute to altered patterns of gene expression and DNA methylation in 37 susceptible loci. Combination of our exposure model with epigenetic analysis and other biochemical assays 38 can give insight into the heritable effect of low-dose BPA in human cells. 39 Published by Elsevier Inc. 40 412 45 Introduction 46 Bisphenol A (BPA), first synthesized by A. P. Dianin in 1891, has been 47 widely used as a cross-linking reagent in the manufacture of epoxy 48 resins since 1950s (Vogel, 2009). It is extensively used in a board range 49 of products, including toys, water pipes, drinking bottles, baby bottles, 50 food containers, tubing, and dental sealants (Welshons et al., 2006). 51 Presently, the worldwide production of BPA exceeds 3 billion kg/year 52 (Vandenberg et al., 2009). Studies have shown that BPA can be released 53 from incomplete polymerization upon heating or leached out through 54 normal use (Mountfort et al., 1997; Kang et al., 2003; Goodson et al., 55 2004). Because of its ubiquity in environment, low levels of BPA can be 56 detected in 92.6% of urine samples (6 years of age ranging from 0.4 to 57 149 g/l) in the National Health and Nutrition Examination Survey 58 (NHANES) 20032004 (Calafat et al., 2008; CDC, 2009). Animal studies 59 have shown that these low levels of BPA exposure may alter Corresponding author. Human Cancer Genetics Program, The Ohio State University, Room 814, Biomedical Research Tower, 460 West 12th Avenue, Columbus, OH 43210, USA. Fax: + 1 614 292 5995. E-mail address: Tim.Huang@osumc.edu (T.H.-M. Huang). developmental programs of sensitive end organs, like mammary and 60 prostate gland, during critical stages of early development (Markey et 61 al., 2001; Nikaido et al., 2004; Timms et al., 2005). The changes may 62 persist into puberty and adulthood, suggesting an imprinting process 63 being imposed in differentiated breast epithelial cells (Markey et al., 64 2001; Munoz-de-Toro et al., 2005). 65 The molecular mechanisms by which early memory of BPA 66 exposures can be imprinted in breast progenitor cells and then 67 passed onto their epithelial progeny are not well understood. One 68 distinct possibility is through epigenetic remodeling of DNA 69 structure without altering the nucleotide sequence itself. The 70 changes, including DNA methylation, frequently occur in GC-rich 71 promoter CpG islands of transcriptionally repressed genes (Ohm 72 and Baylin, 2007; Widschwendter et al., 2007). Studies have 73 suggested that DNA methylation of a promoter CpG island, or 74 promoter hypermethylation, can be initiated in progenitor genomes 75 and heritably passed onto the differentiated progeny (Jones and 76 Baylin, 2007; Marotta and Polyak, 2009). This epigenetic process is 77 known to cause phenotypic variations among individuals and 78 contributes to the development of pathological conditions, like 79 cancer (Feinberg et al., 2006; Esteller, 2007). 80 0041-008X/$ see front matter. Published by Elsevier Inc. doi:10.1016/j.taap.2010.07.014 Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 2 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 81 Previous studies of epigenetic effects of BPA preexposure mainly 82 rely on animal models and epidemiological surveys (Ho et al., 2006; 83 Dolinoy et al., 2007; Prins et al., 2008; Yaoi et al., 2008). Whereas 84 these observations strongly implicate that the exposure to low-dose 85 BPA is potentially harmful to human health, the challenge encoun86 tered is validation studies of the findings in primary human cells. In 87 this regard, we have recently established a human preexposure model 88 for epigenetic studies (Cheng et al., 2008; Hsu et al., 2009). In the 89 model, breast progenitor cells were first exposed to different 90 environmental chemicals, and then these cells were differentiated 91 into epithelial cells in the absence of these environmental stimulants. 92 We hypothesize that slow-dividing progenitor cells have a longer life 93 span and thus are more susceptible to environmental injuries and can 94 transmit this injured memory to their differentiated progeny through 95 epigenetic mechanisms. In our previous studies, the preexposure to 96 17-estradiol (E2) and diethylstilbestrol (DES) may trigger epigenetic 97 repression of protein-coding genes and non-coding microRNAs, some 98 of which exhibit promoter hypermethylation in breast cancer cells 99 (Cheng et al., 2008; Hsu et al., 2009). 100 Here, we extended this preexposure study to evaluate epigenetic 101 effects of low-dose BPA in human breast epithelial cells. As a result of 102 chronic exposure to BPA, activation of estrogen receptor (ER)103 mediated signaling and subsequent alterations of responsive gene 104 expression were observed in the differentiated epithelial progeny. 105 This heritable influence on gene expression was similarly observed in 106 ER-positive breast tumors. 107 Materials and methods 108 Tissue samples and cell culture. Breast tissues, obtained from indi109 viduals undergoing mastectomy or reduction mammoplasty, were 110 collected in accordance with the protocols approved by the Institutional 111 Review Boards of the Ohio State University and the National Taiwan 112 University Hospital. For isolation of breast progenitor cells, non113 cancerous tissues (age: 1742 years old) were enzymatically dissoci114 ated via collagenase digestion as described previously (Hsu et al., 2009). 115 Single cells were grown into floating spherical colonies (210,000 cells 116 per colony), called mammospheres, in ultra-low attachment dishes 117 (Corning, Lowell, MA) in serum-free medium. These mammospheres, 118 enriched in breast progenitor cells (Dontu et al., 2004), were exposed 119 with BPA (4 nM) (Sigma, St. Louis, MO), diethylstilbestrol (DES, 70 nM), 120 daidzein (10 M), 1,3,5-tris(4-hydroxyphenyl)-4-propyl-1H-pyrazole 121 (PPT, 0.1 nM), 4-nonylphenol (NP, 1 M), N-butyl-benzyl phthalate 122 (BBP, 10 M), di(2-ethylhexyl)-phthalate (DEHP, 10 M), 4,4-dichloro123 biphnyl (PCB, 0.1 nM) or DMSO in phenol red-free medium for 3 weeks 124 (medium changed twice a week). The concentration of each chemical 125 was selected based on the literature review. Cell viability assay 126 indicated that there is no toxicity effect of each compound under the 127 concentration we selected (HarrEus et al., 2002; Rohrdanz et al., 2002; 128 Buteau-Lozano et al., 2008). After the exposure, mammospheres were 129 washed with PBS to remove BPA and then placed on a collagen-coated 130 dish in phenol red-free DMEM/F12 medium containing 5% charcoal131 dextran-treated FBS (Hyclone, Waltham, MA) for 23 weeks. Under 132 this condition, progenitor cells were differentiated into breast epithelial 133 cells, or called mammosphere-derived epithelial cells (MDECs). A panel 134 of 48 breast cancer cell lines, procured through the Integrative Cancer 135 Biology Program of the National Cancer Institute, were obtained from 136 the American Type Culture Collection (ATCC, Manassas, VA) and 137 routinely propagated in culture dishes for epigenetic analyses. 138 Immunofluorescence staining. Approximately 5,000 MDECs were 139 seeded on a collagen I-coated coverslip (BD Biosciences, San Jose, 140 CA) for overnight and then fixed with 4% paraformaldehyde for 141 15 min and permeabilized with 0.1% Triton X-100 for 10 min. After 142 blocking with 3% bovine serum albumin (Fisher Scientific, Pittsburgh, 143 PA) for 1 h, the coverslip was incubated with anti-ER antibody (D- 12, 1:50) (Santa Cruz Biotechnology, Santa Cruz, CA) overnight at 4 C. 144 The corresponding secondary FITC-conjugated antibody (Invitrogen, 145 Carlsbad, CA) was applied followed by 4,6-diamidino-2-phenylindole 146 staining (DAPI) (Invitrogen) to localize cell nuclei. The images were 147 captured by confocal laser microscope (Zeiss LSM510) (Zeiss, 148 Thornwood, NY), and percentages of ER subcellular localization 149 were calculated in 10 different optical fields (~ 100 cells) by two 150 independent researchers. 151 Western blot analysis. MDECs preexposed to BPA or DMSO were 152 collected and protein lysates were made. Lysates (30 g) were 153 immunoblotted with antibody against phospho-p42/44 MAPK 154 (1:1000), phospho-Akt (1:2000) (Cell Signaling Technology). 155 GAPDH (Santa Cruz Biotechnology) was used as loading control. 156 Cy5-conjugated goat anti-rabbit and Cy3-conjugated goat anti-mouse 157 antibody (GE Healthcare, Pittsburgh, PA) were used for multiplex 158 detection. The membranes were scanned by Typhoon 9400 scanner 159 (GE Healthcare). 160 Gene expression microarray. Total RNAs of 10 independent MDECs, 161 including BPA-preexposed and control, were isolated with Trizol 162 (Invitrogen) according to the manufacturer's instructions. RNA (5 g/ 163 sample) was used for microarray hybridization to the Affymetrix 164 Human Genome U133 Plus 2.0 Arrays (Affymetrix, Santa Clara, CA) by 165 the Microarray Core Facility at the Ohio State University Comprehen- 166 sive Cancer Center (Columbus, OH). Gene expression estimates of the 167 54,675 probe sets on arrays were obtained using robust multi-array 168 analysis (RMA) method with quantile normalization and background 169 correction (Irizarry et al., 2003). Gene expression microarray data of 48 170 breast cancer cell lines (BCC48, Neve et al., 2006) and breast tumors 171 (GSE2109, International Genomics Consortium, http://www.intgen. 172 org/expo) were available for downloading. Quantile normalization 173 and background correction were also applied to these individual 174 datasets. 175 Comparison between BPA-preexposed and control samples was 176 performed using BRB Array Tools software developed by the Biometric 177 Research Branch of the National Cancer Institute (http://www.linus. 178 nci.nih.gov/BRB-ArrayTools). A paired t-test with random variance 179 model was applied in order to identify differentially expressed genes 180 between BPA-preexposed and control samples. Initial filtering was 181 performed by selecting genes with P 0.05 and by removing genes 182 with lower expression for all samples (genes with expression values 183 less than or equal to 100 for all samples were removed). An un-paired 184 t-test was applied for the BCC48 and GSE2109 datasets in order to 185 identify differentially expressed genes between ER-positive and 186 ER-negative samples. The final 170 loci were obtained by restricting 187 the initial list of candidates to commonly expressed genes in breast 188 cancer based on two microarray datasets, BCC48 and GSE2109. 189 Functional and network analyses of these genes were performed 190 using Ingenuity Systems' IPA software (Ingenuity Systems Inc., www. 191 ingenuity.com). 192 Epigenetic treatments and reverse transcription-quantitative PCR (RT-qPCR). 193 MCF-7 breast cancer cells were treated with 1 M 5-aza-2-deoxycytidine 194 (DAC) in phenol red-free MEM containing 10% FBS and 6 ng/ml insulin. 195 During the final 24 h, some cells were additionally treated with 0.5 M 196 trichostatin A (TSA). RNA (1 g) was isolated and reversely transcribed 197 into cDNA using the SuperScript III Reverse Transcriptase (Invitrogen). 198 RT-qPCR was performed by using 2 SYBR Green Master Mix (Applied 199 Biosystems, Foster City, CA) on a 7500 Real-Time PCR System apparatus 200 (Applied Biosystems). Levels of the 36B4 mRNA transcript were also 201 measured as internal controls (Akamine et al., 2007). The reactions were 202 performed in triplicate, and the standard deviation was calculated using 203 the Comparative Method (ABI Prism 7700 Sequence Detection System 204 User Bulletin #2). Primer sequences and conditions for amplification are 205 available in Supplemental Table S1. 206 Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 3 Fig. 1. Preexposure of MDECs to bisphenol A (BPA) and immunofluorescence analysis of nuclear ER. (A) Subcellular localization of ER in MDECs on acute BPA treatment. MDECs were exposed to BPA (1000 nM) for the indicated time periods. The observed translocation of ER protein (green) from the cytoplasm to nucleus is indicative of functional estrogen signaling. Nuclei were stained with DAPI (blue). Bar = 10 m. (B) Mammospheres were treated with BPA 4 nM or DMSO for 3 weeks. After the exposure, mammospheres were washed with PBS to remove BPA and then placed on a collagen coated dishes for differentiation. Immunofluorescence staining showed that translocation of ER protein (green) from the cytoplasm to nucleus was observed. Nuclei were stained with DAPI (blue). Bar = 10 m. (C) Increased internalization of ER in BPA-preexposed MDECs (#124). Mammospheres were treated with DMSO, BPA (4 nM) or E2 (70 nM) for 3 weeks. The distribution of ER in mammospheres was monitored each week as indicated in the bottom graphic (7, 14, and 21 days). After the exposure, BPA was removed, and progenitor cells underwent epithelial differentiation in the collagen-coated dishes for 3 weeks. ER localization was also monitored as indicated in figure (42 days). Yellow bars indicate the percentage of ER-negative cells within the total population. Green bars (nuclear ER) and red bars (cytoplasmic ER) represent ER-positive cells within total population. (D) Increased nuclear localization of ER in BPA-preexposed MDECs. After the preexposure to BPA (4 nM) or DMSO, MDECs were subjected to immunofluorescence analysis. (CD) The percentage of subcellular localization of ER-positive cells, independently scored by two researchers, is shown. These results were collected from nine independent sets of MDECs samples. *, indicates samples were also subjected to gene expression analysis. (E) BPA induced p42/44 phosphorylation. Phosphorylated levels of p42/44 MAPK and Akt were analyzed in MDECs preexposed to BPA (4 nM) or DMSO by western blotting. GAPDH was used as loading control. 207 DNA methylation analysis by Pyrosequencing. To determine meth208 ylation levels of candidate genes in samples, the Pyrosequencing 209 system (Qiagen, Valencia, CA) was used to detect methylated CpG sites in sequencing reactions (Tost and Gut, 2007). Genomic DNA 210 (500 ng) was treated with sodium bisulfite using the EZ DNA 211 Methylation kit (Zymo Research, Orange, CA). Bisulfite-treated DNA 212 Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 4 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 5 23210987654321 was amplified with specific primers for each gene of interest. The Pyro 231 Mark Assay Design program and the Pyro Q-CpG software were used 232 for primer designs and data analysis, respectively. Average methyl233 ation levels of individual CpG sites for each DNA sample were 234 calculated. Murphy et al., 2009). In this regard, we observed an increased level 287 of phospho-p42/44 MAPK, likely attributed to this internalization in 288 BPA-preexposed MDECs without further ligand stimulation (Fig. 1E). 289 Effect of low-dose BPA on differential gene expression in MDECs 290 235 Statistical analysis. All data derived from subcellular localization, 236 RT-qPCR, and Pyrosequencing were presented as mean SD of n 237 independent measurements. Statistical comparisons between two 238 groups (DMSO vs. BPA) were made by Student's t-test using GraphPad 239 Prism 6 (GraphPad Software, La Jolla, CA). To avoid any violation of 240 normal distribution assumption for DNA methylation analysis, we 241 applied non-parametric MannWhitney rank-sum test (GraphPad 242 Prism 6). A significance was assigned if P b 0.05. 243 Results 244 Effect of low-dose BPA on the nuclear localization of ER in MDECs 245 Environmental chemicals, such as BPA, are known to act as estro246 genic ligands that activate or deactivate gene transcription in breast 247 epithelial cells (Soto et al., 2006; Dairkee et al., 2008). To determine 248 whether BPA causes an estrogen-like effect, we performed immuno249 fluorescence analyses in MDECs (un-exposed) transiently treated with 250 different doses (ranges: 11,000 nM) of BPA at 0, 5, 30, 60, and 120 min 251 (Fig. 1A). BPA, as a weak estrogenic ligand, caused maximized ER 252 internalization at 30 min in a higher dose (1,000 nM) of exposure. 253 Because prolonged exposure of breast progenitor cells to xenoestro254 gens also causes ER internalization in their differentiated progeny 255 (Hsu et al., 2009), we determined whether BPA has this effect. 256 Progenitor-containing mammospheres from an individual (#124) 257 were continuously exposed to 4 nM BPA for 3 weeks. After the exposure, 258 BPA was removed, and progenitor cells underwent epithelial differen259 tiation in the collagen-coated dishes for 23 weeks. Immunofluores260 cence analysis showed an increase of ER-positive population during 261 the mammosphere and MDEC stages (Fig. 1C). After 7-day preexposure, 262 the majority (90%) of mammospheres were ER-negative (yellow bar 263 in Fig. 1C). However, cell lineages were greatly shifted from ER264 negative to ER-positive (green-plus-red bar, 80%) after 42-day 265 incubation. Among ER-positive cells at 42 days, nuclear expression of 266 ER (green bar/green-plus-red bar, 78%) in BPA-preexposed MDECs 267 was increased compared to that of control (DMSO) cells (11.6%), 268 suggesting that BPA preexposure contributes to ER internalization in 269 MDECs (Fig. 1B). As a control, we also observed similar effect in E2 270 (70 nM)-preexposed MDECs. 271 When the analysis was extended to different primary MDECs 272 (n = 9), we noticed individual variations in response to this low-dose 273 BPA preexposure. As shown in Fig. 1D, five (#96, 124, 98, 99, and 117) 274 of these MDEC sets exhibited greater effects (up to 80%) of ER 275 internalization compared to the other four sets (#111, 120, 113, and 276 119) showing lesser effects (1840%). This initial result suggests that 277 as a weak estrogenic ligand, high-dose BPA (at least 1000 nM) is 278 needed to acutely activate ER-mediated signaling while chronic 279 exposure of a lower dose (4 nM) can similarly bring about this signal 280 transduction in breast epithelial cells. Furthermore, our observations 281 indicate that the genetic background of individuals may influence 282 differential responses to the exposure of low-dose BPA. 283 It is known that other exogenous stimulants, such as growth 284 factors, may act through mitogen-activated protein kinase (MAPK) or 285 Akt pathways to promote the nuclear internalization of ER for 286 transcriptional regulation in proliferating cells (Lannigan, 2003; To investigate whether this effect altered gene expression, we 291 conducted microarray analysis in 10 sets of preexposed (BPA, 4 nM) 292 and control MDECs using the Affymetrix Human Genome U133 Plus 293 2.0 Array. One set of the MDECs (#119) was removed from gene 294 expression analysis because of the low nuclear localization of ER. 295 Differential expression of genes at P b 0.05 within 9 set samples was 296 scored, yielding a total of 2,234 candidate loci (1,162 down-regulated 297 and 1,072 up-regulated) likely influenced by this BPA preexposure. 298 Scatter plots and the number of differentially expressed genes for 299 individual MDECs are presented in Fig. 2. Consistent with the 300 observation of ER internalization, we observed individual variations 301 of gene expression in these primary MDECs preexposed to low-dose 302 BPA. In this regard, greater numbers of differentially expressed genes 303 were seen in #124, 99, 100, and 128 (i.e., the high-responder group) 304 while the rest of six primary MDECs had fewer changes of expression 305 (i.e., the low-responder group). We additionally compared these 306 expression profiles with the status of ER internalization available for 307 seven MDEC sets (#124, 99, 117, 120, 111, 113 and 119). Though not 308 statistically significant, we observed a general trend that greater 309 degrees of ER internalization seemed to be associated with increased 310 numbers of differentially expressed genes in MDECs. 311 Effect of BPA-influenced gene signatures in ER-positive breast cancer 312 In silico analysis was conducted to determine whether specific 313 expression profiles of BPA-influenced genes are associated with the 314 development of breast cancer. When the 2,234 candidate loci were 315 compared with those of two microarray datasets, BCC48 (Neve et al., 316 2006) and GSE2109 (International Genomics Consortium, http:// 317 www.intgen.org/expo), we found a total of 170 BPA-influenced genes 318 (57 up-regulated and 113 down-regulated), the aberrant expression 319 of which may contribute to breast tumorigenesis (Fig. 3A and B; see 320 also Supplemental Table S2). Hierarchical clustering of 48 breast 321 cancer cell lines (i.e., BCC48) and 244 breast tumors (i.e., GSE2109) 322 revealed that specific up- and down-regulated patterns of these 170 323 genes are distinctly related to ER-positive cell lines (Fig. 3C and D). 324 This observation further indicates that 1) BPA may aberrantly regulate 325 gene expression through an ER-dependent pathway and 2) this 326 regulatory mechanism may be epigenetically imprinted in ER- 327 positive breast cancer. 328 To validate this potential imprinting effect, we choose 15 down- 329 regulated genes for expression analysis (Fig. 4). The reason to focus on 330 these loci was that the BPA-influenced repression might be associated 331 with hypermethylation of their CpG islands, which are located in the 332 transcription start sites of these selected genes. First, RT-qPCR was used 333 to confirm the expression status of these loci in the aforementioned six 334 MDECs preexposed to BPA (4 nM). The expression of these loci was 335 consistently down-regulated in three high-responders, #124, 99, and 336 128 (P b 0.05). Down-regulation of these loci, however, could not be 337 confirmed in one high-responder (#100), likely attributed to a small 338 sampling of down-regulated loci. Though this down-regulation was also 339 seen in the low-responder group by the sensitive RT-qPCR assay, the 340 repressive effect was usually less apparent (e.g., #120 and 113). In the 341 rest of low-responders (#111, 129, 117, and 119), significant changes of 342 expression between pre-exposed and control MDECs were not noted. 343 Q1 Fig. 2. Gene expression profiling of ten sets of the MDECs. A total of genes in control (DMSO) and BPA-preexposed cells are shown in the scatter plot. The number of significant down- regulated (, green dots) and up-regulated (, red dots) genes (2 fold differences between DMSO vs. BPA) are shown below the ID#. *, samples were also analyzed for ER nuclear localization. Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 6 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 7 344 Epigenetic repression of a BPA-influenced locus, LAMP3, in ER-positive 345 breast cancer cells 346 To further investigate a potential role of epigenetic repression, we 347 focused the expression analysis on a candidate gene, lysosomal348 associated membrane protein 3 (LAMP3), in the well-characterized 349 ER-positive MCF-7 cell line (see functional analysis of this gene in 350 MCF-7 cells in supplemental Figure S1). A low level of LAMP3 351 expression was detected in MCF-7 cells. To determine whether this 352 reduced expression is mediated by epigenetic mechanisms, we 353 treated these cells with the demethylating agent DAC (1 M) and/or 354 the histone deacetylase inhibitor TSA (0.5 M), known to reactivate 355 epigenetically repressed genes (Dworkin et al., 2009; Huang et al., 356 2009). As shown in Fig. 5A (lanes 14), the expression of LAMP3 was 357 significantly reactivated by single treatments (i.e., DAC or TSA, 358 P b 0.01). Furthermore, synergistic re-expression of this gene was 359 observed in cells with the combined treatment (DAC plus TSA, 360 P b 0.001). Additional results of seven other repressed genes are 361 presented in supplemental Figure S2. 362 To investigate whether this epigenetic repression could be 363 attributed to an estrogen-mediated pathway, MCF-7 cells were 364 additionally treated with E2 and/or an ER antagonist, ICI182780. 365 The subsequent E2 treatment led to re-silencing of LAMP3, suggesting 366 a role of estrogen signaling in mediating this epigenetic repression 367 (Fig. 5A, lanes 58). Treatment of ICI182780 abolished the down368 regulation, additionally indicating that this regulation is partly 369 mediated through an ER-dependent pathway (Fig. 5A, lane 9). The 370 repression was partially attenuated in the presence of additional 371 epigenetic treatments (i.e., DAC and TSA, lanes 1012). 372 Based on the results of these pharmacological experiments, our 373 observations suggest that 1) estrogen signaling initiates the repression 374 of the BPA-influenced loci in breast epithelial cells; 2) this repression is 375 partly mediated trough an ER-dependent pathway; and 3) persistent 376 repression of the BPA-influenced loci in cancer cells may be further 377 maintained by DNA methylation and histone modifications. 378 LAMP3 repression in MDECs preexposed to other estrogen-like chemicals 379 To determine whether long-term exposure of other estrogen-like 380 chemicals can additionally initiate this epigenetic repression, mam381 mospheres were exposed to diethylstilbestrol (DES, 70 nM), daidzein 382 (10 M), 1,3,5-tris(4-hydroxyphenyl)-4-propyl-1H-pyrazole (PPT, 383 0.1 nM), 4-nonylphenol (NP, 1 M), N-butyl-benzyl phthalate (BBP, 384 10 M), di(2-ethylhexyl)-phthalate (DEHP, 10 M), and 4,4-dichloro385 biphnyl (PCB, 0.1 nM) for 3 weeks. After the exposure, MDECs were 386 subjected to RT-qPCR analysis for LAMP3 expression. As shown in 387 Fig. 5B, downregulation of LAMP3 was confirmed in MDECs pre388 exposed to these estrogen-like chemicals (2.3 to 12.5-fold decrease). 389 Suppressive effects varied for the different environmental exposures, 390 indicating differential sensitivity of progenitors to these chemicals. 391 Promoter hypermethylation of LAMP3 in ER-positive breast cancer 392 To confirm the in vitro epigenetic findings, we conducted DNA 393 methylation analysis in the promoter CpG island regions of LAMP3 394 loci, in 48 breast cancer cell lines, 484 primary breast tumors (Taiwan cohort, n = 336; US cohort, n = 148), and 10 noncancerous breast 395 tissues as normal controls. Pyrosequencing analysis of LAMP3 (9 CpG 396 sites) revealed that DNA methylation levels were significantly 397 increased in breast cancer cell lines relative to those of normal 398 controls (Supplemental Figure S3B). Moreover, promoter hyper- 399 methylation of LAMP3 (P = 0.008) was significantly associated with 400 the ER-positive status. In close agreement with these results, 401 hypermethylation of LAMP3 was observed in ER-positive tumors in 402 the US cohort (P b 0.0001) (Fig. 6A, and Supplemental Figure S3A) and 403 in the Taiwan cohort (Fig. 6B, and Supplemental S3A). The cut-off 404 points of age groups used in the further analysis were based on 405 menopausal status--premenopausal (age 50 years) and postmeno- 406 pausal (age 50 years) groups. The young age group defined by age 407 35 years appears to have distinct biological characteristics and 408 display poor prognosis compared to those 35 years. Interestingly, 409 while the hypermethylation event occurred in both age groups (35 410 50 and N50 years old) in the US cohort, this trend was only seen in the 411 old age group (N50 years old) of the Taiwan cohort. Association of this 412 hypermethylation with other clinicopathological features of patients 413 was not apparent. 414 Discussion 415 When acutely exposed to estrogenic ligands, signal transduction is 416 mediated in part through nuclear hormone receptors, such as ER 417 (Bjornstrom and Sjoberg, 2005). We have previously shown that the 418 hallmark of this signal transduction is the translocation of cytoplasmic 419 ER into the nucleus of a normal breast epithelial cell (Hsu et al., 2009). 420 Unlike E2 and DES, BPA is considered to be a weak estrogenic ligand 421 based on the present immunofluorescence analysis and previous 422 receptor binding assays (Okada et al., 2008). In our case, up to 1000 423 nM BPA is needed to initiate this ligand-dependent function, which 424 mobilizes ER into the nucleus for transcriptional activation and 425 deactivation (Bjornstrom and Sjoberg, 2005; Okada et al., 2008). 426 We also observed that long-term exposure of breast progenitor 427 cells to low-dose BPA (4 nM) is capable of triggering ER internal- 428 ization later observed in the differentiated progeny. In this case, 429 exogenous stimulants (e.g., growth factors) may elicit ligand- 430 independent activation by promoting the nuclear internalization of 431 phosphorylated ER for transcriptional regulation in proliferating 432 cells (Lannigan, 2003; Murphy et al., 2009). This ligand-independent 433 genomic function likely co-regulates a subset of target genes (e.g., 434 LAMP3) governed through the ligand-dependent pathway. We 435 speculate that persistent exposure of progenitor cells to low-dose 436 PBA likely renders a permanent alteration of their differentiated 437 transcriptomes that are maintained by epigenetic mechanisms. 438 Deacetylated modifications of histone and promoter hypermethyla- 439 tion are heritably established in an inactive gene while acetylated 440 histone and promoter hypomethylation may be present to mark an 441 active locus (Jones and Baylin, 2007; Vaissiere et al., 2008). However, 442 individuals may have different susceptibility to these epigenetic 443 modifications. Based on our expression profile analysis of primary 444 MDECs, the high-responder group is more sensitive than the low- 445 responder group to the BPA preexposure. Whereas genetic variations 446 in response to xenoestrogens are well documented in different strains 447 of mice or rats (Richter et al., 2007; Tyl, 2009), this study provides the 448 Fig. 3. Comparison of gene expression profiles in BPA-preexposed MDECs, 48 breast cancer cell lines (Neve et al. 2006) and breast tumor samples (GSE2109). (A) Venn diagram showing 170 common genes identified in separate analyses of the three different data sets, i.e., MDECs (2234 genes altered by BPA), breast cancer cell lines (2564 genes preferential difference between ER-positive and ER-negative), and primary tumor samples (7848 genes preferential difference between ER-positive and ER-negative). Gene tree cluster analysis was performed on the 170 genes altered by BPA in MDECs (B), ER-stratified cell line (C) and patient tumor data sets (D). This analysis identified 113 repressed (panel B, group I) and 57 activated (panel B, group II) genes in BPA-preexposed MDECs. Group I genes were also identified in ER-positive breast cancer cell lines (panel C) and primary tumor samples (panel D); group II genes were identified in ER-negative breast cancer cell lines (panel C) and primary tumor samples (panel D). Color bar, magnitude of gene expression; green, repression; red, stimulation. Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 8 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 9 mental Figure S4). The expression of some of these BPA-influenced 477 genes may be epigenetically imprinted in breast cancer cells. We were 478 able to validate one candidate gene, LAMP3, the promoter hyper- 479 methylation of which is preferentially linked to transcriptional 480 silencing in ER breast cancer cells. This gene is known to encode 481 proteins associated with cell mobility and adhesion, and its over- 482 expression is usually linked to invasiveness in cancer (Kanao et al., 483 2005). Since LAMP3 may not be epigenetically silenced in ER- 484 negative tumors, its aberrant expression could contribute to more 485 aggressive phenotypes in this type of breast cancer. The hypermethy- 486 lation finding of LAMP3 independently observed in MCF-7 cells and 487 primary breast tumors suggests that this epigenetic event can be 488 initiated in normal breast epithelial cells and then heritably passed on 489 to cancer cells during the course of malignant progression. We further 490 speculate that DNA methylation of LAMP3 is potentially acquired as a 491 result of long-term exposure of progenitor cells to BPA and other 492 xenoestrogens. 493 Interestingly, promoter hypermethylation of this locus was found 494 to be associated with older ER-positive breast patients in the US 495 cohort and Taiwan cohort. However, the levels of DNA methylation 496 distribution showed significantly differences between age groups. 497 This epigenetic disparity could be attributed in part to the geograph- 498 ical differences of breast cancer incidence in these cohorts. Compared 499 to the US patient population, there has been an increased trend of 500 ER-positive young breast cancers (b50 years old) in Taiwan (Lin et 501 al., 2009). Future population study is needed to additionally 502 determine whether different exposure history of BPA and other 503 related chemicals contribute to this epigenetic disparity in the two 504 patient populations. 505 Conclusions 506 Fig. 5. Epigenetic reactivation of ER-mediated LAMP3 repression in MCF-7. (A) MCF-7 were treated with DAC (1 M), TSA (1 M) and/or ER antagonist, ICI182780 (ICI, 1 M) 6 h before E2 stimulation. Total RNA was subjected to RT-qPCR analysis. 36B4 was used as internal control. Mean SD; **P b 0.01, ***P b 0.001 compared with DMSO treated control (lanes 14), E2 treated alone (lanes 58), and ICI +E2 alone (lanes 912). #Pb 0.001 compared lane 5 (E2 treatment) with lane 1 (DMSO). (B) Mammospheres were exposed to E2 (70 nM), diethylstilbestrol (DES, 70 nM), daidzein (10 M), 1,3,5-tris(4-hydroxyphenyl)-4-propyl-1H-pyrazole (PPT, 0.1 nM), 4-nonylphenol (NP, 1 M), N-butylbenzyl phthalate (BBP, 10 M), di(2-ethylhexyl)-phthalate (DEHP, 10 M), and 4,4dichloro-biphnyl (PCB, 0.1 nM) for 3 weeks. After the exposure, the MDECs were subjected to RT-qPCR analysis for LAMP3 expression. Mean SD; ***Pb 0.001 compared with DMSO treated control. ##Pb 0.01, ###Pb 0.001 compared with BPA treated sample. In the present study, we have shown that the mammosphere 507 exposure system is a valuable tool for validation studies of BPA 508 findings based on animal models. We observed heritable effects of 509 low-dose BPA on the nuclear localization of ER and differential gene 510 expression in primary MDECs. Long-term exposure of breast progen- 511 itor cells to BPA may promote ligand-independent ER actions in 512 differentiated progeny. Furthermore, genetic variations of individuals 513 may contribute to differential susceptibility of breast epithelial cells to 514 the environmental exposure. We have also identified 170 BPA- 515 influenced genes that likely play a role in the development of ER- 516 positive breast cancer. These loci are potential biomarkers for 517 assessing the risk of developing breast cancer from exposure to 518 other environmental chemicals. 519 Conflict of interest statement 520 4656543210987 first evidence that differential susceptibility to low-dose BPA 467 exposure may also be present in human populations. 468 It has been observed that exposure to low-dose BPA during early 469 stages of mammary gland development may increase the risk of 470 developing breast neoplasm in adult animals (Durando et al., 2007; 471 Murray et al., 2007). Supporting this finding, we have identified 170 472 human candidate genes that may play a critical role in tumorigenesis. 473 Ingenuity pathway analysis has uncovered their putative functions 474 primarily related to aminoacyl-tRNA biosynthesis, circadian rhythm 475 signaling, GM-CSF signaling, and HER-2 signaling, the deregulation of 476 which can promote the development of breast cancer (see Supple- All authors declare that they have no competing financial interests. 521 Acknowledgments 522 This work was supported by the National Institutes of Health [U01 523 ES015986, R01 CA069065, and R01 ES017594 to T.H.-M.H]. We thank 524 Jeff Apostolos for his technical assistance. 525 Appendix A. Supplementary data 526 Supplementary data associated with this article can be found, in 527 the online version, at doi:10.1016/j.taap.2010.07.014. 528 Fig. 4. Validation of differentially expressed loci by RT-qPCR. Gene-specific RT-qPCR on 10 independent sets of DMSO and BPA-preexposed MDECs was conducted to validate 15 down-regulated loci. Data were analyzed by Ct method using 36B4 as the internal control and are presented relative to DMSO treatment for each MDEC sample. Mean SD (n = 3). *P b 0.05 (Student's t-test) for down-regulated genes compared with DMSO treated control. Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 10 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx Fig. 6. DNA methylation analysis of LAMP3. Quantitative methylation profiles of tumor samples from the US and Taiwan cohorts are shown in Supplemental Figure S3A. (A) Box plots indicate that the level of LAMP3 promoter methylation is positively correlated with ER status in primary tumors from the US cohort (left panel). A positive correlation between LAMP3 methylation and ER status is also observed in patient age 3550 years, and N50 years (right panel). (B) Box plots indicate that the level of LAMP3 promoter methylation is positively correlated with ER status in primary tumors from the Taiwan cohort (left panel). Further analysis shows a positive correlation between LAMP3 methylation and ER status is observed in the age N 50 years (right panel). 529 References 530 Akamine, R., Yamamoto, T., Watanabe, M., Yamazaki, N., Kataoka, M., Ishikawa, M., Ooie, 531 T., Baba, Y., Shinohara, Y., 2007. Usefulness of the 5 region of the cDNA encoding 532 acidic ribosomal phosphoprotein P0 conserved among rats, mice, and humans as a 533 standard probe for gene expression analysis in different tissues and animal species. 534 J. Biochem. Bioph. Methods 70, 481486. 535 Bjornstrom, L., Sjoberg, M., 2005. Mechanisms of estrogen receptor signaling: 536 convergence of genomic and nongenomic actions on target genes. Mol. Endocrinol. 537 19, 833842. 538 Buteau-Lozano, H., Velasco, G., Cristofari, M., Balaguer, P., Perrot-Applanat, M., 2008. 539 Xenoestrogens modulate vascular endothelial growth factor secretion in breast 540 cancer cells through an estrogen receptor-dependent mechanism. J. Endocrinol. 541 196, 399412. 542 Calafat, A.M., Ye, X., Wong, L.Y., Reidy, J.A., Needham, L.L., 2008. Exposure of the U.S. 543 population to bisphenol A and 4-tertiary-octylphenol: 20032004. Environ. Health 544 Perspect. 116, 3944. 545 CDC, 2009. Fourth National Report on Human Exposure to Environmental Chemicals. 546 Centers for Disease Control and Prevention; National Center for Environmental 547 Health; Division of Laboratory Sciences, Atlanta, GA. 548 Cheng, A.S., Culhane, A.C., Chan, M.W., Venkataramu, C.R., Ehrich, M., Nasir, A., 549 Rodriguez, B.A., Liu, J., Yan, P.S., Quackenbush, J., Nephew, K.P., Yeatman, T.J., Huang, 550 T.H., 2008. Epithelial progeny of estrogen-exposed breast progenitor cells display a 551 cancer-like methylome. Cancer Res. 68, 17861796. 552 Dairkee, S.H., Seok, J., Champion, S., Sayeed, A., Mindrinos, M., Xiao, W., Davis, R.W., 553 Goodson, W.H., 2008. Bisphenol A induces a profile of tumor aggressiveness in 554 high-risk cells from breast cancer patients. Cancer Res. 68, 20762080. 555 Dolinoy, D.C., Huang, D., Jirtle, R.L., 2007. Maternal nutrient supplementation counter556 acts bisphenol A-induced DNA hypomethylation in early development. Proc. Natl 557 Acad. Sci. USA 104, 1305613061. 558 Dontu, G., El-Ashry, D., Wicha, M.S., 2004. Breast cancer, stem/progenitor cells and the 559 estrogen receptor. Trends Endocrinol. Metab. 15, 193197. Durando, M., Kass, L., Piva, J., Sonnenschein, C., Soto, A.M., Luque, E.H., Munoz-de-Toro, M., 560 2007. Prenatal bisphenol A exposure induces preneoplastic lesions in the mammary 561 gland in Wistar rats. Environ. Health Perspect. 115, 8086. 562 Dworkin, A.M., Huang, T.H., Toland, A.E., 2009. Epigenetic alterations in the breast: 563 implications for breast cancer detection, prognosis and treatment. Semin. Cancer 564 Biol. 19, 165171. 565 Esteller, M., 2007. Cancer epigenomics: DNA methylomes and histone-modification 566 maps. Nat. Rev. Genet. 8, 286298. 567 Feinberg, A.P., Ohlsson, R., Henikoff, S., 2006. The epigenetic progenitor origin of human 568 cancer. Nat. Rev. Genet. 7, 2133. 569 Goodson, A., Robin, H., Summerfield, W., Cooper, I., 2004. Migration of bisphenol A from 570 can coatings--effects of damage, storage conditions and heating. Food Addit. 571 Contam. 21, 10151026. 572 HarrEus, U.A., Wallner, B.C., Kastenbauer, E.R., Kleinsasser, N.H., 2002. Genotoxicity and 573 cytotoxicity of 4-nonylphenol ethoxylate on lymphocytes as assessed by the comet 574 assay. Int. J. Environ. Anal. Chem. 82, 395401. 575 Ho, S.M., Tang, W.Y., Belmonte de Frausto, J., Prins, G.S., 2006. Developmental exposure 576 to estradiol and bisphenol A increases susceptibility to prostate carcinogenesis and 577 epigenetically regulates phosphodiesterase type 4 variant 4. Cancer Res. 66, 578 56245632. 579 Hsu, P.Y., Deatherage, D.E., Rodriguez, B.A., Liyanarachchi, S., Weng, Y.I., Zuo, T., Liu, J., 580 Cheng, A.S., Huang, T.H., 2009. Xenoestrogen-induced epigenetic repression of 581 microRNA-9-3 in breast epithelial cells. Cancer Res. 69, 59365945. 582 Huang, Y.W., Liu, J.C., Deatherage, D.E., Luo, J., Mutch, D.G., Goodfellow, P.J., Miller, D.S., 583 Huang, T.H., 2009. Epigenetic repression of microRNA-129-2 leads to over- 584 expression of SOX4 oncogene in endometrial cancer. Cancer Res. 69, 90389046. 585 Irizarry, R.A., Hobbs, B., Collin, F., Beazer-Barclay, Y.D., Antonellis, K.J., Scherf, U., Speed, 586 T.P., 2003. Exploration, normalization, and summaries of high density oligonucle- 587 otide array probe level data. Biostatistics 4, 249264. 588 Jones, P.A., Baylin, S.B., 2007. The epigenomics of cancer. Cell 128, 683692. 589 Kanao, H., Enomoto, T., Kimura, T., Fujita, M., Nakashima, R., Ueda, Y., Ueno, Y., 590 Miyatake, T., Yoshizaki, T., Buzard, G.S., Tanigami, A., Yoshino, K., Murata, Y., 2005. 591 Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014 Y.-I. Weng et al. / Toxicology and Applied Pharmacology xxx (2010) xxxxxx 11 592 Overexpression of LAMP3/TSC403/DC-LAMP promotes metastasis in uterine 593 cervical cancer. Cancer Res. 65, 86408645. 594 Kang, J.H., Kito, K., Kondo, F., 2003. Factors influencing the migration of bisphenol A 595 from cans. J. Food Prot. 66, 14441447. 596 Lannigan, D.A., 2003. Estrogen receptor phosphorylation. Steroids 68, 19. 597 Lin, C.H., Liau, J.Y., Lu, Y.S., Huang, C.S., Lee, W.C., Kuo, K.T., Shen, Y.C., Kuo, S.H., Lan, C., 598 Liu, J.M., Kuo, W.H., Chang, K.J., Cheng, A.L., 2009. Molecular subtypes of breast 599 cancer emerging in young women in Taiwan: evidence for more than just 600 westernization as a reason for the disease in Asia. Cancer Epidemiol. Biomarkers 601 Prev. 18, 18071814. 602 Markey, C.M., Luque, E.H., Munoz De Toro, M., Sonnenschein, C., Soto, A.M., 2001. In 603 utero exposure to bisphenol A alters the development and tissue organization of 604 the mouse mammary gland. Biol. Reprod. 65, 12151223. 605 Marotta, L.L., Polyak, K., 2009. Cancer stem cells: a model in the making. Curr. Opin. 606 Genet. Dev. 19, 4450. 607 Mountfort, K.A., Kelly, J., Jickells, S.M., Castle, L., 1997. Investigations into the potential 608 degradation of polycarbonate baby bottles during sterilization with consequent 609 release of bisphenol A. Food Addit. Contam. 14, 737740. 610 Munoz-de-Toro, M., Markey, C.M., Wadia, P.R., Luque, E.H., Rubin, B.S., Sonnenschein, C., 611 Soto, A.M., 2005. Perinatal exposure to bisphenol-A alters peripubertal mammary 612 gland development in mice. Endocrinology 146, 41384147. 613 Murphy, L.C., Skliris, G.P., Rowan, B.G., Al-Dhaheri, M., Williams, C., Penner, C., Troup, S., 614 Begic, S., Parisien, M., Watson, P.H., 2009. The relevance of phosphorylated forms of 615 estrogen receptor in human breast cancer in vivo. J. Steroid Biochem. Mol. Biol. 114, 616 9095. 617 Murray, T.J., Maffini, M.V., Ucci, A.A., Sonnenschein, C., Soto, A.M., 2007. Induction of 618 mammary gland ductal hyperplasias and carcinoma in situ following fetal 619 bisphenol A exposure. Reprod. Toxicol. 23, 383390. 620 Neve, R.M., Chin, K., Fridlyand, J., Yeh, J., Baehner, F.L., Fevr, T., Clark, L., Bayani, N., 621 Coppe, J.P., Tong, F., Speed, T., Spellman, P.T., DeVries, S., Lapuk, A., Wang, N.J., Kuo, 622 W.L., Stilwell, J.L., Pinkel, D., Albertson, D.G., Waldman, F.M., McCormick, F., 623 Dickson, R.B., Johnson, M.D., Lippman, M., Ethier, S., Gazdar, A., Gray, J.W., 2006. A 624 collection of breast cancer cell lines for the study of functionally distinct cancer 625 subtypes. Cancer Cell 10, 515527. 626 Nikaido, Y., Yoshizawa, K., Danbara, N., Tsujita-Kyutoku, M., Yuri, T., Uehara, N., Tsubura, 627 A., 2004. Effects of maternal xenoestrogen exposure on development of the 628 reproductive tract and mammary gland in female CD-1 mouse offspring. Reprod. 629 Toxicol. 18, 803811. 630 Ohm, J.E., Baylin, S.B., 2007. Stem cell chromatin patterns: an instructive mechanism for 631 DNA hypermethylation? Cell Cycle 6, 10401043. 672 Okada, H., Tokunaga, T., Liu, X., Takayanagi, S., Matsushima, A., Shimohigashi, Y., 2008. 632 Direct evidence revealing structural elements essential for the high binding ability 633 of bisphenol A to human estrogen-related receptor-gamma. Environ. Health 634 Perspect. 116, 3238. 635 Prins, G.S., Tang, W.Y., Belmonte, J., Ho, S.M., 2008. Developmental exposure to 636 bisphenol A increases prostate cancer susceptibility in adult rats: epigenetic mode 637 of action is implicated. Fertil. Steril. 89, e41. 638 Richter, C.A., Birnbaum, L.S., Farabollini, F., Newbold, R.R., Rubin, B.S., Talsness, C.E., 639 Vandenbergh, J.G., Walser-Kuntz, D.R., vom Saal, F.S., 2007. In vivo effects of 640 bisphenol A in laboratory rodent studies. Reprod. Toxicol. 24, 199224. 641 Rohrdanz, E., Ohler, S., Tran-Thi, Q.H., Kahl, R., 2002. The phytoestrogen daidzein 642 affects the antioxidant enzyme system of rat hepatoma H4IIE cells. J. Nutr. 132, 643 370375. 644 Soto, A.M., Maffini, M.V., Schaeberle, C.M., Sonnenschein, C., 2006. Strengths and 645 weaknesses of in vitro assays for estrogenic and androgenic activity. Best Pract. Res. 646 Clin. Endocrinol. Metab. 20, 1533. 647 Timms, B.G., Howdeshell, K.L., Barton, L., Bradley, S., Richter, C.A., vom Saal, F.S., 2005. 648 Estrogenic chemicals in plastic and oral contraceptives disrupt development of the 649 fetal mouse prostate and urethra. Proc. Natl Acad. Sci. USA 102, 70147019. 650 Tost, J., Gut, I.G., 2007. DNA methylation analysis by pyrosequencing. Nat. Protoc. 2, 651 22652275. 652 Tyl, R.W., 2009. Basic exploratory research versus guideline-compliant studies used for 653 hazard evaluation and risk assessment: bisphenol A as a case study. Environ. Health 654 Perspect. 117, 16441651. 655 Vaissiere, T., Sawan, C., Herceg, Z., 2008. Epigenetic interplay between histone 656 modifications and DNA methylation in gene silencing. Mutat. Res. 659, 4048. 657 Vandenberg, L.N., Maffini, M.V., Sonnenschein, C., Rubin, B.S., Soto, A.M., 2009. 658 Bisphenol-A and the great divide: a review of controversies in the field of 659 endocrine disruption. Endocr. Rev. 30, 7595. 660 Vogel, S.A., 2009. The politics of plastics: the making and unmaking of bisphenol a 661 "safety". Am. J. Public Health 99 (Suppl 3), S559566. 662 Welshons, W.V., Nagel, S.C., vom Saal, F.S., 2006. Large effects from small exposures: III. 663 Endocrine mechanisms mediating effects of bisphenol A at levels of human 664 exposure. Endocrinology 147, S5669. 665 Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C., 666 Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., Laird, P.W., 2007. Epigenetic 667 stem cell signature in cancer. Nat. Genet. 39, 157158. 668 Yaoi, T., Itoh, K., Nakamura, K., Ogi, H., Fujiwara, Y., Fushiki, S., 2008. Genome-wide 669 analysis of epigenomic alterations in fetal mouse forebrain after exposure to low 670 doses of bisphenol A. Biochem. Biophys. Res. Commun. 376, 563567. 671 Please cite this article as: Weng, Y.-I., et al., Epigenetic influences of low-dose bisphenol A in primary human breast epithelial cells, Toxicol. Appl. Pharmacol. (2010), doi:10.1016/j.taap.2010.07.014