Low Dose of Bisphenol A Modulates Ovarian Cancer Gene Expression Profile and Promotes Epithelial to Mesenchymal Transition Via Canonical Wnt Pathway

Low Dose of Bisphenol A Modulates Ovarian Cancer Gene Expression Profile and Promotes Epithelial... Abstract The xenoestrogen bisphenol A (BPA) is a synthetic endocrine disrupting chemical, having the potential to increase the risk of hormone-dependent ovarian cancer. Thus, a deeper understanding of the molecular and cellular mechanisms is urgently required in the novel cell models of ovarian cancer which express estrogen receptors. To understand the possible mechanisms underlying the effects of BPA, human ovarian adenocarcinoma SKOV3 cells were exposed to BPA (10 or 100 nM) or 0.1% DMSO for 24 h, and then global gene expression profile was determined by high-throughput RNA sequencing. Also, enrichment analysis was carried out to find out relevant functions and pathways within which differentially expressed genes were significantly enriched. Transcriptomic analysis revealed 94 differential expression genes. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses indicated that these genes related to tumorigenesis and metastasis. Further studies were carried out to validate the results of functional annotation, which indicated that BPA (10 and 100 nM) increased migration and invasion as well as induced epithelial to mesenchymal transitions in SKOV3 and A2780 cells. Accordingly, environmentally relevant-dose BPA activated the canonical Wnt signaling pathway. Our study first comprehensively analyzed the possible mechanisms underlying the effects of BPA on ovarian cancer. Environmentally relevant doses of BPA modulated the gene expression profile, promoted epithelial to mesenchymal transition progress via canonical Wnt signaling pathway of ovarian cancer. BPA, gene expression profile, ovarian cancer, metastasis, EMT, canonical Wnt pathway Endocrine disrupting chemicals (EDCs) are natural or man-made agents in the environment that could interfere with some aspect of hormone action in the body, thus cause disrupting effects on the endocrine system (Gore et al., 2015). Circulating levels of natural hormones may be strongly implicated in the risks of several reproductive system cancers. For example, the abnormal level of estrogen is related to the pathogenesis and development of ovarian (Rodriguez et al., 2001), breast (Missmer et al., 2004), cervical (Chung et al., 2010), and endometrial (Grady et al., 1995) tumors which are highly estrogen receptor (ER)-positive or estrogen-responsive. Hence, EDCs that have estrogenicity or estrogenic activity may be important risk factors of estrogen-responsive cancers. The xenoestrogen bisphenol A (BPA) is a synthetic EDC, having the potential to promote the progression of some estrogen-responsive tumors, such as ovarian (Seachrist et al., 2016), and breast (DeMatteo et al., 2013) tumors. BPA is extensively used as an ingredient in many household appliances, including epoxy resin linings of beverage containers and food cans, polycarbonate plastics, and dental sealants (Rochester 2013). What’s more, in the epoxy resin and polycarbonate, heating time and temperature can cause hydrolysis of ester bonds linked to BPA, thus BPA monomer leaching from these products, and contaminating the food (Kang et al., 2003). An epidemiological study has reported that in a population of 394 human subjects, up to 95% have detectable levels of BPA in serum, urine, and saliva owing to low but sustained exposure (Calafat et al., 2005). In a previous study, BPA was detected in 36 human ovarian follicular fluid aspirates in the range of nanomolar concentrations (Ikezuki et al., 2002). Determined from measurable concentrations of BPA in body fluids and tissues, it has been estimated that among the general population, the theoretical internal concentration is ranging from 10 to 100 nM (Vandenberg et al., 2007). Nowadays, adverse endocrine disruptive effects and hormone-dependent cancer formation caused by BPA have risen global concern. In vitro studies have found that BPA not only enhances proliferation and glycolysis-based metabolism (Ptak et al., 2011; Park et al., 2009; Shi et al., 2017), but also stimulates the cell metastasis of the ovarian carcinoma cells (Kim et al., 2015b; Ptak et al., 2014). However, most of these studies fasten on the cancer-promoting effects of BPA at high concentrations, thus neglecting the effects of low concentrations that more closely reflect environmental exposure levels. Moreover, due to the large variations in BPA dosage used, previous researches draw inconsistent conclusions on the transcriptional and apoptotic properties of BPA (Hwang et al., 2011; Ptak et al., 2011). The effect of BPA usually exhibits a U-shaped or an inverted U-shaped dose-response curve (Shi et al., 2017; Wang et al., 2013). Therefore, it is unwarranted to take putative low-dose biological activity for granted as high-dose BPA effects. Thus, the aim of the present study was to assess the cellular effects of environmentally related-dose BPA on ovarian cancer, and 10 and 100 nM were chosen as the concentrations of BPA in this study. The ovarian carcinoma is one the most commonly malignant tumors among women worldwide, owing to the phenomena that it’s always diagnosed in advanced stages and has intra-abdominal metastasis (Siegel et al., 2018). When the lesion is entirely localized to the ovary, the 5-year survival rate is >90%, but when the lesion is disseminated intra-abdominally, the incidence falls below 20% (Ptak et al., 2014). As a result, its death to incidence ratio is up to 60%–65% (National Academies of Sciences, Engineering, and Medicine, 2016). Previous studies have demonstrated that BPA plays a potential part in the progression of ovarian cancer (Kim et al., 2015b; Ptak et al., 2014). However, how BPA induces the progression of ovarian cancer remains largely undefined. Human ovarian carcinomas were more often ER-positive than normal ovary cells or benign tumors (Lantta 1984; Willcocks et al., 1983). Estrogen’s classical nuclear receptors, ERα/β, and membrane receptor G-protein-coupled estrogen receptor (GPER) have been detected in a variety of ovarian carcinoma cell lines (Hoffmann et al., 2017; Hwang et al., 2011). However, it still remains unknown whether exposure to BPA can cause abnormal estrogenic signaling mediated by these ERs. A deeper understanding of the molecular and cellular mechanisms is urgently required in the novel cell models of ovarian cancer which express ERs. Therefore, with the help of high-throughput sequencing techniques, the main objective of our research is to unravel the altered gene expression profile and identify molecular and cellular mechanisms mediating the effects of environmentally related-dose BPA exposure on ovarian cancer. Materials and Methods Chemicals and cell lines See “Chemicals and cell lines” in the Supplementary Material. RNA sequencing Human ovarian adenocarcinoma SKOV3 cells were treated with BPA (10 or 100 nM) or dimethyl sulfoxide (DMSO) for 24 h, and then cells were washed twice with phosphate-buffered saline, total RNA was extracted from cells using Trizol reagent (Sangon Biotech) according to the manufacturer’s instructions. RNA Sequencing was performed on an Illumina Hisreq 4000 platform with 150 bp paired-end reads at Novogene Bioinformatics Technology Co., Ltd (see “RNA Sequencing” in the Supplementary Material for additional details, including RNA extraction, Library construction and Illumina sequencing, and in silico gene expression analysis). RNA-seq data have been deposited in the ArrayExpress database at EMBL-EBI (www.ebi.ac.uk/arrayexpress; last accessed November 28, 2017) under accession number E-MTAB-6295. Functional annotation The shared differentially expressed genes (DEGs) of SKOV3 cells treated with 10-nM BPA and 100-nM BPA in comparison with DMSO-treatment SKOV3 cells were used for further functional annotation (see “Functional annotation” in the Supplementary Material for additional details). Cell viability The cell viability was determined using the CCK-8 kit assay (Beyotime) according to the manufacturer’s protocol (see “Cell viability” in the Supplementary Material for additional details of the CCK-8 assay procedure). Cell migration Wound-healing assay was used to assess cell migration by observing the ability of the cells to move into an acellular space (see “Cell migration” in the Supplementary Material for additional details of the wound-healing assay procedure). Cell invasion Transwell assay was performed to assess cell invasion ability. The details of transwell assay procedure see “Cell invasion” in the Supplementary Material. Cytoplasmic and nuclear protein extraction and Western blotting The cells were collected and lysed with cell lysis buffer (Beyotime). Cytoplasmic and nuclear protein of cultured ovarian cancer cells were separated by Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime) according to the manufacturer’s protocol. The details of Western blotting procedure and antibodies used in this study see “Western blotting” in the Supplementary Material. Immunofluorescence staining Immunofluorescence staining was performed to detect the translocation of β-catenin. The details of immunofluorescence staining procedure see “Immunofluorescence staining” in the Supplementary Material. Quantitative real-time PCR We performed qPCR to detect mRNA or microRNA expression levels (see “Quantitative real-time PCR” in the Supplementary Material for additional details). Primer sequences were provided in Supplementary Tables 1 and 2. Statistical analyses Statistical analyses were the same as our previous study (Lu et al., 2017) (see “Statistical analyses” in the Supplementary Material for additional details). RESULTS Confirmation of the Expression of ERs According to the human protein atlas database (Uhlen et al., 2017), classical nuclear receptors, ERα/β, membrane receptor GPER have been detected in ovarian cancer tissues (see Supplementary Figure 1). We confirmed the expression of ERα/β and GPER by qPCR and western blotting in SKOV3 and A2780 cells. As shown in Figure 1, SKOV3 cells were ERα/β and GPER expressing, A2780 cells were ERα negative and had lower basal ERβ and GPER protein levels. Figure 1. View largeDownload slide Expression of ERα/β, GPER mRNAs and proteins in SKOV3 and A2780 cells. A, ERs mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. B, Basal ERα/β and GPER protein expression in SKOV3 and A2780 cells. β-actin servers as the loading control for western blotting. Data are representative of 3 independent experiments. Figure 1. View largeDownload slide Expression of ERα/β, GPER mRNAs and proteins in SKOV3 and A2780 cells. A, ERs mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. B, Basal ERα/β and GPER protein expression in SKOV3 and A2780 cells. β-actin servers as the loading control for western blotting. Data are representative of 3 independent experiments. Identification of Gene Expression Changes Gene expression changes in SKOV3 cells after physiological-dose BPA treatment for 24 h were identified using the multiple-testing module of the Cuffdiff program in Cufflinks package. The genes with q-value (the false discovery rate (FDR)-adjusted p-value of the test statistic) < .05 were defined as DEGs. To visualize the gene expression profiles across all doses and replicates, the clustering analysis was performed, which showed that replicates of each group clustered closely together (Figs. 2B and 2C), indicating the quality of the gene expression data sets was high enough for further exploration. Furthermore, hierarchical clustering demonstated similarities in the gene expression profiles for 10- and 100-nM BPA treatment. This may be due to the fact that both doses were relatively low in physiological level and hence had similar gene expression alterations. Thus, the genes with q-value < .05 over these 2 doses were considered as DEGs, which increased the analysis stringency, namely, reduced false positives because of multiple testing. When compared with the control group, a total of 142 gene expressions were significantly altered in 10-nM BPA treated cells while 148 in 100-nM BPA treated cells. Among those genes identified in both groups, only 94 genes were found to have significant altered expression in both BPA dose groups. The Venn diagram showed the portion of shared DEGs after exposure of SKOV3 cells to either 10- or 100-nM BPA in SKOV3 cells (Figure 2D). Of the 94 genes meeting the criterion, 36 were found to be up-regulated, and 58 were down-regulated, which were listed in Supplementary Table 3. And we also applied a heatmap analysis of the shared DEGs (Figure 2A). In parallel with the clustering analysis results, the fold change of DEGs in the case of 10- and 100-nM BPA treatment showed no significant difference (Figure 2A; see Supplementary Table 3). Figure 2. View largeDownload slide Gene expression analysis after 24-h exposure to 0.1% DMSO (control), 10- or 100-nM BPA in SKOV3 cells with 3 biological replicates. A, Heatmap reflected expression profile for genes differently regulated (q-value < .05) over 2 BPA doses. B and C, B Dendrogram and C multi-dimensional scaling (MDS) plot generated by hierarchical clustering of SKOV3 gene expression. D, Venn diagram showed the overlap of DEGs induced by treatment with 10- and 100-nM BPA compared with the control group. Figure 2. View largeDownload slide Gene expression analysis after 24-h exposure to 0.1% DMSO (control), 10- or 100-nM BPA in SKOV3 cells with 3 biological replicates. A, Heatmap reflected expression profile for genes differently regulated (q-value < .05) over 2 BPA doses. B and C, B Dendrogram and C multi-dimensional scaling (MDS) plot generated by hierarchical clustering of SKOV3 gene expression. D, Venn diagram showed the overlap of DEGs induced by treatment with 10- and 100-nM BPA compared with the control group. Functional Annotation Analysis of the Shared DEGs Over 2 Doses To understand the biological significance of altered genes caused by BPA exposure, the DEGs were searched in the functional annotation Database for Annotation, Visualization and Integrated Discovery (DAVID) database and categorized by putative functions. For gene ontology (GO) analysis, transcripts were divided into 3 major categories: biological processes, cellular component, and molecular function. As shown in Figure 3, among the cluster of biological processes, it exhibited that the top 3 scores were extracellular matrix (ECM) organization (GO: 0030198), positive regulation of cell proliferation (GO: 0008284) and cellular response to interleukin-1 (GO: 0071347). In addition, the top 3 scores of cellular component were extracellular space (GO: 0005615), ECM (GO: 0031012), and extracellular region (GO: 0005576). Moreover, insulin-like growth factor binding (GO: 0005520), structural molecule activity (GO: 0005198) were the molecular functions most targeted by BPA treatment. The enriched DEGs of each GO term were listed in Supplementary Table 4. Figure 3. View largeDownload slide GO of altered genes. The expression profile of the entire list of the DEGs was used as input data in GO analysis using DAVID database. Annotated terms were classified into 3 major categories (biological process, cellular component, and molecular function) and 31 subgroups. The numbers on the right of bars indicated related “related DEGs count”/“pop hits” of each GO term. Figure 3. View largeDownload slide GO of altered genes. The expression profile of the entire list of the DEGs was used as input data in GO analysis using DAVID database. Annotated terms were classified into 3 major categories (biological process, cellular component, and molecular function) and 31 subgroups. The numbers on the right of bars indicated related “related DEGs count”/“pop hits” of each GO term. Then, using the shared DEGs over 2 doses as input data, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to visualize the effects of BPA exposure on specific cell signaling pathway-related gene expression changes, and screened for biological pathways that might be involved. The results, including representative pathways and the number of genes related to these pathways were shown in Figure 4. The enriched DEGs of each KEGG pathway were listed in Supplementary Table 5. Figure 4. View largeDownload slide KEGG pathway analysis of altered genes. The expression profile of the entire list of the DEGs was used as input data in KEGG pathway analysis using DAVID database. The numbers on the right of bars indicated “related DEGs count”/“pop hits” of each pathway. Figure 4. View largeDownload slide KEGG pathway analysis of altered genes. The expression profile of the entire list of the DEGs was used as input data in KEGG pathway analysis using DAVID database. The numbers on the right of bars indicated “related DEGs count”/“pop hits” of each pathway. Among the enriched KEGG pathways (Figure 4), the top score was microRNAs in cancer (hsa05206) (see Supplementary Figure 3), suggesting microRNAs such as miR-21, miR-221, miR-222, miR-19a, and miR-7 might be involved in tumorigenesis following BPA exposure. To confirm the functional annotation results and sequencing data, these expression changes of microRNAs were further detected by qPCR. Specifically, the expression of miR-21 and miR-222 were up-regulated with increasing doses of BPA exposure (10 and 100 nM) in SKOV3 and A2780 cells (see Figure 5). The SERPINB5 (maspin) mRNA is a known target of miR-21, and the mRNA encoding TIMP3 is miR-21 and miR-222’s common target. We therefore confirmed that BPA decreased the transcript levels of SERPINB5 (maspin) and TIMP3 in SKOV3 and A2780 cells (see Supplementary Figure 4), consistent with a BPA-mediated increase in miR-21 and miR-222 levels. Figure 5. View largeDownload slide MicroRNA expression in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. The expression of (A, C) miR-21 and (B, D) miR-222 were determined using qPCR. * p < .05, ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 5. View largeDownload slide MicroRNA expression in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. The expression of (A, C) miR-21 and (B, D) miR-222 were determined using qPCR. * p < .05, ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. These functional annotation results suggested that BPA exposure significantly modified genes related to tumorigenesis and metastasis in ovarian carcinoma cells. It is worth mentioning that genes related to biological processes such as positive regulation of cell proliferation (GO: 0008284), positive regulation of cysteine-type endopeptidase activity involved in the apoptotic process (GO: 0043280) (Figure 3; see Supplementary Table 4), and microRNAs in cancer (hsa05206) (Figure 4; see Supplementary Table 5) were found to be significantly representative in the final list of genes. Regarding to the regulation of carcinoma metastasis, the 94 shared DEGs were related to GO terms and KEGG pathways like ECM organization (GO: 0030198), ECM-receptor interaction (hsa04512) and positive regulation of cell migration (GO: 0030335), and these pathways included PI3K-Akt signaling pathway (hsa04151), hippo signaling pathway (hsa04390), p53 signaling pathway (hsa04115) (see Supplementary Figure 5), cellular response to transforming growth factor β (TGF-β) stimulus (GO: 0071560), TGF-β signaling pathway (hsa04350), and Wnt signaling pathway (hsa04310) (Figs. 3 and 4; see Supplementary Tables 4 and 5), indicating nanomolar-dose BPA exposure induced alteration of the genes expression was closely involved in proliferation and cancer development. Effects of BPA on OC Cell Proliferation To determine the effects of BPA on OC cell proliferation, we used 2 human ovarian adenocarcinoma cell lines SKOV3 and A2780. SKOV3 and A2780 cells were stimulated with series concentrations of BPA which were in the range of 10 to 4 × 105 nM for 24, 48, or 72 h, and then we performed CCK-8 kit assay to measure cell viability. As shown in Supplementary Figure 2A, compared with vehicle-treated cells, concentrations of BPA > 105 nM significantly reduced SKOV3 cell viability in a concentration-dependent and time-dependent manner. As for A2780 cells, BPA showed cytotoxicity at 104 nM after 24-h exposure (see Supplementary Figure 2B). Besides, BPA had no statistically significant effects on cell proliferation at 10- and 100-nM doses in both cell lines (Figure 6), which represented environmentally equivalent concentrations. Figure 6. View largeDownload slide Effects of BPA on proliferation of SKOV3 and A2780 cells. (A) SKOV3 and (B) A2780 cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24, 48, and 72 h, and then assessed by CCK-8 kit assay. ns, no statistical significance. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 6. View largeDownload slide Effects of BPA on proliferation of SKOV3 and A2780 cells. (A) SKOV3 and (B) A2780 cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24, 48, and 72 h, and then assessed by CCK-8 kit assay. ns, no statistical significance. Values are mean ± SEM. Data are representative of 3 independent experiments. Effect of Nanomolar-Dose BPA on OC Cell Migration and Invasion Tumor metastasis is one of the most life-threatening causes of ovarian cancer (Ptak et al., 2014). We further evaluated the effects of low-dose BPA (10 or 100 nM) on the migration and invasion ability of SKOV3 and A2780 cells in vitro. In parallel with functional annotation results, compared with the vehicle group, we found that stimulation with 10- or 100-nM BPA for 24 or 48 h accelerated wound closure in wound scratch assay (Figs. 7A and 7B), suggesting that BPA induced the migration of ovarian carcinoma cells. Furthermore, exposure to BPA led to significantly more invasive cells (Figs. 7C–F). Figure 7. View largeDownload slide Nanomolar concentrations of BPA induced migration and invasion of SKOV3 and A2780 cells. A and B, Representative images of the wound scratch assay in a 100× light microscope utilizing the A SKOV3 and B A2780 cell lines after scratching 24 and 48 h. Cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA. Effects of BPA on SKOV3 or A2780 cell invasion. Both SKOV3 and A2780 cells were exposed to 10- or 100-nM BPA for 24 h, and then measured using transwell assay with Matrigel after incubation for 18 h. Invasion cells were fixed, stained, photographed, and counted in 5 random views. C and E, Representative digital images of the transwell assay taken randomly by a 200× light microscope. D and F, The quantitative analysis of invaded cells after 24-h BPA exposure and 18-h incubation. ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 7. View largeDownload slide Nanomolar concentrations of BPA induced migration and invasion of SKOV3 and A2780 cells. A and B, Representative images of the wound scratch assay in a 100× light microscope utilizing the A SKOV3 and B A2780 cell lines after scratching 24 and 48 h. Cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA. Effects of BPA on SKOV3 or A2780 cell invasion. Both SKOV3 and A2780 cells were exposed to 10- or 100-nM BPA for 24 h, and then measured using transwell assay with Matrigel after incubation for 18 h. Invasion cells were fixed, stained, photographed, and counted in 5 random views. C and E, Representative digital images of the transwell assay taken randomly by a 200× light microscope. D and F, The quantitative analysis of invaded cells after 24-h BPA exposure and 18-h incubation. ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. BPA Exposure Triggers Epithelial to Mesenchymal Transitions In the functional annotation results, it is noteworthy that some of the DEGs including MMP7, BAMBI, FOSL1, MYC, and THBS1, significantly overrepresented pathways including Wnt signaling pathway (see Supplementary Figure 6) and TGF-β signaling pathway (see Supplementary Figure 7), which have been reported to collaborate to induce activation of epithelial to mesenchymal transition (EMT) program (Moustakas and Heldin, 2007). This suggested that BPA may promote ovarian cancer metastasis via induction of EMT. EMT has been thought to be the first step of cancer metastasis, with epithelial characteristic like ZO-1 is down-regulated, while mesenchymal characteristic like MMP9 is up-regulated. After stimulation with 10- or 100-nM BPA for 24 h, western blotting revealed that BPA exposure down-regulated the expression of epithelial marker ZO-1, while up-regulated the expression of mesenchymal marker MMP9 in a dose-dependent manner (Figs. 8A and 8B). And those gene expression changes were confirmed by qPCR (Figs. 8C–F). Compared with control cells, we observed statistically significant increase in MMP9 mRNA level and decrease in ZO-1 mRNA level after BPA exposure for 24 h. Figure 8. View largeDownload slide BPA treatment induced EMT. Cells were treated with 10- or 100-nM BPA for 24 h to induce the EMT. A and B, Western blotting showed that BPA induced a down-regulation of the epithelial maker ZO-1 and an up-regulation of the mesenchymal marker MMP9. GAPDH servers as a loading control. C–F, qPCR analysis of ZO-1 and MMP9 mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. * p < .05, **p < .01, ***p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 8. View largeDownload slide BPA treatment induced EMT. Cells were treated with 10- or 100-nM BPA for 24 h to induce the EMT. A and B, Western blotting showed that BPA induced a down-regulation of the epithelial maker ZO-1 and an up-regulation of the mesenchymal marker MMP9. GAPDH servers as a loading control. C–F, qPCR analysis of ZO-1 and MMP9 mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. * p < .05, **p < .01, ***p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. BPA Exposure Activated the Canonical Wnt Signaling Pathway The EMT process is mediated by some cellular signaling pathways, such as TGF-β, Wnt-β-catenin, receptor tyrosine kinases, Notch, Hedgehog pathways and microRNA network (Lee et al., 2017). TGF-β pathway is one the most well-described signaling pathways to trigger EMT process, which involves in diverse developmental processes and cellular functions (Gonzalez and Medici, 2014). However, as indicated in Figure 4, Wnt signaling pathway ranked higher than TGF-β signaling pathway. Additionally, although the relationship between BPA and the TGF-β signaling pathway has been reported (Kim et al., 2015a; Park and Choi, 2014), there is now a consensus that micromolar BPA, in fact, inhibits TGF-β signaling pathway in ovarian cancer cells. Therefore, we emphasized on studying the undefined relationship between BPA and the Wnt signaling pathway. To this end, we analyzed genetic changes related to Wnt signaling pathway by the GeneMANIA algorithm and identified the functional network of Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, and MYC) (Figure 9). Figure 9. View largeDownload slide Functional network of the Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, MYC). GeneMANIA retrieved known and predicted interactions between these genes and added extra genes (small gray circles) that are strongly connected to query genes (large circles). Figure 9. View largeDownload slide Functional network of the Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, MYC). GeneMANIA retrieved known and predicted interactions between these genes and added extra genes (small gray circles) that are strongly connected to query genes (large circles). So far, previous study indicated that Wnt signaling pathways include the canonical Wnt pathway, the noncanonical planar cell polarity pathway, and the noncanonical Wnt/calcium pathway (Arend et al., 2013). As shown in Supplementary Figure 6, the DEGs of MMP7 (Uterine), BAMBI, FOSL1 (fra-1), and MYC (c-myc) regulate the canonical Wnt pathway. Previous studies suggested that in the canonical Wnt pathway, β-catenin is the key mediator (Willert and Nusse, 1998), we therefore investigated the effects of BPA on nuclear translocation of β-catenin in SKOV3 and A2780 cells. After stimulation with 10- or 100-nM BPA for 24 h, SKOV3 as well as A2780 cells showed increased β-catenin translocation to the nucleus in a dose-dependent manner in western blotting (Figs. 10A and 10B), suggesting that BPA promoted ovarian cancer EMT through activation of canonical Wnt/β-catenin signaling. Immunofluorescence staining revealed the BPA-induced β-catenin translocation (Figs. 10C and 10D). In the control cells, β-catenin was expressed at a low level, and located mostly at the cytoplasm. After BPA exposure, the protein level of β-catenin was significantly elevated in the nucleus. This finding validated the results of functional annotation. Figure 10. View largeDownload slide BPA exposure activated the canonical Wnt signaling pathway in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. A and B, Western blotting of nuclear and cytoplasmic localization of β-catenin. Lamin A/C serves as a nuclear protein loading control, while GAPDH served as a cytoplasmic loading control. C and D, Immunofluorescence staining of β-catenin expression in nuclear and cytoplasm, demonstrating the nuclear translocation of β-catenin upon a 24-h BPA treatment. Data are representative of 3 independent experiments. Figure 10. View largeDownload slide BPA exposure activated the canonical Wnt signaling pathway in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. A and B, Western blotting of nuclear and cytoplasmic localization of β-catenin. Lamin A/C serves as a nuclear protein loading control, while GAPDH served as a cytoplasmic loading control. C and D, Immunofluorescence staining of β-catenin expression in nuclear and cytoplasm, demonstrating the nuclear translocation of β-catenin upon a 24-h BPA treatment. Data are representative of 3 independent experiments. DISCUSSION Up to now, most studies focused on the cancer-promoting effects of BPA at high concentrations (Fernandez et al., 2012; Hwang et al., 2011; Kang et al., 2013; Kim et al., 2015a,b; Zhang et al., 2014), which have been known well for a long time. However, the studies on the effects of low-dose BPA that more closely reflect environmental exposure levels were insufficient (Ptak and Gregoraszczuk, 2012; Ptak et al., 2014), and the molecular and cellular mechanism of low-dose BPA has not been clarified in ovarian cancer cells. Therefore, our present study used high-throughput RNA-seq technology to comprehensively elucidate the possible mechanisms underlying the effects of low-dose BPA on ovarian cancer. Numerous studies have shown that BPA exposure modulates the global gene expression that affects signal transduction pathways and cellular regulatory networks (Castillo Sanchez et al., 2016; Kim et al., 2015a), which can be successfully characterized using gene expression profile analysis. Gene expression profiles have been determined in BG-1 ovarian cancer cells (Hwang et al., 2011), Michigan Cancer Foundation-7 (MCF-7) breast cancer cells (Mesnage et al., 2017), LNCaP prostate cancer cells (Hess-Wilson et al., 2007), and human osteosarcoma (HOS) cells (Fic et al., 2015) by microarray after BPA exposure. However, limited researches performed enrichment analysis to overview the whole signal pathways of BPA (Gong et al., 2017), especially low-dose BPA in ovarian cancer. Our study first examined transcriptomic changes by RNA-seq after nanomolar BPA exposure in the human ovarian adenocarcinoma cell line SKOV3. 94 shared DEGs after exposure to 10- and 100-nM BPA were identified in this study. Rather than simply cataloged DEGs caused by BPA exposure, we sought to study the biological significance of these altered genes on biological functions and pathways. To this end, we searched DAVID databases using the 94 shared DEGs (see Supplementary Table 3) as input data. The results suggested that BPA significantly changed the expression of genes related to tumorigenesis and metastasis in ovarian carcinoma cells. Previous studies showed that the effect of BPA in proliferation usually exhibits an inverted U-shaped dose-response curve in ovarian cancer (Shi et al., 2017) and mesenchymal stem (Wang et al., 2013) cells. High-dose BPA inhibited cell growth, and a recent study reported that nanomolar-dose BPA can regulate cell proliferation positively in OVCAR-3 ovarian cancer cell lines (Shi et al., 2017). Nevertheless, our data showed that 10- and 100-nM BPA had no significant effects on the proliferation of either SKOV3 or A2780 cells (Figure 6), which was confirmed by cytotoxic tests as well (see Supplementary Figure 2). But the transcriptomic study revealed that gene expression involved in cell proliferation and apoptosis were significantly regulated by BPA. It suggested that BPA at noncytotoxic levels might cause alterations of related gene expression. Although dosing condition and exposure time in the literature can be highly variable, most studies draw a conclusion that BPA increases the migration and invasion of various cancer cell lines, such as ovarian (Kim et al., 2015b; Ptak et al., 2014), breast (Castillo Sanchez et al., 2016; Zhang et al., 2016), and cervical (Ma et al., 2015) cancer cells. Consistent with the reports in the literature, BPA treatment can significantly improve migration and invasion ability of ovarian cancer cells (Figure 7). Results from phenotypic features and functional annotation confirmed the validity of the sequencing data. Recent data indicated that BPA induces migration of BG-1 and OVCAR-3 human epithelial ovarian cancer cells (Kim et al., 2015b; Ptak et al., 2014), which are consistent with the effects of nanomolar BPA on SKOV3 and A2780 cell migration reported here. To date, there is very limited published data showing the effects of low-dose BPA exposure on the invasion ability of ovarian cancer. Previous study has demonstrated that micromolar BPA can trigger invasion in BG-1 cells (Kim et al., 2015a). Our present study revealed that exposure to low-dose BPA resulted in greater invasiveness in SKOV3 and A2780 cells for the first time, which suggested that observations were representative in various ovarian carcinoma cells. Metastasis is the main cause for deaths from cancer and is a common characteristic of advanced solid tumors (Harvey et al., 2013). It is a multistep process which commonly starts with an EMT process before dissociation from the primary tumor and cells must resist anoikis during dissemination (Harvey et al., 2013; Simpson et al., 2008). Functional annotation revealed that many relevant pathways were significantly modulated by BPA. Hippo signaling pathway (see Supplementary Figure 8) deregulation might favor metastasis, as Yes-associated Protein (YAP) over-expression can suppress anoikis of cultured cells and promote EMT at the same time (Harvey et al., 2013). However, little work has been performed on the effects of BPA on the hippo pathway and further investigation is needed. AKT is a downstream effector of integrin binding (GO: 0005178) and Focal adhesion (hsa04510) (see Supplementary Figure 9), which was enriched in DAVID database as well. Recent studies have assessed the effects of BPA on the PI3K/Akt signaling pathway and suggested that PI3K/Akt signaling pathway may be related to BPA-mediated migration in ovarian (Ptak et al., 2014) and colorectal (Chen et al., 2015) cancer cells. Cancer-associated EMT is an important process to acquire migration and invasion ability (Chaffer et al., 2016). Micromolar BPA has been reported to induce EMT of ovarian carcinoma cells through an ER-dependent pathway (Kim et al., 2015b). Results from this study revealed, for the first time, that nanomolar concentration of BPA can trigger EMT in ovarian cancer cells, leading to the down-regulation of the expression of epithelial characteristic ZO-1 and the up-regulation of the expression of mesenchymal characteristic MMP9. MMP9 is highly involved in EMT process by facilitating the protein degradation of the ECM (Polyak and Weinberg. 2009). Up-regulation of MMP9 has also been found in triple-negative breast cancer (Zhang et al., 2016). These studies suggested that environmentally equivalent doses of BPA could induce the metastasis of ovarian carcinoma cells by triggering the MMP9-dependent EMT process. The EMT process is mediated by some cellular signaling pathways, such as TGF-β, Wnt-β-catenin, receptor tyrosine kinases, Notch, Hedgehog, and microRNA pathways (Lee et al., 2017). As indicated in the KEGG pathway analysis of our study (Figure 4), Wnt signaling pathway ranked higher than TGF-β signaling pathway. Wnt/β-catenin target genes regulate cell proliferation, apoptosis, as well as EMT, consequently mediating tumorigenesis and cancer development (Arend et al., 2013). However, the effects of BPA in canonical Wnt pathway remained to be controversial. Determined by subcellular localization, aberrant catenin expression is a key marker of the activation of Wnt signaling pathway, and has been reported in male mouse reproductive cells (Fang et al., 2015). Another study demonstrated that BPA treatment suppresses Wnt/β-catenin pathway in the rat, thus impairs neural stem cells proliferation and differentiation (Tiwari et al., 2015). To confirm the functional annotation results, in this study, our results found for the first time that SKOV3 and A2780 cells treated with nanomolar-dose BPA showed significant β-catenin translocation to the nucleus, indicating that environmentally relevant-dose BPA exposure activated the canonical Wnt/β-catenin signaling pathway in ovarian carcinoma cells, and it might play a part in BPA-induced EMT process. In addition, previous studies reported that TGF-β signaling pathway is rather blocked by ER signaling resulting from BPA (10−6 M) exposure in ovarian cancer cells (Kim et al., 2015b; Park and Choi, 2014). These findings are inconsistent with our present study that BPA exposure triggered EMT and metastasis process in SKOV3 and A2780 cells. It may due to the dual role of TGF-β in carcinogenesis and development or the wide variation in BPA doses used. Thus, further study is required to elucidate the relationship between BPA and the TGF-β signaling pathway. In this study, the gene expression profile identified the differential expression of 94 genes in the SKOV3 ovarian adenocarcinoma cells following 24-h environmentally relevant-dose BPA treatment. We are aware that the major weakness of this study is that we performed transcriptomic analysis only at a single time point posttreatment, thus the recognition of the temporal effects of BPA exposure on global gene expression is not allowed. However, detailed interpretation of the biological implications of altered genes as presented here does help us understand more about the potential effects of BPA exposure on tumorigenesis and metastasis. Further studies demonstrated that BPA promoted migration and invasion and induced EMTs of ovarian carcinoma cells, which was characterized by increasing expression of MMP9 with a concomitant decrease of ZO-1. Correspondingly, BPA exposure activated the canonical Wnt signaling pathway, which might be involved in BPA-induced EMTs. Besides, the expression of miR-21 and miR-222 were up-regulated with increasing doses of BPA exposure (10 and 100 nM). Our study first comprehensively analyzed the possible mechanisms underlying the effects of BPA on ovarian cancer. Environmentally relevant doses of BPA modulated the gene expression profile, promoted EMT progress via canonical Wnt signaling pathway (Figure 11) and activated the microRNA network of ovarian cancer. Figure 11. View largeDownload slide Schematic plot showed BPA promoted EMT via canonical Wnt pathway. Figure 11. View largeDownload slide Schematic plot showed BPA promoted EMT via canonical Wnt pathway. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. ACKNOWLEDGMENT The authors thank Huiqi Chen and Yexinyi Zhou for the kind suggestions and help in the course of experiment. FUNDING This work was supported by the National Natural Science Foundation of China (Grant No: 81302455, 31471297, and 81773016), the Zhejiang Provincial Natural Science Foundation of China (Grant No: LY18C060001), and the Fundamental Research Funds for the Central Universities (No. 2017XZZX011-01). REFERENCES Arend R. C., Londono-Joshi A. I., Straughn J. M.Jr, Buchsbaum D. J. ( 2013). The Wnt/beta-catenin pathway in ovarian cancer: A review. Gynecol. Oncol.  131, 772– 779. Google Scholar CrossRef Search ADS PubMed  Calafat A. M., Kuklenyik Z., Reidy J. A., Caudill S. P., Ekong J., Needham L. L. ( 2005). Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ. Health Perspect.  113, 391– 395. Google Scholar CrossRef Search ADS PubMed  Castillo Sanchez R., Gomez R., Perez Salazar E. ( 2016). Bisphenol A induces migration through a GPER-, FAK-, Src-, and ERK2-dependent pathway in MDA-MB-231 breast cancer cells. Chem. Res. Toxicol.  29, 285– 295. Google Scholar CrossRef Search ADS PubMed  Chaffer C. L., San Juan B. P., Lim E., Weinberg R. A. ( 2016). EMT, cell plasticity and metastasis. Cancer Metastasis Rev.  35, 645– 654. Google Scholar CrossRef Search ADS PubMed  Chen Z.-J., Yang X.-L., Liu H., Wei W., Zhang K.-S., Huang H.-B., Giesy J. P., Liu H.-L., Du J., Wang H.-S.et al.  , ( 2015). Bisphenol A modulates colorectal cancer protein profile and promotes the metastasis via induction of epithelial to mesenchymal transitions. Arch. Toxicol.  89, 1371– 1381. Google Scholar CrossRef Search ADS PubMed  Chung S. H., Franceschi S., Lambert P. F. ( 2010). Estrogen and ERalpha: Culprits in cervical cancer?. Trends Endocrinol. Metab.  21, 504– 511. Google Scholar CrossRef Search ADS PubMed  Dematteo R., Keith M. M., Brophy J. T., Wordsworth A., Watterson A. E., Beck M., Rochon A., Michael F., Jyoti G., Magali P.et al.  , ( 2013). Chemical exposures of women workers in the plastics industry with particular reference to breast cancer and reproductive hazards. New Solut.  22, 427– 448. Google Scholar CrossRef Search ADS   Fang Z., Liu X., Yang X., Song X., Chen X. ( 2015). Effects of Wnt/beta-catenin signaling on bisphenol A exposure in male mouse reproductive cells. Mol. Med. Rep.  12, 5561– 5567. Google Scholar CrossRef Search ADS PubMed  Fernandez S. V., Huang Y., Snider K. E., Zhou Y., Pogash T. J., Russo J. ( 2012). Expression and DNA methylation changes in human breast epithelial cells after bisphenol A exposure. Int. J. Oncol.  41, 369– 377. Google Scholar PubMed  Fic A., Jurković Mlakar S., Juvan P., Mlakar V., Marc J., Sollner Dolenc M., Broberg K., Peterlin Mašič L. ( 2015). Genome-wide gene expression profiling of low-dose, long-term exposure of human osteosarcoma cells to bisphenol A and its analogs bisphenols AF and S. Toxicol. In Vitro  29, 1060– 1069. Google Scholar CrossRef Search ADS PubMed  Gong X., Xie H., Li X., Wu J., Lin Y. ( 2017). Bisphenol A induced apoptosis and transcriptome differences of spermatogonial stem cells in vitro. Acta Biochim. Biophys. Sin.  49, 780– 791. Google Scholar CrossRef Search ADS   Gonzalez D. M., Medici D. ( 2014). Signaling mechanisms of the epithelial-mesenchymal transition. Sci. Signal.  7, re8. Google Scholar CrossRef Search ADS PubMed  Gore A. C., Chappell V. A., Fenton S. E., Flaws J. A., Nadal A., Prins G. S., Toppari J., Zoeller R. T. ( 2015). Executive summary to EDC-2: The endocrine society’s second scientific statement on endocrine-disrupting chemicals. Endocr. Rev.  36, 593– 602. Google Scholar CrossRef Search ADS PubMed  Grady D., Gebretsadik T., Kerlikowske K., Ernster V., Petitti D. ( 1995). Hormone replacement therapy and endometrial cancer risk: A meta-analysis. Obstet. Gynecol.  85, 304– 313. Google Scholar CrossRef Search ADS PubMed  Harvey K. F., Zhang X., Thomas D. M. ( 2013). The Hippo pathway and human cancer. Nat. Rev. Cancer  13, 246– 257. Google Scholar CrossRef Search ADS PubMed  Hess-Wilson J. K., Webb S. L., Daly H. K., Leung Y.-K., Boldison J., Comstock C. E. S., Sartor M. A., Ho S.-M., Knudsen K. E. ( 2007). Unique bisphenol A transcriptome in prostate cancer: Novel effects on ERbeta expression that correspond to androgen receptor mutation status. Environ. Health Perspect.  115, 1646– 1653. Google Scholar CrossRef Search ADS PubMed  Hoffmann M., Gogola J., Kotula-Balak M., Ptak A. ( 2017). Stimulation of ovarian cell proliferation by tetrabromobisphenol A but not tetrachlorobisphenol A through G protein-coupled receptor 30. Toxicol. In Vitro  45, 54– 59. Google Scholar CrossRef Search ADS PubMed  Hwang K. A., Park S. H., Yi B. R., Choi K. C. ( 2011). Gene alterations of ovarian cancer cells expressing estrogen receptors by estrogen and bisphenol a using microarray analysis. Lab. Anim. Res.  27, 99– 107. Google Scholar CrossRef Search ADS PubMed  Ikezuki Y., Tsutsumi O., Takai Y., Kamei Y., Taketani Y. ( 2002). Determination of bisphenol A concentrations in human biological fluids reveals significant early prenatal exposure. Hum. Reprod.  17, 2839– 2841. Google Scholar CrossRef Search ADS PubMed  Kang J. H., Kito K., Kondo F. ( 2003). Factors influencing the migration of bisphenol A from cans. J. Food Prot.  66, 1444– 1447. Google Scholar CrossRef Search ADS PubMed  Kang N. H., Hwang K. A., Lee H. R., Choi D. W., Choi K. C. ( 2013). Resveratrol regulates the cell viability promoted by 17beta-estradiol or bisphenol A via down-regulation of the cross-talk between estrogen receptor alpha and insulin growth factor-1 receptor in BG-1 ovarian cancer cells. Food Chem. Toxicol.  59, 373– 379. Google Scholar CrossRef Search ADS PubMed  Kim Y. S., Choi K. C., Hwang K. A. ( 2015a). Genistein suppressed epithelial-mesenchymal transition and migration efficacies of BG-1 ovarian cancer cells activated by estrogenic chemicals via estrogen receptor pathway and downregulation of TGF-beta signaling pathway. Phytomedicine  22, 993– 999. Google Scholar CrossRef Search ADS   Kim Y. S., Hwang K. A., Hyun S. H., Nam K. H., Lee C. K., Choi K. C. ( 2015b). Bisphenol A and nonylphenol have the potential to stimulate the migration of ovarian cancer cells by inducing epithelial-mesenchymal transition via an estrogen receptor dependent pathway. Chem. Res. Toxicol.  28, 662– 671. Google Scholar CrossRef Search ADS   Lantta M. ( 1984). Estradiol and progesterone receptors in normal ovary and ovarian tumors. Acta Obstet. Gynecol. Scand.  63, 497– 503. Google Scholar CrossRef Search ADS PubMed  Lee H. M., Hwang K. A., Choi K. C. ( 2017). Diverse pathways of epithelial mesenchymal transition related with cancer progression and metastasis and potential effects of endocrine disrupting chemicals on epithelial mesenchymal transition process. Mol. Cell Endocrinol.  457, 103– 113. Google Scholar CrossRef Search ADS PubMed  Lu W., Zhang H., Niu Y., Wu Y., Sun W., Li H., Kong J., Ding K., Shen H.-M., Wu H.et al.  , ( 2017). Long non-coding RNA linc00673 regulated non-small cell lung cancer proliferation, migration, invasion and epithelial mesenchymal transition by sponging miR-150-5p. Mol. Cancer  16, 118. Google Scholar CrossRef Search ADS PubMed  Ma X. F., Zhang J., Shuai H. L., Guan B. Z., Luo X., Yan R. L. ( 2015). IKKbeta/NF-kappaB mediated the low doses of bisphenol A induced migration of cervical cancer cells. Arch. Biochem. Biophys.  573, 52– 58. Google Scholar CrossRef Search ADS PubMed  Mesnage R., Phedonos A., Arno M., Balu S., Corton J. C., Antoniou M. N. ( 2017). Editor’s highlight: Transcriptome profiling reveals bisphenol A alternatives activate estrogen receptor alpha in human breast cancer cells. Toxicol. Sci.  158, 431– 443. Google Scholar CrossRef Search ADS PubMed  Missmer S. A., Eliassen A. H., Barbieri R. L., Hankinson S. E. ( 2004). Endogenous estrogen, androgen, and progesterone concentrations and breast cancer risk among postmenopausal women. J. Natl. Cancer Inst.  96, 1856– 1865. Google Scholar CrossRef Search ADS PubMed  Moustakas A., Heldin C. H. ( 2007). Signaling networks guiding epithelial-mesenchymal transitions during embryogenesis and cancer progression. Cancer Sci.  98, 1512– 1520. Google Scholar CrossRef Search ADS PubMed  National Academies of Sciences, Engineering, and Medicine. ( 2016) Ovarian Cancers: Evolving Paradigms in Research and Care , pp. 3. The National Academies Press, Washington, DC. Park M. A., Choi K. C. ( 2014). Effects of 4-nonylphenol and bisphenol A on stimulation of cell growth via disruption of the transforming growth factor-beta signaling pathway in ovarian cancer models. Chem. Res. Toxicol.  27, 119– 128. Google Scholar CrossRef Search ADS PubMed  Park S.-H., Kim K.-Y., An B.-S., Choi J.-H., Jeung E.-B., Leung P. C. K., Choi K.-C. ( 2009). Cell growth of ovarian cancer cells is stimulated by xenoestrogens through an estrogen-dependent pathway, but their stimulation of cell growth appears not to be involved in the activation of the mitogen-activated protein kinases ERK-1 and p38. J. Reprod. Dev.  55, 23. Google Scholar CrossRef Search ADS PubMed  Polyak K., Weinberg R. A. ( 2009). Transitions between epithelial and mesenchymal states: Acquisition of malignant and stem cell traits. Nat. Rev. Cancer  9, 265– 273. Google Scholar CrossRef Search ADS PubMed  Ptak A., Gregoraszczuk E. L. ( 2012). Bisphenol A induces leptin receptor expression, creating more binding sites for leptin, and activates the JAK/Stat, MAPK/ERK and PI3K/Akt signalling pathways in human ovarian cancer cell. Toxicol. Lett.  210, 332– 337. Google Scholar CrossRef Search ADS PubMed  Ptak A., Hoffmann M., Gruca I., Barc J. ( 2014). Bisphenol A induce ovarian cancer cell migration via the MAPK and PI3K/Akt signalling pathways. Toxicol. Lett.  229, 357– 365. Google Scholar CrossRef Search ADS PubMed  Ptak A., Wrobel A., Gregoraszczuk E. L. ( 2011). Effect of bisphenol-A on the expression of selected genes involved in cell cycle and apoptosis in the OVCAR-3 cell line. Toxicol. Lett.  202, 30– 35. Google Scholar CrossRef Search ADS PubMed  Rochester J. R. ( 2013). Bisphenol A and human health: A review of the literature. Reprod. Toxicol.  42, 132– 155. Google Scholar CrossRef Search ADS PubMed  Rodriguez C., Patel A. V., Calle E. E., Jacob E. J., Thun M. J. ( 2001). Estrogen replacement therapy and ovarian cancer mortality in a large prospective study of US women. Jama  285, 1460– 1465. Google Scholar CrossRef Search ADS PubMed  Seachrist D. D., Bonk K. W., Ho S. M., Prins G. S., Soto A. M., Keri R. A. ( 2016). A review of the carcinogenic potential of bisphenol A. Reprod. Toxicol.  59, 167– 182. Google Scholar CrossRef Search ADS PubMed  Shi X.-Y., Wang Z., Liu L., Feng L.-M., Li N., Liu S., Gao H. ( 2017). Low concentrations of bisphenol A promote human ovarian cancer cell proliferation and glycolysis-based metabolism through the estrogen receptor-alpha pathway. Chemosphere  185, 361– 367. Google Scholar CrossRef Search ADS PubMed  Siegel R. L., Miller K. D., Jemal A. ( 2018). Cancer statistics, 2018. CA Cancer J. Clin.  68, 7– 30. Google Scholar CrossRef Search ADS PubMed  Simpson C. D., Anyiwe K., Schimmer A. D. ( 2008). Anoikis resistance and tumor metastasis. Cancer Lett.  272, 177– 185. Google Scholar CrossRef Search ADS PubMed  Tiwari S. K., Agarwal S., Seth B., Yadav A., Ray R. S., Mishra V. N., Chaturvedi R. K. ( 2015). Inhibitory effects of bisphenol-A on neural stem cells proliferation and differentiation in the rat brain are dependent on wnt/beta-catenin pathway. Mol. Neurobiol.  52, 1735– 1757. Google Scholar CrossRef Search ADS PubMed  Uhlen M., Zhang C., Lee S., Sjöstedt E., Fagerberg L., Bidkhori G., Benfeitas R., Arif M., Liu Z., Edfors F.et al.  , ( 2017). A pathology atlas of the human cancer transcriptome. Science  357, eaan2507. Available at: www.proteinatlas.org. Accessed 8 October 2017. Google Scholar CrossRef Search ADS PubMed  Vandenberg L. N., Maffini M. V., Wadia P. R., Sonnenschein C., Rubin B. S., Soto A. M. ( 2007). Exposure to environmentally relevant doses of the xenoestrogen bisphenol-A alters development of the fetal mouse mammary gland. Endocrinology  148, 116– 127. Google Scholar CrossRef Search ADS PubMed  Wang K. H., Kao A. P., Chang C. C., Lin T. C., Kuo T. C. ( 2013). Bisphenol A at environmentally relevant doses induces cyclooxygenase-2 expression and promotes invasion of human mesenchymal stem cells derived from uterine myoma tissue. Taiwan J. Obstet. Gynecol.  52, 246– 252. Google Scholar CrossRef Search ADS PubMed  Willcocks D., Toppila M., Hudson C. N., Tyler J. P., Baird P. J., Eastman C. J. ( 1983). Estrogen and progesterone receptors in human ovarian tumors. Gynecol. Oncol.  16, 246– 253. Google Scholar CrossRef Search ADS PubMed  Willert K., Nusse R. ( 1998). Beta-catenin: A key mediator of Wnt signaling. Curr. Opin. Genet. Dev.  8, 95– 102. Google Scholar CrossRef Search ADS PubMed  Zhang K.-S., Chen H.-Q., Chen Y.-S., Qiu K.-F., Zheng X.-B., Li G.-C., Yang H.-D., Wen C.-J. ( 2014). Bisphenol A stimulates human lung cancer cell migration via upregulation of matrix metalloproteinases by GPER/EGFR/ERK1/2 signal pathway. Biomed. Pharmacother.  68, 1037– 1043. Google Scholar CrossRef Search ADS PubMed  Zhang X. L., Liu N., Weng S. F., Wang H. S. ( 2016). Bisphenol A increases the migration and invasion of triple-negative breast cancer cells via oestrogen-related receptor gamma. Basic Clin. Pharmacol. Toxicol.  119, 389– 395. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Toxicological Sciences Oxford University Press

Low Dose of Bisphenol A Modulates Ovarian Cancer Gene Expression Profile and Promotes Epithelial to Mesenchymal Transition Via Canonical Wnt Pathway

Loading next page...
 
/lp/ou_press/low-dose-of-bisphenol-a-modulates-ovarian-cancer-gene-expression-dPsEl3lHIt
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1096-6080
eISSN
1096-0929
D.O.I.
10.1093/toxsci/kfy107
Publisher site
See Article on Publisher Site

Abstract

Abstract The xenoestrogen bisphenol A (BPA) is a synthetic endocrine disrupting chemical, having the potential to increase the risk of hormone-dependent ovarian cancer. Thus, a deeper understanding of the molecular and cellular mechanisms is urgently required in the novel cell models of ovarian cancer which express estrogen receptors. To understand the possible mechanisms underlying the effects of BPA, human ovarian adenocarcinoma SKOV3 cells were exposed to BPA (10 or 100 nM) or 0.1% DMSO for 24 h, and then global gene expression profile was determined by high-throughput RNA sequencing. Also, enrichment analysis was carried out to find out relevant functions and pathways within which differentially expressed genes were significantly enriched. Transcriptomic analysis revealed 94 differential expression genes. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analyses indicated that these genes related to tumorigenesis and metastasis. Further studies were carried out to validate the results of functional annotation, which indicated that BPA (10 and 100 nM) increased migration and invasion as well as induced epithelial to mesenchymal transitions in SKOV3 and A2780 cells. Accordingly, environmentally relevant-dose BPA activated the canonical Wnt signaling pathway. Our study first comprehensively analyzed the possible mechanisms underlying the effects of BPA on ovarian cancer. Environmentally relevant doses of BPA modulated the gene expression profile, promoted epithelial to mesenchymal transition progress via canonical Wnt signaling pathway of ovarian cancer. BPA, gene expression profile, ovarian cancer, metastasis, EMT, canonical Wnt pathway Endocrine disrupting chemicals (EDCs) are natural or man-made agents in the environment that could interfere with some aspect of hormone action in the body, thus cause disrupting effects on the endocrine system (Gore et al., 2015). Circulating levels of natural hormones may be strongly implicated in the risks of several reproductive system cancers. For example, the abnormal level of estrogen is related to the pathogenesis and development of ovarian (Rodriguez et al., 2001), breast (Missmer et al., 2004), cervical (Chung et al., 2010), and endometrial (Grady et al., 1995) tumors which are highly estrogen receptor (ER)-positive or estrogen-responsive. Hence, EDCs that have estrogenicity or estrogenic activity may be important risk factors of estrogen-responsive cancers. The xenoestrogen bisphenol A (BPA) is a synthetic EDC, having the potential to promote the progression of some estrogen-responsive tumors, such as ovarian (Seachrist et al., 2016), and breast (DeMatteo et al., 2013) tumors. BPA is extensively used as an ingredient in many household appliances, including epoxy resin linings of beverage containers and food cans, polycarbonate plastics, and dental sealants (Rochester 2013). What’s more, in the epoxy resin and polycarbonate, heating time and temperature can cause hydrolysis of ester bonds linked to BPA, thus BPA monomer leaching from these products, and contaminating the food (Kang et al., 2003). An epidemiological study has reported that in a population of 394 human subjects, up to 95% have detectable levels of BPA in serum, urine, and saliva owing to low but sustained exposure (Calafat et al., 2005). In a previous study, BPA was detected in 36 human ovarian follicular fluid aspirates in the range of nanomolar concentrations (Ikezuki et al., 2002). Determined from measurable concentrations of BPA in body fluids and tissues, it has been estimated that among the general population, the theoretical internal concentration is ranging from 10 to 100 nM (Vandenberg et al., 2007). Nowadays, adverse endocrine disruptive effects and hormone-dependent cancer formation caused by BPA have risen global concern. In vitro studies have found that BPA not only enhances proliferation and glycolysis-based metabolism (Ptak et al., 2011; Park et al., 2009; Shi et al., 2017), but also stimulates the cell metastasis of the ovarian carcinoma cells (Kim et al., 2015b; Ptak et al., 2014). However, most of these studies fasten on the cancer-promoting effects of BPA at high concentrations, thus neglecting the effects of low concentrations that more closely reflect environmental exposure levels. Moreover, due to the large variations in BPA dosage used, previous researches draw inconsistent conclusions on the transcriptional and apoptotic properties of BPA (Hwang et al., 2011; Ptak et al., 2011). The effect of BPA usually exhibits a U-shaped or an inverted U-shaped dose-response curve (Shi et al., 2017; Wang et al., 2013). Therefore, it is unwarranted to take putative low-dose biological activity for granted as high-dose BPA effects. Thus, the aim of the present study was to assess the cellular effects of environmentally related-dose BPA on ovarian cancer, and 10 and 100 nM were chosen as the concentrations of BPA in this study. The ovarian carcinoma is one the most commonly malignant tumors among women worldwide, owing to the phenomena that it’s always diagnosed in advanced stages and has intra-abdominal metastasis (Siegel et al., 2018). When the lesion is entirely localized to the ovary, the 5-year survival rate is >90%, but when the lesion is disseminated intra-abdominally, the incidence falls below 20% (Ptak et al., 2014). As a result, its death to incidence ratio is up to 60%–65% (National Academies of Sciences, Engineering, and Medicine, 2016). Previous studies have demonstrated that BPA plays a potential part in the progression of ovarian cancer (Kim et al., 2015b; Ptak et al., 2014). However, how BPA induces the progression of ovarian cancer remains largely undefined. Human ovarian carcinomas were more often ER-positive than normal ovary cells or benign tumors (Lantta 1984; Willcocks et al., 1983). Estrogen’s classical nuclear receptors, ERα/β, and membrane receptor G-protein-coupled estrogen receptor (GPER) have been detected in a variety of ovarian carcinoma cell lines (Hoffmann et al., 2017; Hwang et al., 2011). However, it still remains unknown whether exposure to BPA can cause abnormal estrogenic signaling mediated by these ERs. A deeper understanding of the molecular and cellular mechanisms is urgently required in the novel cell models of ovarian cancer which express ERs. Therefore, with the help of high-throughput sequencing techniques, the main objective of our research is to unravel the altered gene expression profile and identify molecular and cellular mechanisms mediating the effects of environmentally related-dose BPA exposure on ovarian cancer. Materials and Methods Chemicals and cell lines See “Chemicals and cell lines” in the Supplementary Material. RNA sequencing Human ovarian adenocarcinoma SKOV3 cells were treated with BPA (10 or 100 nM) or dimethyl sulfoxide (DMSO) for 24 h, and then cells were washed twice with phosphate-buffered saline, total RNA was extracted from cells using Trizol reagent (Sangon Biotech) according to the manufacturer’s instructions. RNA Sequencing was performed on an Illumina Hisreq 4000 platform with 150 bp paired-end reads at Novogene Bioinformatics Technology Co., Ltd (see “RNA Sequencing” in the Supplementary Material for additional details, including RNA extraction, Library construction and Illumina sequencing, and in silico gene expression analysis). RNA-seq data have been deposited in the ArrayExpress database at EMBL-EBI (www.ebi.ac.uk/arrayexpress; last accessed November 28, 2017) under accession number E-MTAB-6295. Functional annotation The shared differentially expressed genes (DEGs) of SKOV3 cells treated with 10-nM BPA and 100-nM BPA in comparison with DMSO-treatment SKOV3 cells were used for further functional annotation (see “Functional annotation” in the Supplementary Material for additional details). Cell viability The cell viability was determined using the CCK-8 kit assay (Beyotime) according to the manufacturer’s protocol (see “Cell viability” in the Supplementary Material for additional details of the CCK-8 assay procedure). Cell migration Wound-healing assay was used to assess cell migration by observing the ability of the cells to move into an acellular space (see “Cell migration” in the Supplementary Material for additional details of the wound-healing assay procedure). Cell invasion Transwell assay was performed to assess cell invasion ability. The details of transwell assay procedure see “Cell invasion” in the Supplementary Material. Cytoplasmic and nuclear protein extraction and Western blotting The cells were collected and lysed with cell lysis buffer (Beyotime). Cytoplasmic and nuclear protein of cultured ovarian cancer cells were separated by Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime) according to the manufacturer’s protocol. The details of Western blotting procedure and antibodies used in this study see “Western blotting” in the Supplementary Material. Immunofluorescence staining Immunofluorescence staining was performed to detect the translocation of β-catenin. The details of immunofluorescence staining procedure see “Immunofluorescence staining” in the Supplementary Material. Quantitative real-time PCR We performed qPCR to detect mRNA or microRNA expression levels (see “Quantitative real-time PCR” in the Supplementary Material for additional details). Primer sequences were provided in Supplementary Tables 1 and 2. Statistical analyses Statistical analyses were the same as our previous study (Lu et al., 2017) (see “Statistical analyses” in the Supplementary Material for additional details). RESULTS Confirmation of the Expression of ERs According to the human protein atlas database (Uhlen et al., 2017), classical nuclear receptors, ERα/β, membrane receptor GPER have been detected in ovarian cancer tissues (see Supplementary Figure 1). We confirmed the expression of ERα/β and GPER by qPCR and western blotting in SKOV3 and A2780 cells. As shown in Figure 1, SKOV3 cells were ERα/β and GPER expressing, A2780 cells were ERα negative and had lower basal ERβ and GPER protein levels. Figure 1. View largeDownload slide Expression of ERα/β, GPER mRNAs and proteins in SKOV3 and A2780 cells. A, ERs mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. B, Basal ERα/β and GPER protein expression in SKOV3 and A2780 cells. β-actin servers as the loading control for western blotting. Data are representative of 3 independent experiments. Figure 1. View largeDownload slide Expression of ERα/β, GPER mRNAs and proteins in SKOV3 and A2780 cells. A, ERs mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. B, Basal ERα/β and GPER protein expression in SKOV3 and A2780 cells. β-actin servers as the loading control for western blotting. Data are representative of 3 independent experiments. Identification of Gene Expression Changes Gene expression changes in SKOV3 cells after physiological-dose BPA treatment for 24 h were identified using the multiple-testing module of the Cuffdiff program in Cufflinks package. The genes with q-value (the false discovery rate (FDR)-adjusted p-value of the test statistic) < .05 were defined as DEGs. To visualize the gene expression profiles across all doses and replicates, the clustering analysis was performed, which showed that replicates of each group clustered closely together (Figs. 2B and 2C), indicating the quality of the gene expression data sets was high enough for further exploration. Furthermore, hierarchical clustering demonstated similarities in the gene expression profiles for 10- and 100-nM BPA treatment. This may be due to the fact that both doses were relatively low in physiological level and hence had similar gene expression alterations. Thus, the genes with q-value < .05 over these 2 doses were considered as DEGs, which increased the analysis stringency, namely, reduced false positives because of multiple testing. When compared with the control group, a total of 142 gene expressions were significantly altered in 10-nM BPA treated cells while 148 in 100-nM BPA treated cells. Among those genes identified in both groups, only 94 genes were found to have significant altered expression in both BPA dose groups. The Venn diagram showed the portion of shared DEGs after exposure of SKOV3 cells to either 10- or 100-nM BPA in SKOV3 cells (Figure 2D). Of the 94 genes meeting the criterion, 36 were found to be up-regulated, and 58 were down-regulated, which were listed in Supplementary Table 3. And we also applied a heatmap analysis of the shared DEGs (Figure 2A). In parallel with the clustering analysis results, the fold change of DEGs in the case of 10- and 100-nM BPA treatment showed no significant difference (Figure 2A; see Supplementary Table 3). Figure 2. View largeDownload slide Gene expression analysis after 24-h exposure to 0.1% DMSO (control), 10- or 100-nM BPA in SKOV3 cells with 3 biological replicates. A, Heatmap reflected expression profile for genes differently regulated (q-value < .05) over 2 BPA doses. B and C, B Dendrogram and C multi-dimensional scaling (MDS) plot generated by hierarchical clustering of SKOV3 gene expression. D, Venn diagram showed the overlap of DEGs induced by treatment with 10- and 100-nM BPA compared with the control group. Figure 2. View largeDownload slide Gene expression analysis after 24-h exposure to 0.1% DMSO (control), 10- or 100-nM BPA in SKOV3 cells with 3 biological replicates. A, Heatmap reflected expression profile for genes differently regulated (q-value < .05) over 2 BPA doses. B and C, B Dendrogram and C multi-dimensional scaling (MDS) plot generated by hierarchical clustering of SKOV3 gene expression. D, Venn diagram showed the overlap of DEGs induced by treatment with 10- and 100-nM BPA compared with the control group. Functional Annotation Analysis of the Shared DEGs Over 2 Doses To understand the biological significance of altered genes caused by BPA exposure, the DEGs were searched in the functional annotation Database for Annotation, Visualization and Integrated Discovery (DAVID) database and categorized by putative functions. For gene ontology (GO) analysis, transcripts were divided into 3 major categories: biological processes, cellular component, and molecular function. As shown in Figure 3, among the cluster of biological processes, it exhibited that the top 3 scores were extracellular matrix (ECM) organization (GO: 0030198), positive regulation of cell proliferation (GO: 0008284) and cellular response to interleukin-1 (GO: 0071347). In addition, the top 3 scores of cellular component were extracellular space (GO: 0005615), ECM (GO: 0031012), and extracellular region (GO: 0005576). Moreover, insulin-like growth factor binding (GO: 0005520), structural molecule activity (GO: 0005198) were the molecular functions most targeted by BPA treatment. The enriched DEGs of each GO term were listed in Supplementary Table 4. Figure 3. View largeDownload slide GO of altered genes. The expression profile of the entire list of the DEGs was used as input data in GO analysis using DAVID database. Annotated terms were classified into 3 major categories (biological process, cellular component, and molecular function) and 31 subgroups. The numbers on the right of bars indicated related “related DEGs count”/“pop hits” of each GO term. Figure 3. View largeDownload slide GO of altered genes. The expression profile of the entire list of the DEGs was used as input data in GO analysis using DAVID database. Annotated terms were classified into 3 major categories (biological process, cellular component, and molecular function) and 31 subgroups. The numbers on the right of bars indicated related “related DEGs count”/“pop hits” of each GO term. Then, using the shared DEGs over 2 doses as input data, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to visualize the effects of BPA exposure on specific cell signaling pathway-related gene expression changes, and screened for biological pathways that might be involved. The results, including representative pathways and the number of genes related to these pathways were shown in Figure 4. The enriched DEGs of each KEGG pathway were listed in Supplementary Table 5. Figure 4. View largeDownload slide KEGG pathway analysis of altered genes. The expression profile of the entire list of the DEGs was used as input data in KEGG pathway analysis using DAVID database. The numbers on the right of bars indicated “related DEGs count”/“pop hits” of each pathway. Figure 4. View largeDownload slide KEGG pathway analysis of altered genes. The expression profile of the entire list of the DEGs was used as input data in KEGG pathway analysis using DAVID database. The numbers on the right of bars indicated “related DEGs count”/“pop hits” of each pathway. Among the enriched KEGG pathways (Figure 4), the top score was microRNAs in cancer (hsa05206) (see Supplementary Figure 3), suggesting microRNAs such as miR-21, miR-221, miR-222, miR-19a, and miR-7 might be involved in tumorigenesis following BPA exposure. To confirm the functional annotation results and sequencing data, these expression changes of microRNAs were further detected by qPCR. Specifically, the expression of miR-21 and miR-222 were up-regulated with increasing doses of BPA exposure (10 and 100 nM) in SKOV3 and A2780 cells (see Figure 5). The SERPINB5 (maspin) mRNA is a known target of miR-21, and the mRNA encoding TIMP3 is miR-21 and miR-222’s common target. We therefore confirmed that BPA decreased the transcript levels of SERPINB5 (maspin) and TIMP3 in SKOV3 and A2780 cells (see Supplementary Figure 4), consistent with a BPA-mediated increase in miR-21 and miR-222 levels. Figure 5. View largeDownload slide MicroRNA expression in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. The expression of (A, C) miR-21 and (B, D) miR-222 were determined using qPCR. * p < .05, ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 5. View largeDownload slide MicroRNA expression in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. The expression of (A, C) miR-21 and (B, D) miR-222 were determined using qPCR. * p < .05, ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. These functional annotation results suggested that BPA exposure significantly modified genes related to tumorigenesis and metastasis in ovarian carcinoma cells. It is worth mentioning that genes related to biological processes such as positive regulation of cell proliferation (GO: 0008284), positive regulation of cysteine-type endopeptidase activity involved in the apoptotic process (GO: 0043280) (Figure 3; see Supplementary Table 4), and microRNAs in cancer (hsa05206) (Figure 4; see Supplementary Table 5) were found to be significantly representative in the final list of genes. Regarding to the regulation of carcinoma metastasis, the 94 shared DEGs were related to GO terms and KEGG pathways like ECM organization (GO: 0030198), ECM-receptor interaction (hsa04512) and positive regulation of cell migration (GO: 0030335), and these pathways included PI3K-Akt signaling pathway (hsa04151), hippo signaling pathway (hsa04390), p53 signaling pathway (hsa04115) (see Supplementary Figure 5), cellular response to transforming growth factor β (TGF-β) stimulus (GO: 0071560), TGF-β signaling pathway (hsa04350), and Wnt signaling pathway (hsa04310) (Figs. 3 and 4; see Supplementary Tables 4 and 5), indicating nanomolar-dose BPA exposure induced alteration of the genes expression was closely involved in proliferation and cancer development. Effects of BPA on OC Cell Proliferation To determine the effects of BPA on OC cell proliferation, we used 2 human ovarian adenocarcinoma cell lines SKOV3 and A2780. SKOV3 and A2780 cells were stimulated with series concentrations of BPA which were in the range of 10 to 4 × 105 nM for 24, 48, or 72 h, and then we performed CCK-8 kit assay to measure cell viability. As shown in Supplementary Figure 2A, compared with vehicle-treated cells, concentrations of BPA > 105 nM significantly reduced SKOV3 cell viability in a concentration-dependent and time-dependent manner. As for A2780 cells, BPA showed cytotoxicity at 104 nM after 24-h exposure (see Supplementary Figure 2B). Besides, BPA had no statistically significant effects on cell proliferation at 10- and 100-nM doses in both cell lines (Figure 6), which represented environmentally equivalent concentrations. Figure 6. View largeDownload slide Effects of BPA on proliferation of SKOV3 and A2780 cells. (A) SKOV3 and (B) A2780 cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24, 48, and 72 h, and then assessed by CCK-8 kit assay. ns, no statistical significance. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 6. View largeDownload slide Effects of BPA on proliferation of SKOV3 and A2780 cells. (A) SKOV3 and (B) A2780 cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24, 48, and 72 h, and then assessed by CCK-8 kit assay. ns, no statistical significance. Values are mean ± SEM. Data are representative of 3 independent experiments. Effect of Nanomolar-Dose BPA on OC Cell Migration and Invasion Tumor metastasis is one of the most life-threatening causes of ovarian cancer (Ptak et al., 2014). We further evaluated the effects of low-dose BPA (10 or 100 nM) on the migration and invasion ability of SKOV3 and A2780 cells in vitro. In parallel with functional annotation results, compared with the vehicle group, we found that stimulation with 10- or 100-nM BPA for 24 or 48 h accelerated wound closure in wound scratch assay (Figs. 7A and 7B), suggesting that BPA induced the migration of ovarian carcinoma cells. Furthermore, exposure to BPA led to significantly more invasive cells (Figs. 7C–F). Figure 7. View largeDownload slide Nanomolar concentrations of BPA induced migration and invasion of SKOV3 and A2780 cells. A and B, Representative images of the wound scratch assay in a 100× light microscope utilizing the A SKOV3 and B A2780 cell lines after scratching 24 and 48 h. Cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA. Effects of BPA on SKOV3 or A2780 cell invasion. Both SKOV3 and A2780 cells were exposed to 10- or 100-nM BPA for 24 h, and then measured using transwell assay with Matrigel after incubation for 18 h. Invasion cells were fixed, stained, photographed, and counted in 5 random views. C and E, Representative digital images of the transwell assay taken randomly by a 200× light microscope. D and F, The quantitative analysis of invaded cells after 24-h BPA exposure and 18-h incubation. ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 7. View largeDownload slide Nanomolar concentrations of BPA induced migration and invasion of SKOV3 and A2780 cells. A and B, Representative images of the wound scratch assay in a 100× light microscope utilizing the A SKOV3 and B A2780 cell lines after scratching 24 and 48 h. Cells were treated with 0.1% DMSO (vehicle), 10- or 100-nM BPA. Effects of BPA on SKOV3 or A2780 cell invasion. Both SKOV3 and A2780 cells were exposed to 10- or 100-nM BPA for 24 h, and then measured using transwell assay with Matrigel after incubation for 18 h. Invasion cells were fixed, stained, photographed, and counted in 5 random views. C and E, Representative digital images of the transwell assay taken randomly by a 200× light microscope. D and F, The quantitative analysis of invaded cells after 24-h BPA exposure and 18-h incubation. ** p < .01, *** p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. BPA Exposure Triggers Epithelial to Mesenchymal Transitions In the functional annotation results, it is noteworthy that some of the DEGs including MMP7, BAMBI, FOSL1, MYC, and THBS1, significantly overrepresented pathways including Wnt signaling pathway (see Supplementary Figure 6) and TGF-β signaling pathway (see Supplementary Figure 7), which have been reported to collaborate to induce activation of epithelial to mesenchymal transition (EMT) program (Moustakas and Heldin, 2007). This suggested that BPA may promote ovarian cancer metastasis via induction of EMT. EMT has been thought to be the first step of cancer metastasis, with epithelial characteristic like ZO-1 is down-regulated, while mesenchymal characteristic like MMP9 is up-regulated. After stimulation with 10- or 100-nM BPA for 24 h, western blotting revealed that BPA exposure down-regulated the expression of epithelial marker ZO-1, while up-regulated the expression of mesenchymal marker MMP9 in a dose-dependent manner (Figs. 8A and 8B). And those gene expression changes were confirmed by qPCR (Figs. 8C–F). Compared with control cells, we observed statistically significant increase in MMP9 mRNA level and decrease in ZO-1 mRNA level after BPA exposure for 24 h. Figure 8. View largeDownload slide BPA treatment induced EMT. Cells were treated with 10- or 100-nM BPA for 24 h to induce the EMT. A and B, Western blotting showed that BPA induced a down-regulation of the epithelial maker ZO-1 and an up-regulation of the mesenchymal marker MMP9. GAPDH servers as a loading control. C–F, qPCR analysis of ZO-1 and MMP9 mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. * p < .05, **p < .01, ***p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. Figure 8. View largeDownload slide BPA treatment induced EMT. Cells were treated with 10- or 100-nM BPA for 24 h to induce the EMT. A and B, Western blotting showed that BPA induced a down-regulation of the epithelial maker ZO-1 and an up-regulation of the mesenchymal marker MMP9. GAPDH servers as a loading control. C–F, qPCR analysis of ZO-1 and MMP9 mRNA expression in SKOV3 and A2780 cells. The gene GAPDH was used as an internal control. * p < .05, **p < .01, ***p < .001. Values are mean ± SEM. Data are representative of 3 independent experiments. BPA Exposure Activated the Canonical Wnt Signaling Pathway The EMT process is mediated by some cellular signaling pathways, such as TGF-β, Wnt-β-catenin, receptor tyrosine kinases, Notch, Hedgehog pathways and microRNA network (Lee et al., 2017). TGF-β pathway is one the most well-described signaling pathways to trigger EMT process, which involves in diverse developmental processes and cellular functions (Gonzalez and Medici, 2014). However, as indicated in Figure 4, Wnt signaling pathway ranked higher than TGF-β signaling pathway. Additionally, although the relationship between BPA and the TGF-β signaling pathway has been reported (Kim et al., 2015a; Park and Choi, 2014), there is now a consensus that micromolar BPA, in fact, inhibits TGF-β signaling pathway in ovarian cancer cells. Therefore, we emphasized on studying the undefined relationship between BPA and the Wnt signaling pathway. To this end, we analyzed genetic changes related to Wnt signaling pathway by the GeneMANIA algorithm and identified the functional network of Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, and MYC) (Figure 9). Figure 9. View largeDownload slide Functional network of the Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, MYC). GeneMANIA retrieved known and predicted interactions between these genes and added extra genes (small gray circles) that are strongly connected to query genes (large circles). Figure 9. View largeDownload slide Functional network of the Wnt signaling pathway-related DEGs (MMP7, BAMBI, FOSL1, MYC). GeneMANIA retrieved known and predicted interactions between these genes and added extra genes (small gray circles) that are strongly connected to query genes (large circles). So far, previous study indicated that Wnt signaling pathways include the canonical Wnt pathway, the noncanonical planar cell polarity pathway, and the noncanonical Wnt/calcium pathway (Arend et al., 2013). As shown in Supplementary Figure 6, the DEGs of MMP7 (Uterine), BAMBI, FOSL1 (fra-1), and MYC (c-myc) regulate the canonical Wnt pathway. Previous studies suggested that in the canonical Wnt pathway, β-catenin is the key mediator (Willert and Nusse, 1998), we therefore investigated the effects of BPA on nuclear translocation of β-catenin in SKOV3 and A2780 cells. After stimulation with 10- or 100-nM BPA for 24 h, SKOV3 as well as A2780 cells showed increased β-catenin translocation to the nucleus in a dose-dependent manner in western blotting (Figs. 10A and 10B), suggesting that BPA promoted ovarian cancer EMT through activation of canonical Wnt/β-catenin signaling. Immunofluorescence staining revealed the BPA-induced β-catenin translocation (Figs. 10C and 10D). In the control cells, β-catenin was expressed at a low level, and located mostly at the cytoplasm. After BPA exposure, the protein level of β-catenin was significantly elevated in the nucleus. This finding validated the results of functional annotation. Figure 10. View largeDownload slide BPA exposure activated the canonical Wnt signaling pathway in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. A and B, Western blotting of nuclear and cytoplasmic localization of β-catenin. Lamin A/C serves as a nuclear protein loading control, while GAPDH served as a cytoplasmic loading control. C and D, Immunofluorescence staining of β-catenin expression in nuclear and cytoplasm, demonstrating the nuclear translocation of β-catenin upon a 24-h BPA treatment. Data are representative of 3 independent experiments. Figure 10. View largeDownload slide BPA exposure activated the canonical Wnt signaling pathway in SKOV3 and A2780 cells. Cells were exposed to 0.1% DMSO (vehicle), 10- or 100-nM BPA for 24 h. A and B, Western blotting of nuclear and cytoplasmic localization of β-catenin. Lamin A/C serves as a nuclear protein loading control, while GAPDH served as a cytoplasmic loading control. C and D, Immunofluorescence staining of β-catenin expression in nuclear and cytoplasm, demonstrating the nuclear translocation of β-catenin upon a 24-h BPA treatment. Data are representative of 3 independent experiments. DISCUSSION Up to now, most studies focused on the cancer-promoting effects of BPA at high concentrations (Fernandez et al., 2012; Hwang et al., 2011; Kang et al., 2013; Kim et al., 2015a,b; Zhang et al., 2014), which have been known well for a long time. However, the studies on the effects of low-dose BPA that more closely reflect environmental exposure levels were insufficient (Ptak and Gregoraszczuk, 2012; Ptak et al., 2014), and the molecular and cellular mechanism of low-dose BPA has not been clarified in ovarian cancer cells. Therefore, our present study used high-throughput RNA-seq technology to comprehensively elucidate the possible mechanisms underlying the effects of low-dose BPA on ovarian cancer. Numerous studies have shown that BPA exposure modulates the global gene expression that affects signal transduction pathways and cellular regulatory networks (Castillo Sanchez et al., 2016; Kim et al., 2015a), which can be successfully characterized using gene expression profile analysis. Gene expression profiles have been determined in BG-1 ovarian cancer cells (Hwang et al., 2011), Michigan Cancer Foundation-7 (MCF-7) breast cancer cells (Mesnage et al., 2017), LNCaP prostate cancer cells (Hess-Wilson et al., 2007), and human osteosarcoma (HOS) cells (Fic et al., 2015) by microarray after BPA exposure. However, limited researches performed enrichment analysis to overview the whole signal pathways of BPA (Gong et al., 2017), especially low-dose BPA in ovarian cancer. Our study first examined transcriptomic changes by RNA-seq after nanomolar BPA exposure in the human ovarian adenocarcinoma cell line SKOV3. 94 shared DEGs after exposure to 10- and 100-nM BPA were identified in this study. Rather than simply cataloged DEGs caused by BPA exposure, we sought to study the biological significance of these altered genes on biological functions and pathways. To this end, we searched DAVID databases using the 94 shared DEGs (see Supplementary Table 3) as input data. The results suggested that BPA significantly changed the expression of genes related to tumorigenesis and metastasis in ovarian carcinoma cells. Previous studies showed that the effect of BPA in proliferation usually exhibits an inverted U-shaped dose-response curve in ovarian cancer (Shi et al., 2017) and mesenchymal stem (Wang et al., 2013) cells. High-dose BPA inhibited cell growth, and a recent study reported that nanomolar-dose BPA can regulate cell proliferation positively in OVCAR-3 ovarian cancer cell lines (Shi et al., 2017). Nevertheless, our data showed that 10- and 100-nM BPA had no significant effects on the proliferation of either SKOV3 or A2780 cells (Figure 6), which was confirmed by cytotoxic tests as well (see Supplementary Figure 2). But the transcriptomic study revealed that gene expression involved in cell proliferation and apoptosis were significantly regulated by BPA. It suggested that BPA at noncytotoxic levels might cause alterations of related gene expression. Although dosing condition and exposure time in the literature can be highly variable, most studies draw a conclusion that BPA increases the migration and invasion of various cancer cell lines, such as ovarian (Kim et al., 2015b; Ptak et al., 2014), breast (Castillo Sanchez et al., 2016; Zhang et al., 2016), and cervical (Ma et al., 2015) cancer cells. Consistent with the reports in the literature, BPA treatment can significantly improve migration and invasion ability of ovarian cancer cells (Figure 7). Results from phenotypic features and functional annotation confirmed the validity of the sequencing data. Recent data indicated that BPA induces migration of BG-1 and OVCAR-3 human epithelial ovarian cancer cells (Kim et al., 2015b; Ptak et al., 2014), which are consistent with the effects of nanomolar BPA on SKOV3 and A2780 cell migration reported here. To date, there is very limited published data showing the effects of low-dose BPA exposure on the invasion ability of ovarian cancer. Previous study has demonstrated that micromolar BPA can trigger invasion in BG-1 cells (Kim et al., 2015a). Our present study revealed that exposure to low-dose BPA resulted in greater invasiveness in SKOV3 and A2780 cells for the first time, which suggested that observations were representative in various ovarian carcinoma cells. Metastasis is the main cause for deaths from cancer and is a common characteristic of advanced solid tumors (Harvey et al., 2013). It is a multistep process which commonly starts with an EMT process before dissociation from the primary tumor and cells must resist anoikis during dissemination (Harvey et al., 2013; Simpson et al., 2008). Functional annotation revealed that many relevant pathways were significantly modulated by BPA. Hippo signaling pathway (see Supplementary Figure 8) deregulation might favor metastasis, as Yes-associated Protein (YAP) over-expression can suppress anoikis of cultured cells and promote EMT at the same time (Harvey et al., 2013). However, little work has been performed on the effects of BPA on the hippo pathway and further investigation is needed. AKT is a downstream effector of integrin binding (GO: 0005178) and Focal adhesion (hsa04510) (see Supplementary Figure 9), which was enriched in DAVID database as well. Recent studies have assessed the effects of BPA on the PI3K/Akt signaling pathway and suggested that PI3K/Akt signaling pathway may be related to BPA-mediated migration in ovarian (Ptak et al., 2014) and colorectal (Chen et al., 2015) cancer cells. Cancer-associated EMT is an important process to acquire migration and invasion ability (Chaffer et al., 2016). Micromolar BPA has been reported to induce EMT of ovarian carcinoma cells through an ER-dependent pathway (Kim et al., 2015b). Results from this study revealed, for the first time, that nanomolar concentration of BPA can trigger EMT in ovarian cancer cells, leading to the down-regulation of the expression of epithelial characteristic ZO-1 and the up-regulation of the expression of mesenchymal characteristic MMP9. MMP9 is highly involved in EMT process by facilitating the protein degradation of the ECM (Polyak and Weinberg. 2009). Up-regulation of MMP9 has also been found in triple-negative breast cancer (Zhang et al., 2016). These studies suggested that environmentally equivalent doses of BPA could induce the metastasis of ovarian carcinoma cells by triggering the MMP9-dependent EMT process. The EMT process is mediated by some cellular signaling pathways, such as TGF-β, Wnt-β-catenin, receptor tyrosine kinases, Notch, Hedgehog, and microRNA pathways (Lee et al., 2017). As indicated in the KEGG pathway analysis of our study (Figure 4), Wnt signaling pathway ranked higher than TGF-β signaling pathway. Wnt/β-catenin target genes regulate cell proliferation, apoptosis, as well as EMT, consequently mediating tumorigenesis and cancer development (Arend et al., 2013). However, the effects of BPA in canonical Wnt pathway remained to be controversial. Determined by subcellular localization, aberrant catenin expression is a key marker of the activation of Wnt signaling pathway, and has been reported in male mouse reproductive cells (Fang et al., 2015). Another study demonstrated that BPA treatment suppresses Wnt/β-catenin pathway in the rat, thus impairs neural stem cells proliferation and differentiation (Tiwari et al., 2015). To confirm the functional annotation results, in this study, our results found for the first time that SKOV3 and A2780 cells treated with nanomolar-dose BPA showed significant β-catenin translocation to the nucleus, indicating that environmentally relevant-dose BPA exposure activated the canonical Wnt/β-catenin signaling pathway in ovarian carcinoma cells, and it might play a part in BPA-induced EMT process. In addition, previous studies reported that TGF-β signaling pathway is rather blocked by ER signaling resulting from BPA (10−6 M) exposure in ovarian cancer cells (Kim et al., 2015b; Park and Choi, 2014). These findings are inconsistent with our present study that BPA exposure triggered EMT and metastasis process in SKOV3 and A2780 cells. It may due to the dual role of TGF-β in carcinogenesis and development or the wide variation in BPA doses used. Thus, further study is required to elucidate the relationship between BPA and the TGF-β signaling pathway. In this study, the gene expression profile identified the differential expression of 94 genes in the SKOV3 ovarian adenocarcinoma cells following 24-h environmentally relevant-dose BPA treatment. We are aware that the major weakness of this study is that we performed transcriptomic analysis only at a single time point posttreatment, thus the recognition of the temporal effects of BPA exposure on global gene expression is not allowed. However, detailed interpretation of the biological implications of altered genes as presented here does help us understand more about the potential effects of BPA exposure on tumorigenesis and metastasis. Further studies demonstrated that BPA promoted migration and invasion and induced EMTs of ovarian carcinoma cells, which was characterized by increasing expression of MMP9 with a concomitant decrease of ZO-1. Correspondingly, BPA exposure activated the canonical Wnt signaling pathway, which might be involved in BPA-induced EMTs. Besides, the expression of miR-21 and miR-222 were up-regulated with increasing doses of BPA exposure (10 and 100 nM). Our study first comprehensively analyzed the possible mechanisms underlying the effects of BPA on ovarian cancer. Environmentally relevant doses of BPA modulated the gene expression profile, promoted EMT progress via canonical Wnt signaling pathway (Figure 11) and activated the microRNA network of ovarian cancer. Figure 11. View largeDownload slide Schematic plot showed BPA promoted EMT via canonical Wnt pathway. Figure 11. View largeDownload slide Schematic plot showed BPA promoted EMT via canonical Wnt pathway. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. ACKNOWLEDGMENT The authors thank Huiqi Chen and Yexinyi Zhou for the kind suggestions and help in the course of experiment. FUNDING This work was supported by the National Natural Science Foundation of China (Grant No: 81302455, 31471297, and 81773016), the Zhejiang Provincial Natural Science Foundation of China (Grant No: LY18C060001), and the Fundamental Research Funds for the Central Universities (No. 2017XZZX011-01). REFERENCES Arend R. C., Londono-Joshi A. I., Straughn J. M.Jr, Buchsbaum D. J. ( 2013). The Wnt/beta-catenin pathway in ovarian cancer: A review. Gynecol. Oncol.  131, 772– 779. Google Scholar CrossRef Search ADS PubMed  Calafat A. M., Kuklenyik Z., Reidy J. A., Caudill S. P., Ekong J., Needham L. L. ( 2005). Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ. Health Perspect.  113, 391– 395. Google Scholar CrossRef Search ADS PubMed  Castillo Sanchez R., Gomez R., Perez Salazar E. ( 2016). Bisphenol A induces migration through a GPER-, FAK-, Src-, and ERK2-dependent pathway in MDA-MB-231 breast cancer cells. Chem. Res. Toxicol.  29, 285– 295. Google Scholar CrossRef Search ADS PubMed  Chaffer C. L., San Juan B. P., Lim E., Weinberg R. A. ( 2016). EMT, cell plasticity and metastasis. Cancer Metastasis Rev.  35, 645– 654. Google Scholar CrossRef Search ADS PubMed  Chen Z.-J., Yang X.-L., Liu H., Wei W., Zhang K.-S., Huang H.-B., Giesy J. P., Liu H.-L., Du J., Wang H.-S.et al.  , ( 2015). Bisphenol A modulates colorectal cancer protein profile and promotes the metastasis via induction of epithelial to mesenchymal transitions. Arch. Toxicol.  89, 1371– 1381. Google Scholar CrossRef Search ADS PubMed  Chung S. H., Franceschi S., Lambert P. F. ( 2010). Estrogen and ERalpha: Culprits in cervical cancer?. Trends Endocrinol. Metab.  21, 504– 511. Google Scholar CrossRef Search ADS PubMed  Dematteo R., Keith M. M., Brophy J. T., Wordsworth A., Watterson A. E., Beck M., Rochon A., Michael F., Jyoti G., Magali P.et al.  , ( 2013). Chemical exposures of women workers in the plastics industry with particular reference to breast cancer and reproductive hazards. New Solut.  22, 427– 448. Google Scholar CrossRef Search ADS   Fang Z., Liu X., Yang X., Song X., Chen X. ( 2015). Effects of Wnt/beta-catenin signaling on bisphenol A exposure in male mouse reproductive cells. Mol. Med. Rep.  12, 5561– 5567. Google Scholar CrossRef Search ADS PubMed  Fernandez S. V., Huang Y., Snider K. E., Zhou Y., Pogash T. J., Russo J. ( 2012). Expression and DNA methylation changes in human breast epithelial cells after bisphenol A exposure. Int. J. Oncol.  41, 369– 377. Google Scholar PubMed  Fic A., Jurković Mlakar S., Juvan P., Mlakar V., Marc J., Sollner Dolenc M., Broberg K., Peterlin Mašič L. ( 2015). Genome-wide gene expression profiling of low-dose, long-term exposure of human osteosarcoma cells to bisphenol A and its analogs bisphenols AF and S. Toxicol. In Vitro  29, 1060– 1069. Google Scholar CrossRef Search ADS PubMed  Gong X., Xie H., Li X., Wu J., Lin Y. ( 2017). Bisphenol A induced apoptosis and transcriptome differences of spermatogonial stem cells in vitro. Acta Biochim. Biophys. Sin.  49, 780– 791. Google Scholar CrossRef Search ADS   Gonzalez D. M., Medici D. ( 2014). Signaling mechanisms of the epithelial-mesenchymal transition. Sci. Signal.  7, re8. Google Scholar CrossRef Search ADS PubMed  Gore A. C., Chappell V. A., Fenton S. E., Flaws J. A., Nadal A., Prins G. S., Toppari J., Zoeller R. T. ( 2015). Executive summary to EDC-2: The endocrine society’s second scientific statement on endocrine-disrupting chemicals. Endocr. Rev.  36, 593– 602. Google Scholar CrossRef Search ADS PubMed  Grady D., Gebretsadik T., Kerlikowske K., Ernster V., Petitti D. ( 1995). Hormone replacement therapy and endometrial cancer risk: A meta-analysis. Obstet. Gynecol.  85, 304– 313. Google Scholar CrossRef Search ADS PubMed  Harvey K. F., Zhang X., Thomas D. M. ( 2013). The Hippo pathway and human cancer. Nat. Rev. Cancer  13, 246– 257. Google Scholar CrossRef Search ADS PubMed  Hess-Wilson J. K., Webb S. L., Daly H. K., Leung Y.-K., Boldison J., Comstock C. E. S., Sartor M. A., Ho S.-M., Knudsen K. E. ( 2007). Unique bisphenol A transcriptome in prostate cancer: Novel effects on ERbeta expression that correspond to androgen receptor mutation status. Environ. Health Perspect.  115, 1646– 1653. Google Scholar CrossRef Search ADS PubMed  Hoffmann M., Gogola J., Kotula-Balak M., Ptak A. ( 2017). Stimulation of ovarian cell proliferation by tetrabromobisphenol A but not tetrachlorobisphenol A through G protein-coupled receptor 30. Toxicol. In Vitro  45, 54– 59. Google Scholar CrossRef Search ADS PubMed  Hwang K. A., Park S. H., Yi B. R., Choi K. C. ( 2011). Gene alterations of ovarian cancer cells expressing estrogen receptors by estrogen and bisphenol a using microarray analysis. Lab. Anim. Res.  27, 99– 107. Google Scholar CrossRef Search ADS PubMed  Ikezuki Y., Tsutsumi O., Takai Y., Kamei Y., Taketani Y. ( 2002). Determination of bisphenol A concentrations in human biological fluids reveals significant early prenatal exposure. Hum. Reprod.  17, 2839– 2841. Google Scholar CrossRef Search ADS PubMed  Kang J. H., Kito K., Kondo F. ( 2003). Factors influencing the migration of bisphenol A from cans. J. Food Prot.  66, 1444– 1447. Google Scholar CrossRef Search ADS PubMed  Kang N. H., Hwang K. A., Lee H. R., Choi D. W., Choi K. C. ( 2013). Resveratrol regulates the cell viability promoted by 17beta-estradiol or bisphenol A via down-regulation of the cross-talk between estrogen receptor alpha and insulin growth factor-1 receptor in BG-1 ovarian cancer cells. Food Chem. Toxicol.  59, 373– 379. Google Scholar CrossRef Search ADS PubMed  Kim Y. S., Choi K. C., Hwang K. A. ( 2015a). Genistein suppressed epithelial-mesenchymal transition and migration efficacies of BG-1 ovarian cancer cells activated by estrogenic chemicals via estrogen receptor pathway and downregulation of TGF-beta signaling pathway. Phytomedicine  22, 993– 999. Google Scholar CrossRef Search ADS   Kim Y. S., Hwang K. A., Hyun S. H., Nam K. H., Lee C. K., Choi K. C. ( 2015b). Bisphenol A and nonylphenol have the potential to stimulate the migration of ovarian cancer cells by inducing epithelial-mesenchymal transition via an estrogen receptor dependent pathway. Chem. Res. Toxicol.  28, 662– 671. Google Scholar CrossRef Search ADS   Lantta M. ( 1984). Estradiol and progesterone receptors in normal ovary and ovarian tumors. Acta Obstet. Gynecol. Scand.  63, 497– 503. Google Scholar CrossRef Search ADS PubMed  Lee H. M., Hwang K. A., Choi K. C. ( 2017). Diverse pathways of epithelial mesenchymal transition related with cancer progression and metastasis and potential effects of endocrine disrupting chemicals on epithelial mesenchymal transition process. Mol. Cell Endocrinol.  457, 103– 113. Google Scholar CrossRef Search ADS PubMed  Lu W., Zhang H., Niu Y., Wu Y., Sun W., Li H., Kong J., Ding K., Shen H.-M., Wu H.et al.  , ( 2017). Long non-coding RNA linc00673 regulated non-small cell lung cancer proliferation, migration, invasion and epithelial mesenchymal transition by sponging miR-150-5p. Mol. Cancer  16, 118. Google Scholar CrossRef Search ADS PubMed  Ma X. F., Zhang J., Shuai H. L., Guan B. Z., Luo X., Yan R. L. ( 2015). IKKbeta/NF-kappaB mediated the low doses of bisphenol A induced migration of cervical cancer cells. Arch. Biochem. Biophys.  573, 52– 58. Google Scholar CrossRef Search ADS PubMed  Mesnage R., Phedonos A., Arno M., Balu S., Corton J. C., Antoniou M. N. ( 2017). Editor’s highlight: Transcriptome profiling reveals bisphenol A alternatives activate estrogen receptor alpha in human breast cancer cells. Toxicol. Sci.  158, 431– 443. Google Scholar CrossRef Search ADS PubMed  Missmer S. A., Eliassen A. H., Barbieri R. L., Hankinson S. E. ( 2004). Endogenous estrogen, androgen, and progesterone concentrations and breast cancer risk among postmenopausal women. J. Natl. Cancer Inst.  96, 1856– 1865. Google Scholar CrossRef Search ADS PubMed  Moustakas A., Heldin C. H. ( 2007). Signaling networks guiding epithelial-mesenchymal transitions during embryogenesis and cancer progression. Cancer Sci.  98, 1512– 1520. Google Scholar CrossRef Search ADS PubMed  National Academies of Sciences, Engineering, and Medicine. ( 2016) Ovarian Cancers: Evolving Paradigms in Research and Care , pp. 3. The National Academies Press, Washington, DC. Park M. A., Choi K. C. ( 2014). Effects of 4-nonylphenol and bisphenol A on stimulation of cell growth via disruption of the transforming growth factor-beta signaling pathway in ovarian cancer models. Chem. Res. Toxicol.  27, 119– 128. Google Scholar CrossRef Search ADS PubMed  Park S.-H., Kim K.-Y., An B.-S., Choi J.-H., Jeung E.-B., Leung P. C. K., Choi K.-C. ( 2009). Cell growth of ovarian cancer cells is stimulated by xenoestrogens through an estrogen-dependent pathway, but their stimulation of cell growth appears not to be involved in the activation of the mitogen-activated protein kinases ERK-1 and p38. J. Reprod. Dev.  55, 23. Google Scholar CrossRef Search ADS PubMed  Polyak K., Weinberg R. A. ( 2009). Transitions between epithelial and mesenchymal states: Acquisition of malignant and stem cell traits. Nat. Rev. Cancer  9, 265– 273. Google Scholar CrossRef Search ADS PubMed  Ptak A., Gregoraszczuk E. L. ( 2012). Bisphenol A induces leptin receptor expression, creating more binding sites for leptin, and activates the JAK/Stat, MAPK/ERK and PI3K/Akt signalling pathways in human ovarian cancer cell. Toxicol. Lett.  210, 332– 337. Google Scholar CrossRef Search ADS PubMed  Ptak A., Hoffmann M., Gruca I., Barc J. ( 2014). Bisphenol A induce ovarian cancer cell migration via the MAPK and PI3K/Akt signalling pathways. Toxicol. Lett.  229, 357– 365. Google Scholar CrossRef Search ADS PubMed  Ptak A., Wrobel A., Gregoraszczuk E. L. ( 2011). Effect of bisphenol-A on the expression of selected genes involved in cell cycle and apoptosis in the OVCAR-3 cell line. Toxicol. Lett.  202, 30– 35. Google Scholar CrossRef Search ADS PubMed  Rochester J. R. ( 2013). Bisphenol A and human health: A review of the literature. Reprod. Toxicol.  42, 132– 155. Google Scholar CrossRef Search ADS PubMed  Rodriguez C., Patel A. V., Calle E. E., Jacob E. J., Thun M. J. ( 2001). Estrogen replacement therapy and ovarian cancer mortality in a large prospective study of US women. Jama  285, 1460– 1465. Google Scholar CrossRef Search ADS PubMed  Seachrist D. D., Bonk K. W., Ho S. M., Prins G. S., Soto A. M., Keri R. A. ( 2016). A review of the carcinogenic potential of bisphenol A. Reprod. Toxicol.  59, 167– 182. Google Scholar CrossRef Search ADS PubMed  Shi X.-Y., Wang Z., Liu L., Feng L.-M., Li N., Liu S., Gao H. ( 2017). Low concentrations of bisphenol A promote human ovarian cancer cell proliferation and glycolysis-based metabolism through the estrogen receptor-alpha pathway. Chemosphere  185, 361– 367. Google Scholar CrossRef Search ADS PubMed  Siegel R. L., Miller K. D., Jemal A. ( 2018). Cancer statistics, 2018. CA Cancer J. Clin.  68, 7– 30. Google Scholar CrossRef Search ADS PubMed  Simpson C. D., Anyiwe K., Schimmer A. D. ( 2008). Anoikis resistance and tumor metastasis. Cancer Lett.  272, 177– 185. Google Scholar CrossRef Search ADS PubMed  Tiwari S. K., Agarwal S., Seth B., Yadav A., Ray R. S., Mishra V. N., Chaturvedi R. K. ( 2015). Inhibitory effects of bisphenol-A on neural stem cells proliferation and differentiation in the rat brain are dependent on wnt/beta-catenin pathway. Mol. Neurobiol.  52, 1735– 1757. Google Scholar CrossRef Search ADS PubMed  Uhlen M., Zhang C., Lee S., Sjöstedt E., Fagerberg L., Bidkhori G., Benfeitas R., Arif M., Liu Z., Edfors F.et al.  , ( 2017). A pathology atlas of the human cancer transcriptome. Science  357, eaan2507. Available at: www.proteinatlas.org. Accessed 8 October 2017. Google Scholar CrossRef Search ADS PubMed  Vandenberg L. N., Maffini M. V., Wadia P. R., Sonnenschein C., Rubin B. S., Soto A. M. ( 2007). Exposure to environmentally relevant doses of the xenoestrogen bisphenol-A alters development of the fetal mouse mammary gland. Endocrinology  148, 116– 127. Google Scholar CrossRef Search ADS PubMed  Wang K. H., Kao A. P., Chang C. C., Lin T. C., Kuo T. C. ( 2013). Bisphenol A at environmentally relevant doses induces cyclooxygenase-2 expression and promotes invasion of human mesenchymal stem cells derived from uterine myoma tissue. Taiwan J. Obstet. Gynecol.  52, 246– 252. Google Scholar CrossRef Search ADS PubMed  Willcocks D., Toppila M., Hudson C. N., Tyler J. P., Baird P. J., Eastman C. J. ( 1983). Estrogen and progesterone receptors in human ovarian tumors. Gynecol. Oncol.  16, 246– 253. Google Scholar CrossRef Search ADS PubMed  Willert K., Nusse R. ( 1998). Beta-catenin: A key mediator of Wnt signaling. Curr. Opin. Genet. Dev.  8, 95– 102. Google Scholar CrossRef Search ADS PubMed  Zhang K.-S., Chen H.-Q., Chen Y.-S., Qiu K.-F., Zheng X.-B., Li G.-C., Yang H.-D., Wen C.-J. ( 2014). Bisphenol A stimulates human lung cancer cell migration via upregulation of matrix metalloproteinases by GPER/EGFR/ERK1/2 signal pathway. Biomed. Pharmacother.  68, 1037– 1043. Google Scholar CrossRef Search ADS PubMed  Zhang X. L., Liu N., Weng S. F., Wang H. S. ( 2016). Bisphenol A increases the migration and invasion of triple-negative breast cancer cells via oestrogen-related receptor gamma. Basic Clin. Pharmacol. Toxicol.  119, 389– 395. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Toxicological SciencesOxford University Press

Published: Apr 27, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off