Metabolites of n-Butylparaben and iso-Butylparaben Exhibit Estrogenic Properties in MCF-7 and T47D Human Breast Cancer Cell Lines

Metabolites of n-Butylparaben and iso-Butylparaben Exhibit Estrogenic Properties in MCF-7 and... Abstract Two oxidized metabolites of n-butylparaben (BuP) and iso-butylparaben (IsoBuP) discovered in human urine samples exhibit structural similarity to endogenous estrogens. We hypothesized that these metabolites bind to the human estrogen receptor (ER) and promote estrogen signaling. We tested this using models of ER-mediated cellular proliferation. The estrogenic properties of 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) were determined using the ER-positive, estrogen-dependent human breast cancer cell lines MCF-7, and T47D. The 3OH metabolite induced cellular proliferation with EC50 of 8.2 µM in MCF-7 cells. The EC50 for 3OH in T47D cells could not be reached. The 2OH metabolite induced proliferation with EC50 of 2.2 µM and 43.0 µM in MCF-7 and T47D cells, respectively. The EC50 for the parental IsoBuP and BuP was 0.30 and 1.2 µM in MCF-7 cells, respectively. The expression of a pro-proliferative, estrogen-inducible gene (GREB1) was induced by these compounds and blocked by co-administration of an ER antagonist (ICI 182, 780), confirming the ER-dependence of these effects. The metabolites promoted significant ER-dependent transcriptional activity of an ERE-luciferase reporter construct at 10 and 20 µM for 2OH and 10 µM for 3OH. Computational docking studies showed that the paraben compounds exhibited the potential for favorable ligand-binding domain interactions with human ERα in a manner similar to known x-ray crystal structures of 17ß-estradiol in complex with ERα. We conclude that the hydroxylated metabolites of BuP and IsoBuP are weak estrogens and should be considered as additional components of potential endocrine disrupting effects upon paraben exposure. breast cancer, endocrine, estrogens, paraben, environmental Breast cancer is currently the second most common malignancy diagnosed in the United States (US) with an estimated 250 000 new cases and 40 000 disease-related deaths in 2016 (Siegel et al., 2017). An estimated 7% of breast cancer diagnoses are attributed to an inherited genetic predisposition with the remaining ∼93% being attributed to other likely risk factors such as lifestyle choices, obesity, or exposure to environmental carcinogens (Schwartz et al., 2008). In addition, lifetime exposure to estrogens has been shown to be correlated with breast cancer risk (Lippman, 2001). Both preclinical and clinical studies have shown that estrogens, specifically 17ß-estradiol (E2), can induce breast cancer pathogenesis (Lippman and Bolan, 1975; Lippman et al., 1976). The ability of estrogen to induce breast cancer cellular proliferation is concerning because it is estimated that 70%–80% of all diagnosed breast cancers express the estrogen receptor (ER) and are deemed ER-positive (Lu et al., 2016). Pharmacological approaches have been developed to block ER signaling using ER antagonists such as tamoxifen, and more recently aromatase inhibitors (AIs), which inhibit the synthesis of estrogens (Sikora et al., 2012), have been used. While highly effective, de novo and acquired resistance to AIs is common, and a possible mechanism of resistance is attributed to exposure to environmental estrogen mimicking chemicals, or xenoestrogens (Pastor-Barriuso et al., 2016; Sikora et al., 2009) including the ones we chose to study in this report (Moos et al., 2016a). Recently, there has been increasing public awareness and research into environmental chemicals that may have biologically relevant, endocrine disrupting properties (Delfosse et al., 2014). Xenoestrogens include many of these suspected endocrine disrupting compounds that have previously been shown to display agonistic behavior toward estrogen receptor-α (ERα) (Delfosse et al., 2015; Wielogorska et al., 2015). Alkyl esters of p-hydroxybenzoic acid (parabens) are one type of xenoestrogen that have been investigated for whether current human exposure levels are a cause for concern (Sasseville et al., 2015). Given their antimicrobial properties, parabens are frequently used as a preservative in numerous pharmaceuticals, food products, and personal care products (Guo and Kannan, 2013; Liao et al., 2013). Compared to oral ingestion, dermal contact encompasses a broader range of paraben exposure sources that are found in numerous cosmetics and personal care products (Dodson et al., 2012). As an example, daily paraben exposure for a 63-kg average adult in Korea is estimated to be a sum total of 18 960 µg/day for methyl and ethylparaben combined, and 1580 µg/day for propyl paraben alone (Kang et al., 2016). Similarly, dermal intake of parabens has been estimated to be 31 µg/kg-bw/day for adult females and range between 58.6 and 766 µg/kg-bw/day among infants and toddlers within US in 2012 (Guo and Kannan, 2013). Exposure estimates for parabens in Belgium, Germany, and several other nations have been summarized and reported elsewhere (Calafat et al., 2010; Csiszar et al., 2016; Dewalque et al., 2014; Moos et al., 2016b). The ubiquitous exposure to parabens has led to reports of their detection in breast tissue, breast milk, placental tissue, serum, seminal fluid, and urine samples from numerous general populations around the world (Azzouz et al., 2016; Barr et al., 2012; Frederiksen et al., 2011; Hines et al., 2015; Moos et al., 2016b; Valle-Sistac et al., 2016). In spite of evidence showing the presence of parabens in various human tissues such as adipose (Artacho-Cordón et al., 2017) and breast tissue (Charles and Darbre, 2013), there is still significant debate over their current risk to the general population. To date, studies have focused on the estrogenic properties of paraben parent compounds (Okubo et al., 2001; Wielogorska et al., 2015; Wróbel and Gregoraszczuk, 2013, 2014). Two oxidized paraben metabolites, 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) (Figure 1), have recently been discovered, but their potential estrogenic properties have not yet been studied (Moos et al., 2016a). Unlike E2, which forms 3 hydrogen bonds in the ligand binding pocket of human ER according to X-ray crystallography structures (Delfosse et al., 2012), BuP is unable to hydrogen bond with His524 (Delfosse et al., 2014). The lack of a second hydroxyl group on BuP likely contributes to its relatively weak ER binding affinity. We hypothesized that the metabolites’ additional hydroxyl group might enable binding interactions similar to that of 17ß-estradiol and confer increased estrogenic potency compared to their respective parent compounds. To test this, we used pre-clinical in vitro models of ER-positive breast cancer and demonstrated that the oxidized metabolites promote cell proliferation in an ER-dependent manner comparable to their respective parent compounds. Computational docking studies indicated that the metabolites display hydrogen bonding capabilities similar to 17ß-estradiol within the human ERα ligand-binding domain in support of our in vitro data. Figure 1. View largeDownload slide Chemical structures of 17ß-estradiol, 3OH, and 2OH paraben metabolites. * indicates chiral center. Figure 1. View largeDownload slide Chemical structures of 17ß-estradiol, 3OH, and 2OH paraben metabolites. * indicates chiral center. MATERIALS AND METHODS Reagents 17ß-estradiol (E2), ICI 182, 780 (ICI), and n-butylparaben (BuP) were purchased from Sigma-Aldrich Inc. (St. Louis, MO) with purity ≥98% determined by HPLC. E2 and ICI were dissolved to 10 mM in absolute ethanol and stored at −20°C. Iso-butylparaben (IsoBuP) was purchased from TCI America (Portland, OR) with purity ≥98% determined by HPLC. 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) were kindly provided as a gift by Dr. Vladimir Belov, Max Planck Institute for Biophysical Chemistry in Germany. The 3OH and 2OH compounds had a purity ≥95% determined by HPLC with UV (254 nm) detection. BuP, IsoBuP, 3OH, and 2OH were dissolved to 100 mM in DMSO stored at −20°C. The final concentration of ethanol or DMSO did not exceed 0.1% (v/v) in culture media. All compounds were stored protected from light. Cell lines, culture conditions, and proliferation assays MCF-7 and T47D cells were obtained from the Tissue Culture Shared Resource (TCSR) at the Lombardi Comprehensive Cancer Center (Georgetown University, Washington, DC). Cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco/Life Technologies, Grand Island, NY) supplemented with 10% (v/v) fetal calf serum (Valley Biomedical Inc., Winchester, VA), at 37°C in a humidified 5% (v/v) CO2 atmosphere. The identity of the cells was confirmed by short tandem repeat profiling by the TCSR, and shown to be free of mycoplasma contamination. For assays in defined steroid deplete conditions, cells were repeatedly washed and grown in steroid depleted media over 3 days before proliferation assays (phenol red-free IMEM supplemented with 10% [v/v] charcoal stripped bovine calf serum—CCS) (Valley Biomedical Inc., Winchester, VA) based on a previously described method (Rae et al., 2005). For growth assays, MCF-7 and T47D cells were withdrawn from steroids as previously described (Johnson et al., 2004) and plated in steroid-free media at 1000 and 2000 cells/well, respectively, in 96-well plates and allowed to attach overnight. PrestoBlue cell viability assay MCF-7 and T47D cells were plated in steroid-free media at 1000 and 2000 cells/well, respectively, in 96-well plates and cultured overnight. The following day, cells were treated with the specified compounds at the indicated concentrations and the vehicle controls (ethanol or DMSO) diluted in IMEM supplemented with 10% (v/v) CCS. 6 days after treatment, cell viability was assessed using Presto Blue reagent (ThermoFisher Scientific, Waltham, MA) according to manufacturer’s instructions. In brief, cells were incubated for 3 h in the presence of 10% (v/v) Presto Blue and fluorescence was measured using a POLARstar Omega plate reader with the excitation/emission wavelengths set at 544/590 nm. RNA expression assay MCF-7 cells were repeatedly washed and grown in steroid depleted media over 3 days as described above and plated in steroid depleted media at 400 000 cells/well in 6-well plates at least 10 h before treatment with parabens for 2, 4, and 6 h durations. Total RNA was isolated using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. Yield and quantity were determined by spectrophotometry (NanoDrop ND-1000). All samples were stored at −80°C. Total RNA (1 µg) was reverse transcribed (RT) using Reverse Transcription System (Promega, Madison, WI) and the cDNA amplified in a 25 µl reaction containing Gene Expression Master Mix and gene specific primers both from Thermo Fisher Scientific (Waltham, MA). GREB1 mRNA expression was measured using a TaqMan RT-PCR assay as described previously (Rae et al., 2005). GREB1 expression was normalized against GAPDH with relative expression being calculated using the ΔΔCT method (Livak and Schmittgen, 2001). Estrogen-response element (ERE)–luciferase reporter assay The ERE–luciferase reporter plasmid (a gift from Dr. Anna T. Riegel, Georgetown University) was co-transfected into MCF-7 cells with Fugene 6 transfection reagent (Promega, Madison, WI) according to manufacturer’s instructions with the renilla plasmid (100:1 v:v luciferase to renilla). After 18 h, transfected cells were repeatedly washed with phenol red-free 10% (v/v) CCS I-MEM (Valley Biomedical Inc., Winchester, VA) to produce a steroid-free environment. The following day cells were plated on a 24-well plate (Corning/Costar) in 10% (v/v) CCS I-MEM (Valley Biomedical Inc., Winchester, VA). After 6 h, cells were treated with either 2OH, 3OH, BuP, IsoBuP, E2, or a vehicle control (ethanol or DMSO). After 18 h, cell lysates were processed and relative light unit reading were measured according to manufacturer’s recommendations on a Leader 50 luminometer (Gen-Probe). Molecular docking The crystal structure of the human ERα ligand-binding domain in complex with 17ß-estradiol was downloaded from the protein data bank PDB ID: 3UUD and prepared for docking with the YASARA molecular modeling suite (Krieger and Vriend, 2014). The PDB entry 3UUD was selected for docking due to its high resolution (1.6 Å). The presence of an endogenous ligand was another factor considered when selecting this structure to ensure that the ligand-binding domain was a suitable representation of the agonist conformation. All crystallographic waters were removed, hydrogen atoms were added, bond orders were corrected for S-hydroxycysteine residues, missing loops were repaired (Canutescu and Dunbrack, 2003), and only the A-chain was retained. Docking was performed with VINA (Trott and Olson, 2010) using default parameters, and setup was conducted with YASARA as the graphical front-end (Krieger et al., 2002). A grid size of 25 Å  × 25 Å  × 25 Å was centered on the C9 carbon of E2 before removing the native ligand and set as the search space for the paraben compounds treated as flexible ligands. The best pose of 25 runs was defined as producing the smallest root-mean-square deviation (RMSD) between the benzoyl group of the paraben compounds and comparable carbon atoms (1–5, 9, & 10) and phenol oxygen atom of the phenolic-A ring of E2. Hydrogen bond networks of the docked structures were optimized with YASARA before determining RMSD values (Krieger et al., 2012). Statistical analysis The curve-fitting program GraphPad Prism 7.0 was used to plot all data. The time and relative fold expression of GREB1 within each treatment group was processed by one-way ANOVA (p < .005) followed by a Bonferroni post hoc analysis to determine which time points differed from each other. RESULTS 3OH and 2OH Butyl Paraben Metabolites Induce the Proliferation of Estrogen-Dependent Human Breast Cancer Cells Grown in Steroid-Free Conditions To determine the whether the paraben metabolites could induce human breast cancer growth, estrogen-dependent MCF-7 and T47D cell lines were treated with BuP, IsoBuP, and their respective metabolites. The calculated effective concentrations inducing 50% growth (EC50) for BuP and IsoBuP (1.2 and 0.30 µM, respectively) were similar to those reported previously (Charles and Darbre, 2013; Delfosse et al., 2014; Evans et al., 2012; Watanabe et al., 2013) (Supplementary Figs. 1A–D). MCF-7 cells treated with the 3OH (Figure 2A) and 2OH (Figure 2B) paraben metabolites induced cellular proliferation in a concentration-dependent manner. T47D cells treated with 3OH (Figure 2C) or 2OH (Figure 2D) required slightly higher concentrations than MCF-7 cells to induce proliferation but they did not achieve a sigmoidal concentration dependence. Higher concentrations of the 3OH metabolite would exceed 0.1% DMSO vehicle concentration which prevented an EC50 from being determined in the T47D cell line. The EC50 for 3OH and 2OH in MCF-7 and T47D cells was calculated (3OH: EC50 8.0 µM in MCF-7; 2OH: EC50 2.2 µM in MCF-7 and 43.0 µM in T47D). All PrestoBlue cell viability assay results were confirmed with the crystal violet growth assay as described in Materials and Methods section (data not shown). Previous reports have investigated the androgenic and anti-androgenic behavior of parabens (Chen et al., 2007; Ma et al., 2014) prior to the discovery of the 3OH and 2OH paraben metabolites. Therefore, we also examined whether the 3OH and 2OH paraben metabolites exhibited androgenic properties by treating androgen-dependent LNCaP cells and we found that the paraben metabolites do not exhibit androgenic behavior at concentration ranges of 10 pM–10 µM (Supplementary Figs. 2A and 2B). Non-ER α expressing MDA-MB-231 cells were treated with 3OH, 2OH, and their respective parent compounds to show no effect on cell proliferation. (Supplementary Figs. 3A–D). Figure 2. View largeDownload slide 3OH and 2OH paraben metabolites induce breast cancer cell proliferation. MCF-7 and T47D cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. MCF-7 cells were treated with either 3OH (A) or 2OH (B) and T47D cells were treated with 3OH (C) or 2OH (D) at the indicated concentrations. Growth curves for A–D represent percentage of cell growth compared to DMSO (vehicle) control (0%). Points on dose response curve represent 6-day growth versus vehicle treated control ± SE (n = 6 technical replicates). Dotted line indicates growth induced by 100 pM E2. Figure 2. View largeDownload slide 3OH and 2OH paraben metabolites induce breast cancer cell proliferation. MCF-7 and T47D cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. MCF-7 cells were treated with either 3OH (A) or 2OH (B) and T47D cells were treated with 3OH (C) or 2OH (D) at the indicated concentrations. Growth curves for A–D represent percentage of cell growth compared to DMSO (vehicle) control (0%). Points on dose response curve represent 6-day growth versus vehicle treated control ± SE (n = 6 technical replicates). Dotted line indicates growth induced by 100 pM E2. The Pure Anti-Estrogen Receptor Antagonist (ICI 182, 780) Inhibits 3OH and 2OH Metabolite Induced Proliferation of Estrogen-Dependent Human Breast Cancer To confirm that the cell proliferation induced by 3OH and 2OH paraben metabolites was due to ERα signaling, we assessed whether their proliferative effects could be blocked by the pure anti-estrogen ICI 182, 780 (Fulvestrant). Estrogen-dependent MCF-7 cells were treated at the identified EC50 concentrations of BuP (1.2 µM), IsoBuP (0.3 µM), and their metabolites either alone or in the presence of increasing concentrations of ICI. BuP and IsoBuP induced proliferation was inhibited by ICI at 0.8 and 1.4 nM, respectively, comparable to previous reports (Okubo et al., 2001) (Figs. 3A and B). Cell proliferation induced by the 3OH and 2OH paraben metabolites was inhibited in a concentration-dependent manner by ICI with an IC50 of 0.7 and 1.2 nM, respectively (Figs. 3A and B). Figure 3. View largeDownload slide Pure anti-estrogen blocks 3OH, 2OH, respective parent compound induced breast cancer cell proliferation. MCF-7 cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. Growth induction by a fixed dose of either (A) (▪) 0.3 µM IsoBuP (solid line) or (□) 2.15 µM 2OH (dashed line) or (B) by (▪) 1.2 µM Butylparaben or (□) 8.22 µM 3OH was antagonized by the pure anti-estrogen, ICI 182, 780 (ICI). ICI was added to final concentrations ranging from 1 pM to 1 µM at log intervals. Growth curves represents percentage of cell growth compared to 2OH or IsoBuP at the fixed concentrations indicated above. Data are normalized from maximum growth (100%) to minimum growth (0%) for each treatment. Points on dose response curve represent Points represent 6-day growth versus proliferation with EC50 of indicated paraben without ICI ± SE (n = 6 technical replicates). Figure 3. View largeDownload slide Pure anti-estrogen blocks 3OH, 2OH, respective parent compound induced breast cancer cell proliferation. MCF-7 cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. Growth induction by a fixed dose of either (A) (▪) 0.3 µM IsoBuP (solid line) or (□) 2.15 µM 2OH (dashed line) or (B) by (▪) 1.2 µM Butylparaben or (□) 8.22 µM 3OH was antagonized by the pure anti-estrogen, ICI 182, 780 (ICI). ICI was added to final concentrations ranging from 1 pM to 1 µM at log intervals. Growth curves represents percentage of cell growth compared to 2OH or IsoBuP at the fixed concentrations indicated above. Data are normalized from maximum growth (100%) to minimum growth (0%) for each treatment. Points on dose response curve represent Points represent 6-day growth versus proliferation with EC50 of indicated paraben without ICI ± SE (n = 6 technical replicates). 3OH and 2OH Metabolites Induce Expression of the Estrogen-Regulated Gene, GREB1 To confirm the estrogen agonist action of the paraben metabolites we tested the ability of these compounds to induce the expression of an estrogen responsive gene, GREB1. We measured the effects of paraben-treated ER-positive MCF-7 cell lines on GREB1 mRNA levels. GREB1 is a critical downstream target of ERα signaling the expression of which is induced by exposure to E2, which can be suppressed by ICI 182, 780 as previously described (Rae et al., 2005). GREB1 has also been described as a critical estrogen-specific ER-interacting protein, an interaction that has been shown to be highly enriched upon estrogen exposure (Mohammed et al., 2013). ER-positive MCF-7 cells were treated at 10 µM with BuP, IsoBuP, 3OH, and 2OH as described in Materials and Methods section. E2 was used as a positive control; it induced GREB1 expression (∼29-fold at 6 h; p < .001) at all time-points compared with the vehicle control (Figure 4). Treatment with BuP, IsoBuP, 2OH, and 3OH induced GREB1 expression ∼30, ∼36, ∼20, and ∼10-fold at 6 h, respectively (p < .001) compared to vehicle control. Neither ethanol nor DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Co-treatment with the paraben compounds and the pure anti-estrogen ICI 182, 780 at 6 h blocked these effects (Figure 5). Figure 4. View largeDownload slide Time course induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), or 100 pM E2, for 2, 4, and 6 h. Neither ethanol or DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Cells were also treated at 10 µM with butylparaben (BuP), IsoBuP (IsoBuP), 3OH, and 2OH. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance within each treatment group was determined by one-way ANOVA followed by Bonferroni post-hoc analysis. # = not significant; a, b, c, d, e = p ≤ .005. Figure 4. View largeDownload slide Time course induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), or 100 pM E2, for 2, 4, and 6 h. Neither ethanol or DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Cells were also treated at 10 µM with butylparaben (BuP), IsoBuP (IsoBuP), 3OH, and 2OH. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance within each treatment group was determined by one-way ANOVA followed by Bonferroni post-hoc analysis. # = not significant; a, b, c, d, e = p ≤ .005. Figure 5. View largeDownload slide Induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites is blocked in the presence of ICI 182, 780. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), 100 pM E2, or 100 nM ICI for 6 h. Cells were also treated at 10 µM with BuP, IsoBuP (IsoBuP), 3OH, and 2OH alone or in combination with 100 nM ICI. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance between each lettered pair was determined by t-test and significant at p < .001. Figure 5. View largeDownload slide Induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites is blocked in the presence of ICI 182, 780. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), 100 pM E2, or 100 nM ICI for 6 h. Cells were also treated at 10 µM with BuP, IsoBuP (IsoBuP), 3OH, and 2OH alone or in combination with 100 nM ICI. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance between each lettered pair was determined by t-test and significant at p < .001. 3OH and 2OH Metabolites Induce Luciferase Expression in MCF-7 Cells Transfected With an Estrogen-Responsive Luciferase Reporter Construct To further confirm that the estrogenic activity of the paraben metabolites is mediated by classical ER mediated signaling, we tested the ability of the compounds to induce ER-dependent transcription using an ERE-luciferase reporter construct (El-Ashry et al., 1996). The 2OH metabolite significantly (p < .05) induced transcriptional activity of the ERE-luciferase reporter construct at 10 and 20 µM (Figure 6). Induced transcriptional activity by the 3OH metabolite was significant at 20 µM (p < .05) but not at 10 µM. BuP and IsoBuP also demonstrated increased ERE transcriptional activity consistent with previous reports (Vo et al., 2011; Wielogorska et al., 2015) (data not shown). Figure 6. View largeDownload slide 3OH and 2OH paraben metabolites promote significant ERE-luciferase transcriptional activity. The 2OH and 3OH paraben metabolites significantly induce ERE-luciferase activity. MCF-7 cells were co-transfected with the ERE-luciferase construct and renilla reporter plasmid in estrogen-free conditions and treated at the indicated concentrations for each compound. Relative firefly luciferase activity was plotted over renilla luciferase activity induced by treatment from each specified compound versus ethanol vehicle control for E2 and DMSO vehicle control for each paraben compound. Statistical significance between each indicated pair was determined by t-test. Bars represent the mean from 3 technical replicates ± SE. ns = not significant; *p ≤ .05; **p ≤ .001. Figure 6. View largeDownload slide 3OH and 2OH paraben metabolites promote significant ERE-luciferase transcriptional activity. The 2OH and 3OH paraben metabolites significantly induce ERE-luciferase activity. MCF-7 cells were co-transfected with the ERE-luciferase construct and renilla reporter plasmid in estrogen-free conditions and treated at the indicated concentrations for each compound. Relative firefly luciferase activity was plotted over renilla luciferase activity induced by treatment from each specified compound versus ethanol vehicle control for E2 and DMSO vehicle control for each paraben compound. Statistical significance between each indicated pair was determined by t-test. Bars represent the mean from 3 technical replicates ± SE. ns = not significant; *p ≤ .05; **p ≤ .001. 3OH and 2OH Metabolites Can Be Docked Within the Ligand-Binding Pocket of ERα To provide additional support that the hydroxylated metabolites were promoting ERE-mediated transcription and stimulated cellular proliferation, molecular docking was used to characterize the interactions of the metabolites as flexible ligands within the known active conformation of the ligand-binding site of human ERα. The E2 ligand in the crystal structure of ERα (PDB ID: 3UUD) was removed and re-docked into the active site. The RMSD among all carbon and oxygen atoms for the binding pose of the docked E2 ligand and the pose identified in the crystal structure of ERα was 0.819 Å, indicating that our prepared ERα structure could accurately reproduce experimentally determined binding poses. The high resolution of the 3UUD ERα structure (1.6 Å), compared to PDB entries 4TV1 (1.85 Å) and 4MG9 (2.0 Å) ERα structures complexed with either propyl- or BuP, respectively, was a significant contributing factor in choosing this structure for docking the hydroxylated paraben metabolites. Docking of either propyl- and or BuP to 4TV1 and 4MG9 resulted in relatively poorer RMSD values compared to 3UUD which was another consideration for proceeding with the 3UUD structure for this study. The binding poses of the hydroxylated paraben metabolites and their respective parent compound were analyzed, and each of them displayed a preference to dock to the reported active site of ERα (Delfosse et al., 2014). Hydrogen bonding interactions of the paraben compounds with key side-chain residues in the active site of ERα are shown in Figures 7A–C and Supplementary Figures 4A–C. The 2OH compound displayed hydrogen bonding capabilities with amino acid residues Arg394 and Glu353 (Figure 7A) which have been shown to hydrogen bond to the phenol ring of E2 (Delfosse et al., 2012, 2015). 2D interaction diagrams generated with Discovery Studio Visualizer indicated the potential for 2 additional hydrogen bonds to be formed with L387 and G521 (Supplementary Figure 4A). The calculated RMSD for E2 and the 2OH metabolite was determined as described in the Materials and Methods section and indicated a high degree of binding similarity with E2 (0.658 Å). Due to the presence of a chiral center located at the secondary alcohol group of the 3OH compound and unknown proportions of a potential racemic mixture of our 3OH test substance, it was necessary to dock both the R and S isomer of the 3OH compound. The identified binding poses of the R & S isomers both suggested favorable hydrogen bonding interactions with amino acids Arg394 and Glu353, but the S isomer was additionally capable of forming hydrogen bonds with His524 and Gly521 (Figs. 7B and C). 2D interaction diagrams showed that the R & S isomer may both form additional hydrogen bonds with L387 but only the S isomer may form an additional hydrogen bond with L525 (Supplementary Figs. 4B and 4C). The calculated RMSD for E2 compared with the 3OH R & S isomers both indicated a high degree of binding similarity with E2 (0.905 and 0.562 Å, respectively) and determined as described in the Materials and Methods section. The calculated RMSD for E2 between each of the parent compounds BuP and IsoBuP demonstrated comparable binding similarity as the metabolites (1.001 and 0.976 Å, respectively). The calculated RMSD for each of the metabolite and parent compound pairs also indicated; high binding similarity with each other (2OH & IsoBuP: 0.123 Å; 3OH R & BuP: 0.293 Å; 3OH S & BuP: 0.555 Å). Figure 7. View largeDownload slide 3OH and 2OH paraben metabolites dock to ERα. Human ERα ligand-binding domain docked with (A) 2OH isomer (B) 3OH R isomer or (C) 3OH S isomer. 2OH and 3OH are colored orange and magenta, respectively. 17ß-estradiol has been overlaid with each ligand in grey for comparison. Hydrogen bonds are represented as yellow dashes. Oxygen and nitrogen atoms are colored in red and blue, respectively. The key structural features and hydrogen bonding residues are displayed and labeled. Figure 7. View largeDownload slide 3OH and 2OH paraben metabolites dock to ERα. Human ERα ligand-binding domain docked with (A) 2OH isomer (B) 3OH R isomer or (C) 3OH S isomer. 2OH and 3OH are colored orange and magenta, respectively. 17ß-estradiol has been overlaid with each ligand in grey for comparison. Hydrogen bonds are represented as yellow dashes. Oxygen and nitrogen atoms are colored in red and blue, respectively. The key structural features and hydrogen bonding residues are displayed and labeled. DISCUSSION In addition to the numerous urine biomonitoring studies of parabens in diverse populations around the world (Artacho-Cordón et al., 2017; Calafat et al., 2010; Kang et al., 2016; Moos et al., 2016b), in vivo studies using male rats suggest potential disruption of testosterone and sperm production from propyl- and BuP exposures (Oishi, 2001, 2002). In contrast, human epidemiological studies have not conclusively demonstrated that paraben concentrations in urine are associated with altered male steroid hormone levels or with indicators of sperm DNA damage (Meeker, 2010; Nishihama et al., 2017). However, a common finding among studies measuring paraben compounds in urine is the several fold higher paraben exposure among females compared to males (Guo and Kannan, 2013; Meeker, 2010; Moos et al., 2016b). Urine studies have reported median concentrations of paraben compounds in urine of females generally at low nanomolar concentrations including 1.0–1.9 µM for methylparaben. 41.0–250.0 nM for propylparaben, and 3.1–3.6 nM for butylparaben (Nishihama et al., 2016; Watkins et al., 2015). One report found significantly higher urine BuP levels (as a marker of higher exposure) being associated with shorter menstrual cycle length among female Japanese women age 19–22 despite only reporting a median concentration of 3.6 nM for this compound (Nishihama et al., 2016) which is 2–3 orders of magnitude lower than the EC50 for butylparaben as determined in our study. A larger study showed an association between higher levels of estrogen metabolites in urine, such as estrone sulfate and estrone glucuronide, and shorter menstrual cycles in a group of women that were part of the Women’s Reproductive Health Study (Windham et al., 2002). Therefore, the presence of estrogen mimicking compounds in systemic circulation, such as parabens, may have the potential to alter normal hormone or estrogen regulated processes such as those observed in ER-positive breast cancer. Similarly, it has been well documented that the paraben parent compounds, BuP and IsoBuP, induce estrogen-dependent breast cancer cell proliferation by ERα signaling and behave as ER agonists as observed in reporter gene and ER binding assays (Okubo et al., 2001; Vo et al., 2011; Watanabe et al., 2013; Wielogorska et al., 2015). Two novel oxidized paraben metabolites were recently shown to be present in human urine samples which structurally appeared to meet the key amino acid hydrogen bonding requirements for ERα in complex with E2 (Delfosse et al., 2014, 2015). We also determined the chemical similarity of the oxidized metabolites, indicated by their Tanimoto coefficient (Tc) (Supplementary Table 1), which were found to be comparable to known agonists, BuP and IsoBuP, relative to E2. Given these structural similarities of the oxidized metabolites and their parent compounds with E2, we characterized the potential estrogenic behavior of these novel paraben metabolites and provide evidence as to whether they can promote cell proliferation by an estrogen signaling mechanism. Using in vitro growth assays, we were able to demonstrate that the novel paraben metabolites 2OH and 3OH display characteristics of estrogenic behavior in ER-positive, estrogen-dependent human breast cancer cell lines. Although the 2OH and 3OH metabolites induce the proliferation of MCF-7 breast cancer cells, they appear to be less potent in T47D cells, particularly 3OH which required concentrations above 10 µM to induce significant cell proliferation. The proliferative ability of ER-negative MDA-MB-231 cells treated with the hydroxylated metabolites was not affected further suggesting that the metabolites induce cell proliferation via an ER signaling mechanism. Despite the paraben metabolite induced growth of MCF-7 cells, 2OH and 3OH were found to exhibit estrogenic behavior that was less potent in comparison to their respective parent compounds. A possible explanation for the observed lower potency of the metabolites may be the result of lower membrane permeability and increased hydrophilic characteristics among the metabolites compared to BuP and IsoBuP indicated by their computationally derived log p values (Supplementary Table 1). The presence of the second hydroxyl group on the 2OH and 3OH metabolites likely contributes to an increase in polarity making it more difficult to diffuse across the cell membrane compared their more lipophilic parent compounds. The correlation between membrane permeability and physiochemical properties of small molecules has been widely discussed elsewhere (Guimaraes et al., 2012; Lipinski et al., 2001). We also investigated whether the cell proliferation of the estrogen-dependent breast cancer cells induced by the 3OH and 2OH paraben metabolites was due to ERα signaling. MCF-7 cell proliferation induced by each of the paraben compounds was blocked in the presence of a pure-antiestrogen suggesting that the metabolites promote cell proliferation via ERα. We also examined the effect of the 3OH and 2OH metabolites on the transcriptional activity of an estrogen-inducible promoter by ERE-luciferase assay. We found that 2OH could significantly promote increased transcriptional activity at the concentrations tested; however, the transcriptional activity induced by 3OH was statistically significant only at the highest dose tested. In addition, we observed significant metabolite-induced expression of the estrogen regulated gene, GREB1, which is a well characterized downstream target of ERα signaling. Increased expression of GREB1 was found to be time-dependent upon exposure to the metabolites as observed with the E2 positive-control. GREB1 expression induced by the compounds was blocked in the presence of a pure anti-estrogen for all tested paraben compounds which is consistent with the results from the growth assays as previously discussed. Lastly, our in silico modeling data suggest that the paraben metabolites display a preference for docking to the ligand binding domain of ERα and demonstrate favorable interactions with key amino acid residues as seen in reported crystal structures of ERα with either E2, propylparaben, or BuP (Delfosse et al., 2012, 2014, 2015). Despite the favorable ligand-binding domain interactions predicted from the docking experiments, the computationally derived partition coefficients and greater hydrophilic characteristics of the metabolites suggest that poor bioavailability might explain why the metabolites were not more potent than the parent compounds. It is important to note that our in silico approach cannot be used to make a distinction of whether the oxidized metabolites are indeed true agonists or antagonists. However, future work will focus on elucidating other potential binding modes or allosteric interactions among other hormone receptors and the oxidized metabolites. Collectively, these data, suggest that the novel 3OH and 2OH metabolites demonstrate behavior consistent with their being weak estrogens. Although the paraben metabolites were found to be generally less potent than their parent compounds, their calculated EC50 values were still within a similar order of magnitude as their parent compounds according to previous reports. Furthermore, the extent of oxidative modification for the oxidized metabolites have been shown to be present ∼2.3 or ∼1.1-fold higher than their parent compounds in human urine for the 2OH and 3OH metabolites, respectively (Moos et al., 2016a). This is especially important due to the presence of parabens in breast tissue (Charles and Darbre, 2013) where the tissue concentrations of the oxidized metabolites is currently unknown. Charles and Darbre (2013) have previously shown that 27% of breast tissue samples taken from patients with ER + PR+ primary tumors contained at least 1 measureable paraben compound that was above its lowest-observed-effect concentration in MCF-7 cells (Charles and Darbre, 2013). Although the breast tissue concentrations analyzed by Charles and Darbre (2013) were found to have median concentrations in the low nanomolar range, some of the tissue samples were observed to have measurable paraben compounds in the micromolar range within 1–3 orders of magnitude of their experimental EC50 values. This same study also reported a few breast tissue concentrations that were at or above our experimentally determined EC50 for IsoBuP or near the EC30 for BuP (Barr et al., 2012; Charles and Darbre, 2013). Despite these findings, healthy control tissue was not examined for paraben content making it difficult to interpret what potential biological effect the presence of these paraben compounds in breast tissue might have on ER-positive breast cancer. Furthermore, it is not well understood what potential effect these estrogen-mimicking compounds might have in the context of ER-positive breast cancer patients who are on antiestrogen therapy, such as AIs, and whether or not total paraben exposure could contribute as possible mechanism of resistance for these patients. Therefore, combined exposure from the metabolites and their parent compounds should not be overlooked when being assessed in biomonitoring studies due to the risk of underestimating human exposure. Future work is needed to establish whether the metabolites would have a combined estrogenic effect in the presence of relevant concentrations of other paraben compounds that have been previously measured in tissue samples from patients with ER+ PR+ primary breast cancer (Barr et al., 2012; Charles and Darbre, 2013). To our knowledge, this is the first report to characterize the estrogenic behavior of the novel paraben metabolites, 3OH and 2OH. We have demonstrated that the oxidized 3OH and 2OH paraben metabolites induce breast cancer cell proliferation by estrogen signaling on the same order of magnitude as their parent compounds. However, the derived EC50 for the metabolites suggest that they are relatively less potent than their parent compounds. Given the lower potency of the metabolites relative to their respective parent compounds, existing regulatory standards and industry trends toward safer alternatives may be adequate in limiting human exposure to paraben compounds. However, future biomonitoring studies should attempt to account for the metabolites when determining total daily intake averages in order to prevent an underestimation of an equally important component of potential endocrine disrupting effects upon paraben exposure. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. ACKNOWLEDGMENTS We would to like to thank Dr. Vladimir Belov, Max Planck Institute for Biophysical Chemistry, Germany, for synthesizing the hydroxylated metabolites used in this study. We acknowledge the assistance provided by the GUMC Tissue Culture Shared Resource, which is supported in part by the Lombardi Comprehensive Cancer Center support grant (NIH/NCI Grant P30-CA051008). FUNDING National Institute of Environmental Health Sciences of the National Institutes of Health under Award Number T32ES007062 (T.L.G.) and the Breast Cancer Research Foundation (BCRF) (N003173 to J.M.R.). REFERENCES Artacho-Cordón F., Arrebola J. P., Nielsen O., Hernández P., Skakkebaek N. E., Fernández M. F., Andersson A. M., Olea N., Frederiksen H. ( 2017). Assumed non-persistent environmental chemicals in human adipose tissue; matrix stability and correlation with levels measured in urine and serum. Environ. Res.  156, 120– 127. Google Scholar CrossRef Search ADS   Azzouz A., Rascon A. J., Ballesteros E. ( 2016). Simultaneous determination of parabens, alkylphenols, phenylphenols, bisphenol A and triclosan in human urine, blood and breast milk by continuous solid-phase extraction and gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal.  119, 16– 26. Google Scholar CrossRef Search ADS   Barr L., Metaxas G., Harbach C. A., Savoy L. A., Darbre P. D. ( 2012). Measurement of paraben concentrations in human breast tissue at serial locations across the breast from axilla to sternum. J. Appl. Toxicol.  32, 219– 232. Google Scholar CrossRef Search ADS   Calafat A. M., Ye X., Wong L. Y., Bishop A. M., Needham L. L. ( 2010). Urinary concentrations of four parabens in the U.S. population: NHANES 2005–2006. Environ. Health Perspect.  118, 679– 685. Google Scholar CrossRef Search ADS   Canutescu A. A., Dunbrack R. L. ( 2003). Cyclic coordinate descent: A robotics algorithm for protein loop closure. Protein Sci.  12, 963– 972. Google Scholar CrossRef Search ADS   Charles A. K., Darbre P. D. ( 2013). Combinations of parabens at concentrations measured in human breast tissue can increase proliferation of MCF-7 human breast cancer cells. J. Appl. Toxicol.  33, 390– 398. Google Scholar CrossRef Search ADS   Chen J., Ahn K. C., Gee N. A., Gee S. J., Hammock B. D., Lasley B. L. ( 2007). Antiandrogenic properties of parabens and other phenolic containing small molecules in personal care products. Toxicol. Appl. Pharmacol.  221, 278– 284. Google Scholar CrossRef Search ADS   Csiszar S. A., Ernstoff A. S., Fantke P., Meyer D. E., Jolliet O. ( 2016). High-throughput exposure modeling to support prioritization of chemicals in personal care products. Chemosphere  163, 490– 498. Google Scholar CrossRef Search ADS   Delfosse V., Grimaldi M., Cavailles V., Balaguer P., Bourguet W. ( 2014). Structural and functional profiling of environmental ligands for estrogen receptors. Environ. Health Perspect.  122, 1306– 1313. Delfosse V., Grimaldi M., Pons J. L., Boulahtouf A., le Maire A., Cavailles V., Labesse G., Bourguet W., Balaguer P. ( 2012). Structural and mechanistic insights into bisphenols action provide guidelines for risk assessment and discovery of bisphenol A substitutes. Proc. Natl Acad. Sci . 109, 14930– 14935. Google Scholar CrossRef Search ADS   Delfosse V., Maire A. l., Balaguer P., Bourguet W. ( 2015). A structural perspective on nuclear receptors as targets of environmental compounds. Acta Pharmacol. Sin  36, 88– 104. Google Scholar CrossRef Search ADS   Dewalque L., Pirard C., Charlier C. ( 2014). Measurement of urinary biomarkers of parabens, benzophenone-3, and phthalates in a Belgian population. Biomed. Res. Int.  2014, 1. Google Scholar CrossRef Search ADS   Dodson R. E., Nishioka M., Standley L. J., Perovich L. J., Brody J. G., Rudel R. A. ( 2012). Endocrine disruptors and asthma-associated chemicals in consumer products. Environ. Health Perspect.  120, 935– 943. Google Scholar CrossRef Search ADS   El-Ashry D., Chrysogelos S. A., Lippman M. E., Kern F. G. ( 1996). Estrogen induction of TGF-alpha is mediated by an estrogen response element composed of two imperfect palindromes. J. Steroid Biochem. Mol. Biol.  59, 261– 269. Google Scholar CrossRef Search ADS   Evans R. M., Scholze M., Kortenkamp A. ( 2012). Additive mixture effects of estrogenic chemicals in human cell-based assays can be influenced by inclusion of chemicals with differing effect profiles. PLoS ONE  7, e43606. Google Scholar CrossRef Search ADS   Frederiksen H., Jorgensen N., Andersson A. M. ( 2011). Parabens in urine, serum and seminal plasma from healthy Danish men determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS). J. Expo. Sci. Environ. Epidemiol.  21, 262– 271. Google Scholar CrossRef Search ADS   Guimaraes C. R., Mathiowetz A. M., Shalaeva M., Goetz G., Liras S. ( 2012). Use of 3D properties to characterize beyond rule-of-5 property space for passive permeation. J. Chem. Inf. Model.  52, 882– 890. Google Scholar CrossRef Search ADS   Guo Y., Kannan K. ( 2013). A survey of phthalates and parabens in personal care products from the United States and its implications for human exposure. Environ. Sci. Technol.  47, 14442– 14449. Google Scholar CrossRef Search ADS   Hines E. P., Mendola P., von Ehrenstein O. S., Ye X., Calafat A. M., Fenton S. E. ( 2015). Concentrations of environmental phenols and parabens in milk, urine and serum of lactating North Carolina women. Reprod. Toxicol.  54, 120– 128. Google Scholar CrossRef Search ADS   Johnson M. D., Zuo H., Lee K. H., Trebley J. P., Rae J. M., Weatherman R. V., Desta Z., Flockhart D. A., Skaar T. C. ( 2004). Pharmacological characterization of 4-hydroxy-N-desmethyl tamoxifen, a novel active metabolite of tamoxifen. Breast Cancer Res. Treat.  85, 151– 159. Google Scholar CrossRef Search ADS   Kang H. S., Kyung M. S., Ko A., Park J. H., Hwang M. S., Kwon J. E., Suh J. H., Lee H. S., Moon G. I., Hong J. H.et al.  , ( 2016). Urinary concentrations of parabens and their association with demographic factors: A population-based cross-sectional study. Environ. Res.  146, 245– 251. Google Scholar CrossRef Search ADS   Krieger E., Dunbrack R. L.Jr, Hooft R. W., Krieger B. ( 2012). Assignment of protonation states in proteins and ligands: Combining pKa prediction with hydrogen bonding network optimization. Methods Mol. Biol.  819, 405– 421. Google Scholar CrossRef Search ADS   Krieger E., Koraimann G., Vriend G. ( 2002). Increasing the precision of comparative models with YASARA NOVA—A self-parameterizing force field. Proteins  47, 393– 402. Google Scholar CrossRef Search ADS   Krieger E., Vriend G. ( 2014). YASARA View—Molecular graphics for all devices—From smartphones to workstations. Bioinformatics  30, 2981– 2982. Google Scholar CrossRef Search ADS   Liao C., Liu F., Kannan K. ( 2013). Occurrence of and dietary exposure to parabens in foodstuffs from the United States. Environ. Sci. Technol.  47, 3918– 3925. Google Scholar CrossRef Search ADS   Lipinski C. A., Lombardo F., Dominy B. W., Feeney P. J. ( 2001). Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev.  46, 3– 26. Google Scholar CrossRef Search ADS   Lippman M. E., Bolan G., Huff K. ( 1976). The effects of estrogens and antiestrogens on hormone-responsive breast cancer in long-term tissue culture. Cancer Res.  36, 4595– 4601. Lippman M. E., Bolan G. ( 1975). Oestrogen-responsive human breast cancer in long term tissue culture. Nature  256, 592– 593. Google Scholar CrossRef Search ADS   Lippman M. E., Krueger K. A., Eckert S., Sashegyi A., Walls E. L., Jamal S., Cauley J. A., Cummings S. R. ( 2001). Indicators of lifetime estrogen exposure: effect on breast cancer incidence and interaction with raloxifene therapy in the multiple outcomes of raloxifene evaluation study participants. J. Clin. Oncol.  19, 3111– 3116. Google Scholar CrossRef Search ADS   Livak K. J., Schmittgen T. D. ( 2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods  25, 402– 408. Google Scholar CrossRef Search ADS   Lu R., Hu X., Zhou J., Sun J., Zhu A. Z., Xu X., Zheng H., Gao X., Wang X., Jin H. ( 2016). COPS5 amplification and overexpression confers tamoxifen-resistance in ERalpha-positive breast cancer by degradation of NCoR. Nat. Commun.  7, 12044. Google Scholar CrossRef Search ADS   Ma D., Chen L., Zhu X., Li F., Liu C., Liu R. ( 2014). Assessment of combined antiandrogenic effects of binary parabens mixtures in a yeast-based reporter assay. Environ. Sci. Pollut. Res. Int . 21, 6482– 6494. Google Scholar CrossRef Search ADS   Meeker J. D., Yang T., Ye X., Calafat A. M., Hauser R. ( 2010). Urinary concentrations of parabens and serum hormone levels, semen quality parameters, and sperm DNA damage. Environ. Health Perspect.  119, 252– 257. Google Scholar CrossRef Search ADS   Mohammed H., D’Santos C., Serandour A. A., Ali H. R., Brown G. D., Atkins A., Rueda O. M., Holmes K. A., Theodorou V., Robinson J. L.et al.  , ( 2013). Endogenous purification reveals GREB1 as a key estrogen receptor regulatory factor. Cell Rep.  3, 342– 349. Google Scholar CrossRef Search ADS   Moos R. K., Angerer J., Dierkes G., Bruning T., Koch H. M. ( 2016a). Metabolism and elimination of methyl, iso- and n-butyl paraben in human urine after single oral dosage. Arch. Toxicol.  90, 2699– 2709. Google Scholar CrossRef Search ADS   Moos R. K., Apel P., Schroter-Kermani C., Kolossa-Gehring M., Bruning T., Koch H. M. ( 2016b). Daily intake and hazard index of parabens based upon 24 h urine samples of the German Environmental Specimen Bank from 1995 to 2012. J. Expo. Sci. Environ. Epidemiol . 27, 591– 600. Google Scholar CrossRef Search ADS   Nishihama Y., Toshima H., Yoshinaga J., Mizumoto Y., Yoneyama M., Nakajima D., Shiraishi H., Tokuoka S. ( 2017). Paraben exposure and semen quality of Japanese male partners of subfertile couples. Environ. Health Prev. Med.  22, 5. Google Scholar CrossRef Search ADS   Nishihama Y., Yoshinaga J., Iida A., Konishi S., Imai H., Yoneyama M., Nakajima D., Shiraishi H. ( 2016). Association between paraben exposure and menstrual cycle in female university students in Japan. Reprod. Toxicol.  63, 107– 113. Google Scholar CrossRef Search ADS   Oishi S. ( 2001). Effects of butylparaben on the male reproductive system in rats. Toxicol. Ind. Health  17, 31– 39. Google Scholar CrossRef Search ADS   Oishi S. ( 2002). Effects of propyl paraben on the male reproductive system. Food Chem. Toxicol.  40, 1807– 1813. Google Scholar CrossRef Search ADS   Okubo T., Yokoyama Y., Kano K., Kano I. ( 2001). ER-dependent estrogenic activity of parabens assessed by proliferation of human breast cancer MCF-7 cells and expression of ERa and PR. Food Chem. Toxicol.  39, 1225– 1232. Google Scholar CrossRef Search ADS   Pastor-Barriuso R., Fernandez M. F., Castano-Vinyals G., Whelan D., Perez-Gomez B., Llorca J., Villanueva C. M., Guevara M., Molina-Molina J. M., Artacho C. F. ( 2016). Total effective xenoestrogen burden in serum samples and risk for breast cancer in a population-based multicase-control study in Spain. Environ. Health Perspect.  124, 1575– 1582. Google Scholar CrossRef Search ADS   Rae J. M., Johnson M. D., Scheys J. O., Cordero K. E., Larios J. M., Lippman M. E. ( 2005). GREB 1 is a critical regulator of hormone dependent breast cancer growth. Breast Cancer Res. Treat.  92, 141– 149. Google Scholar CrossRef Search ADS   Sasseville D., Alfalah M., Lacroix J. P. ( 2015). “Parabenoia” debunked, or “Who's Afraid of Parabens?”. Dermatitis  26, 254– 259. Google Scholar CrossRef Search ADS   Schwartz G. F., Hughes K. S., Lynch H. T., Fabian C. J., Fentiman I. S., Robson M. E., Domchek S. M., Hartmann L. C., Holland R., Winchester D. J. ( 2008). Proceedings of the international consensus conference on breast cancer risk, genetics, & risk management, April, 2007. Cancer  113, 2627– 2637. Google Scholar CrossRef Search ADS   Siegel R. L., Miller K. D., Jemal A. ( 2017). Cancer statistics, 2017. CA Cancer J. Clin.  67, 7– 30. Google Scholar CrossRef Search ADS   Sikora M. J., Cordero K. E., Larios J. M., Johnson M. D., Lippman M. E., Rae J. M. ( 2009). The androgen metabolite 5alpha-androstane-3beta, 17beta-diol (3betaAdiol) induces breast cancer growth via estrogen receptor: implications for aromatase inhibitor resistance. Breast Cancer Res. Treat.  115, 289– 296. Google Scholar CrossRef Search ADS   Sikora M. J., Strumba V., Lippman M. E., Johnson M. D., Rae J. M. ( 2012). Mechanisms of estrogen-independent breast cancer growth driven by low estrogen concentrations are unique versus complete estrogen deprivation. Breast Cancer Res. Treat.  134, 1027– 1039. Google Scholar CrossRef Search ADS   Trott O., Olson A. J. ( 2010). AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem.  31, 455– 461. Valle-Sistac J., Molins-Delgado D., Diaz M., Ibanez L., Barcelo D., Silvia D. iaz., Cruz M. ( 2016). Determination of parabens and benzophenone-type UV filters in human placenta. First description of the existence of benzyl paraben and benzophenone-4. Environ. Int.  88, 243– 249. Google Scholar CrossRef Search ADS   Vo T. T., Jung E. M., Choi K. C., Yu F. H., Jeung E. B. ( 2011). Estrogen receptor alpha is involved in the induction of Calbindin-D(9k) and progesterone receptor by parabens in GH3 cells: A biomarker gene for screening xenoestrogens. Steroids  76, 675– 681. Google Scholar CrossRef Search ADS   Watanabe Y., Kojima H., Takeuchi S., Uramaru N., Sanoh S., Sugihara K., Kitamura S., Ohta S., Kitamura S. ( 2013). Comparative study on transcriptional activity of 17 parabens mediated by estrogen receptor alpha and beta and androgen receptor. Food Chem. Toxicol.  57, 227– 234. Google Scholar CrossRef Search ADS   Watkins D. J., Ferguson K. K., Anzalota Del Toro L. V., Alshawabkeh A. N., Cordero J. F., Meeker J. D. ( 2015). Associations between urinary phenol and paraben concentrations and markers of oxidative stress and inflammation among pregnant women in Puerto Rico. Int. J. Hyg. Environ. Health  218, 212– 219. Google Scholar CrossRef Search ADS   Wielogorska E., Elliott C. T., Danaher M., Connolly L. ( 2015). Endocrine disruptor activity of multiple environmental food chain contaminants. Toxicol. In Vitro  29, 211– 220. Google Scholar CrossRef Search ADS   Windham G. C., Elkin E., Fenster L., Waller K., Anderson M., Mitchell P. R., Lasley B., Swan S. H. ( 2002). Ovarian hormones in premenopausal women: variation by demographic, reproductive and menstrual cycle characteristics. Epidemiology  13, 675– 684. Google Scholar CrossRef Search ADS   Wróbel A., Gregoraszczuk E. Ł. ( 2013). Effects of single and repeated in vitro exposure of three forms of parabens, methyl-, butyl- and propylparabens on the proliferation and estradiol secretion in MCF-7 and MCF-10A cells. Pharmacol. Rep.  65, 484– 493. Google Scholar CrossRef Search ADS   Wrobel A. M., Gregoraszczuk E. L. ( 2014). Actions of methyl-, propyl- and butylparaben on estrogen receptor-alpha and -beta and the progesterone receptor in MCF-7 cancer cells and non-cancerous MCF-10A cells. Toxicol. Lett.  230, 375– 381. Google Scholar CrossRef Search ADS   © 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

Metabolites of n-Butylparaben and iso-Butylparaben Exhibit Estrogenic Properties in MCF-7 and T47D Human Breast Cancer Cell Lines

Loading next page...
 
/lp/ou_press/metabolites-of-n-butylparaben-and-iso-butylparaben-exhibit-estrogenic-GnAZNKRN0V
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/kfy063
Publisher site
See Article on Publisher Site

Abstract

Abstract Two oxidized metabolites of n-butylparaben (BuP) and iso-butylparaben (IsoBuP) discovered in human urine samples exhibit structural similarity to endogenous estrogens. We hypothesized that these metabolites bind to the human estrogen receptor (ER) and promote estrogen signaling. We tested this using models of ER-mediated cellular proliferation. The estrogenic properties of 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) were determined using the ER-positive, estrogen-dependent human breast cancer cell lines MCF-7, and T47D. The 3OH metabolite induced cellular proliferation with EC50 of 8.2 µM in MCF-7 cells. The EC50 for 3OH in T47D cells could not be reached. The 2OH metabolite induced proliferation with EC50 of 2.2 µM and 43.0 µM in MCF-7 and T47D cells, respectively. The EC50 for the parental IsoBuP and BuP was 0.30 and 1.2 µM in MCF-7 cells, respectively. The expression of a pro-proliferative, estrogen-inducible gene (GREB1) was induced by these compounds and blocked by co-administration of an ER antagonist (ICI 182, 780), confirming the ER-dependence of these effects. The metabolites promoted significant ER-dependent transcriptional activity of an ERE-luciferase reporter construct at 10 and 20 µM for 2OH and 10 µM for 3OH. Computational docking studies showed that the paraben compounds exhibited the potential for favorable ligand-binding domain interactions with human ERα in a manner similar to known x-ray crystal structures of 17ß-estradiol in complex with ERα. We conclude that the hydroxylated metabolites of BuP and IsoBuP are weak estrogens and should be considered as additional components of potential endocrine disrupting effects upon paraben exposure. breast cancer, endocrine, estrogens, paraben, environmental Breast cancer is currently the second most common malignancy diagnosed in the United States (US) with an estimated 250 000 new cases and 40 000 disease-related deaths in 2016 (Siegel et al., 2017). An estimated 7% of breast cancer diagnoses are attributed to an inherited genetic predisposition with the remaining ∼93% being attributed to other likely risk factors such as lifestyle choices, obesity, or exposure to environmental carcinogens (Schwartz et al., 2008). In addition, lifetime exposure to estrogens has been shown to be correlated with breast cancer risk (Lippman, 2001). Both preclinical and clinical studies have shown that estrogens, specifically 17ß-estradiol (E2), can induce breast cancer pathogenesis (Lippman and Bolan, 1975; Lippman et al., 1976). The ability of estrogen to induce breast cancer cellular proliferation is concerning because it is estimated that 70%–80% of all diagnosed breast cancers express the estrogen receptor (ER) and are deemed ER-positive (Lu et al., 2016). Pharmacological approaches have been developed to block ER signaling using ER antagonists such as tamoxifen, and more recently aromatase inhibitors (AIs), which inhibit the synthesis of estrogens (Sikora et al., 2012), have been used. While highly effective, de novo and acquired resistance to AIs is common, and a possible mechanism of resistance is attributed to exposure to environmental estrogen mimicking chemicals, or xenoestrogens (Pastor-Barriuso et al., 2016; Sikora et al., 2009) including the ones we chose to study in this report (Moos et al., 2016a). Recently, there has been increasing public awareness and research into environmental chemicals that may have biologically relevant, endocrine disrupting properties (Delfosse et al., 2014). Xenoestrogens include many of these suspected endocrine disrupting compounds that have previously been shown to display agonistic behavior toward estrogen receptor-α (ERα) (Delfosse et al., 2015; Wielogorska et al., 2015). Alkyl esters of p-hydroxybenzoic acid (parabens) are one type of xenoestrogen that have been investigated for whether current human exposure levels are a cause for concern (Sasseville et al., 2015). Given their antimicrobial properties, parabens are frequently used as a preservative in numerous pharmaceuticals, food products, and personal care products (Guo and Kannan, 2013; Liao et al., 2013). Compared to oral ingestion, dermal contact encompasses a broader range of paraben exposure sources that are found in numerous cosmetics and personal care products (Dodson et al., 2012). As an example, daily paraben exposure for a 63-kg average adult in Korea is estimated to be a sum total of 18 960 µg/day for methyl and ethylparaben combined, and 1580 µg/day for propyl paraben alone (Kang et al., 2016). Similarly, dermal intake of parabens has been estimated to be 31 µg/kg-bw/day for adult females and range between 58.6 and 766 µg/kg-bw/day among infants and toddlers within US in 2012 (Guo and Kannan, 2013). Exposure estimates for parabens in Belgium, Germany, and several other nations have been summarized and reported elsewhere (Calafat et al., 2010; Csiszar et al., 2016; Dewalque et al., 2014; Moos et al., 2016b). The ubiquitous exposure to parabens has led to reports of their detection in breast tissue, breast milk, placental tissue, serum, seminal fluid, and urine samples from numerous general populations around the world (Azzouz et al., 2016; Barr et al., 2012; Frederiksen et al., 2011; Hines et al., 2015; Moos et al., 2016b; Valle-Sistac et al., 2016). In spite of evidence showing the presence of parabens in various human tissues such as adipose (Artacho-Cordón et al., 2017) and breast tissue (Charles and Darbre, 2013), there is still significant debate over their current risk to the general population. To date, studies have focused on the estrogenic properties of paraben parent compounds (Okubo et al., 2001; Wielogorska et al., 2015; Wróbel and Gregoraszczuk, 2013, 2014). Two oxidized paraben metabolites, 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) (Figure 1), have recently been discovered, but their potential estrogenic properties have not yet been studied (Moos et al., 2016a). Unlike E2, which forms 3 hydrogen bonds in the ligand binding pocket of human ER according to X-ray crystallography structures (Delfosse et al., 2012), BuP is unable to hydrogen bond with His524 (Delfosse et al., 2014). The lack of a second hydroxyl group on BuP likely contributes to its relatively weak ER binding affinity. We hypothesized that the metabolites’ additional hydroxyl group might enable binding interactions similar to that of 17ß-estradiol and confer increased estrogenic potency compared to their respective parent compounds. To test this, we used pre-clinical in vitro models of ER-positive breast cancer and demonstrated that the oxidized metabolites promote cell proliferation in an ER-dependent manner comparable to their respective parent compounds. Computational docking studies indicated that the metabolites display hydrogen bonding capabilities similar to 17ß-estradiol within the human ERα ligand-binding domain in support of our in vitro data. Figure 1. View largeDownload slide Chemical structures of 17ß-estradiol, 3OH, and 2OH paraben metabolites. * indicates chiral center. Figure 1. View largeDownload slide Chemical structures of 17ß-estradiol, 3OH, and 2OH paraben metabolites. * indicates chiral center. MATERIALS AND METHODS Reagents 17ß-estradiol (E2), ICI 182, 780 (ICI), and n-butylparaben (BuP) were purchased from Sigma-Aldrich Inc. (St. Louis, MO) with purity ≥98% determined by HPLC. E2 and ICI were dissolved to 10 mM in absolute ethanol and stored at −20°C. Iso-butylparaben (IsoBuP) was purchased from TCI America (Portland, OR) with purity ≥98% determined by HPLC. 3-hydroxy n-butyl 4-hydroxybenzoate (3OH) and 2-hydroxy iso-butyl 4-hydroxybenzoate (2OH) were kindly provided as a gift by Dr. Vladimir Belov, Max Planck Institute for Biophysical Chemistry in Germany. The 3OH and 2OH compounds had a purity ≥95% determined by HPLC with UV (254 nm) detection. BuP, IsoBuP, 3OH, and 2OH were dissolved to 100 mM in DMSO stored at −20°C. The final concentration of ethanol or DMSO did not exceed 0.1% (v/v) in culture media. All compounds were stored protected from light. Cell lines, culture conditions, and proliferation assays MCF-7 and T47D cells were obtained from the Tissue Culture Shared Resource (TCSR) at the Lombardi Comprehensive Cancer Center (Georgetown University, Washington, DC). Cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco/Life Technologies, Grand Island, NY) supplemented with 10% (v/v) fetal calf serum (Valley Biomedical Inc., Winchester, VA), at 37°C in a humidified 5% (v/v) CO2 atmosphere. The identity of the cells was confirmed by short tandem repeat profiling by the TCSR, and shown to be free of mycoplasma contamination. For assays in defined steroid deplete conditions, cells were repeatedly washed and grown in steroid depleted media over 3 days before proliferation assays (phenol red-free IMEM supplemented with 10% [v/v] charcoal stripped bovine calf serum—CCS) (Valley Biomedical Inc., Winchester, VA) based on a previously described method (Rae et al., 2005). For growth assays, MCF-7 and T47D cells were withdrawn from steroids as previously described (Johnson et al., 2004) and plated in steroid-free media at 1000 and 2000 cells/well, respectively, in 96-well plates and allowed to attach overnight. PrestoBlue cell viability assay MCF-7 and T47D cells were plated in steroid-free media at 1000 and 2000 cells/well, respectively, in 96-well plates and cultured overnight. The following day, cells were treated with the specified compounds at the indicated concentrations and the vehicle controls (ethanol or DMSO) diluted in IMEM supplemented with 10% (v/v) CCS. 6 days after treatment, cell viability was assessed using Presto Blue reagent (ThermoFisher Scientific, Waltham, MA) according to manufacturer’s instructions. In brief, cells were incubated for 3 h in the presence of 10% (v/v) Presto Blue and fluorescence was measured using a POLARstar Omega plate reader with the excitation/emission wavelengths set at 544/590 nm. RNA expression assay MCF-7 cells were repeatedly washed and grown in steroid depleted media over 3 days as described above and plated in steroid depleted media at 400 000 cells/well in 6-well plates at least 10 h before treatment with parabens for 2, 4, and 6 h durations. Total RNA was isolated using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) according to the manufacturer’s instructions. Yield and quantity were determined by spectrophotometry (NanoDrop ND-1000). All samples were stored at −80°C. Total RNA (1 µg) was reverse transcribed (RT) using Reverse Transcription System (Promega, Madison, WI) and the cDNA amplified in a 25 µl reaction containing Gene Expression Master Mix and gene specific primers both from Thermo Fisher Scientific (Waltham, MA). GREB1 mRNA expression was measured using a TaqMan RT-PCR assay as described previously (Rae et al., 2005). GREB1 expression was normalized against GAPDH with relative expression being calculated using the ΔΔCT method (Livak and Schmittgen, 2001). Estrogen-response element (ERE)–luciferase reporter assay The ERE–luciferase reporter plasmid (a gift from Dr. Anna T. Riegel, Georgetown University) was co-transfected into MCF-7 cells with Fugene 6 transfection reagent (Promega, Madison, WI) according to manufacturer’s instructions with the renilla plasmid (100:1 v:v luciferase to renilla). After 18 h, transfected cells were repeatedly washed with phenol red-free 10% (v/v) CCS I-MEM (Valley Biomedical Inc., Winchester, VA) to produce a steroid-free environment. The following day cells were plated on a 24-well plate (Corning/Costar) in 10% (v/v) CCS I-MEM (Valley Biomedical Inc., Winchester, VA). After 6 h, cells were treated with either 2OH, 3OH, BuP, IsoBuP, E2, or a vehicle control (ethanol or DMSO). After 18 h, cell lysates were processed and relative light unit reading were measured according to manufacturer’s recommendations on a Leader 50 luminometer (Gen-Probe). Molecular docking The crystal structure of the human ERα ligand-binding domain in complex with 17ß-estradiol was downloaded from the protein data bank PDB ID: 3UUD and prepared for docking with the YASARA molecular modeling suite (Krieger and Vriend, 2014). The PDB entry 3UUD was selected for docking due to its high resolution (1.6 Å). The presence of an endogenous ligand was another factor considered when selecting this structure to ensure that the ligand-binding domain was a suitable representation of the agonist conformation. All crystallographic waters were removed, hydrogen atoms were added, bond orders were corrected for S-hydroxycysteine residues, missing loops were repaired (Canutescu and Dunbrack, 2003), and only the A-chain was retained. Docking was performed with VINA (Trott and Olson, 2010) using default parameters, and setup was conducted with YASARA as the graphical front-end (Krieger et al., 2002). A grid size of 25 Å  × 25 Å  × 25 Å was centered on the C9 carbon of E2 before removing the native ligand and set as the search space for the paraben compounds treated as flexible ligands. The best pose of 25 runs was defined as producing the smallest root-mean-square deviation (RMSD) between the benzoyl group of the paraben compounds and comparable carbon atoms (1–5, 9, & 10) and phenol oxygen atom of the phenolic-A ring of E2. Hydrogen bond networks of the docked structures were optimized with YASARA before determining RMSD values (Krieger et al., 2012). Statistical analysis The curve-fitting program GraphPad Prism 7.0 was used to plot all data. The time and relative fold expression of GREB1 within each treatment group was processed by one-way ANOVA (p < .005) followed by a Bonferroni post hoc analysis to determine which time points differed from each other. RESULTS 3OH and 2OH Butyl Paraben Metabolites Induce the Proliferation of Estrogen-Dependent Human Breast Cancer Cells Grown in Steroid-Free Conditions To determine the whether the paraben metabolites could induce human breast cancer growth, estrogen-dependent MCF-7 and T47D cell lines were treated with BuP, IsoBuP, and their respective metabolites. The calculated effective concentrations inducing 50% growth (EC50) for BuP and IsoBuP (1.2 and 0.30 µM, respectively) were similar to those reported previously (Charles and Darbre, 2013; Delfosse et al., 2014; Evans et al., 2012; Watanabe et al., 2013) (Supplementary Figs. 1A–D). MCF-7 cells treated with the 3OH (Figure 2A) and 2OH (Figure 2B) paraben metabolites induced cellular proliferation in a concentration-dependent manner. T47D cells treated with 3OH (Figure 2C) or 2OH (Figure 2D) required slightly higher concentrations than MCF-7 cells to induce proliferation but they did not achieve a sigmoidal concentration dependence. Higher concentrations of the 3OH metabolite would exceed 0.1% DMSO vehicle concentration which prevented an EC50 from being determined in the T47D cell line. The EC50 for 3OH and 2OH in MCF-7 and T47D cells was calculated (3OH: EC50 8.0 µM in MCF-7; 2OH: EC50 2.2 µM in MCF-7 and 43.0 µM in T47D). All PrestoBlue cell viability assay results were confirmed with the crystal violet growth assay as described in Materials and Methods section (data not shown). Previous reports have investigated the androgenic and anti-androgenic behavior of parabens (Chen et al., 2007; Ma et al., 2014) prior to the discovery of the 3OH and 2OH paraben metabolites. Therefore, we also examined whether the 3OH and 2OH paraben metabolites exhibited androgenic properties by treating androgen-dependent LNCaP cells and we found that the paraben metabolites do not exhibit androgenic behavior at concentration ranges of 10 pM–10 µM (Supplementary Figs. 2A and 2B). Non-ER α expressing MDA-MB-231 cells were treated with 3OH, 2OH, and their respective parent compounds to show no effect on cell proliferation. (Supplementary Figs. 3A–D). Figure 2. View largeDownload slide 3OH and 2OH paraben metabolites induce breast cancer cell proliferation. MCF-7 and T47D cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. MCF-7 cells were treated with either 3OH (A) or 2OH (B) and T47D cells were treated with 3OH (C) or 2OH (D) at the indicated concentrations. Growth curves for A–D represent percentage of cell growth compared to DMSO (vehicle) control (0%). Points on dose response curve represent 6-day growth versus vehicle treated control ± SE (n = 6 technical replicates). Dotted line indicates growth induced by 100 pM E2. Figure 2. View largeDownload slide 3OH and 2OH paraben metabolites induce breast cancer cell proliferation. MCF-7 and T47D cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. MCF-7 cells were treated with either 3OH (A) or 2OH (B) and T47D cells were treated with 3OH (C) or 2OH (D) at the indicated concentrations. Growth curves for A–D represent percentage of cell growth compared to DMSO (vehicle) control (0%). Points on dose response curve represent 6-day growth versus vehicle treated control ± SE (n = 6 technical replicates). Dotted line indicates growth induced by 100 pM E2. The Pure Anti-Estrogen Receptor Antagonist (ICI 182, 780) Inhibits 3OH and 2OH Metabolite Induced Proliferation of Estrogen-Dependent Human Breast Cancer To confirm that the cell proliferation induced by 3OH and 2OH paraben metabolites was due to ERα signaling, we assessed whether their proliferative effects could be blocked by the pure anti-estrogen ICI 182, 780 (Fulvestrant). Estrogen-dependent MCF-7 cells were treated at the identified EC50 concentrations of BuP (1.2 µM), IsoBuP (0.3 µM), and their metabolites either alone or in the presence of increasing concentrations of ICI. BuP and IsoBuP induced proliferation was inhibited by ICI at 0.8 and 1.4 nM, respectively, comparable to previous reports (Okubo et al., 2001) (Figs. 3A and B). Cell proliferation induced by the 3OH and 2OH paraben metabolites was inhibited in a concentration-dependent manner by ICI with an IC50 of 0.7 and 1.2 nM, respectively (Figs. 3A and B). Figure 3. View largeDownload slide Pure anti-estrogen blocks 3OH, 2OH, respective parent compound induced breast cancer cell proliferation. MCF-7 cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. Growth induction by a fixed dose of either (A) (▪) 0.3 µM IsoBuP (solid line) or (□) 2.15 µM 2OH (dashed line) or (B) by (▪) 1.2 µM Butylparaben or (□) 8.22 µM 3OH was antagonized by the pure anti-estrogen, ICI 182, 780 (ICI). ICI was added to final concentrations ranging from 1 pM to 1 µM at log intervals. Growth curves represents percentage of cell growth compared to 2OH or IsoBuP at the fixed concentrations indicated above. Data are normalized from maximum growth (100%) to minimum growth (0%) for each treatment. Points on dose response curve represent Points represent 6-day growth versus proliferation with EC50 of indicated paraben without ICI ± SE (n = 6 technical replicates). Figure 3. View largeDownload slide Pure anti-estrogen blocks 3OH, 2OH, respective parent compound induced breast cancer cell proliferation. MCF-7 cells were grown in E2-free conditions as described in Materials and Methods section. PrestoBlue cell viability assay was used as a surrogate to determine relative cell number. Growth induction by a fixed dose of either (A) (▪) 0.3 µM IsoBuP (solid line) or (□) 2.15 µM 2OH (dashed line) or (B) by (▪) 1.2 µM Butylparaben or (□) 8.22 µM 3OH was antagonized by the pure anti-estrogen, ICI 182, 780 (ICI). ICI was added to final concentrations ranging from 1 pM to 1 µM at log intervals. Growth curves represents percentage of cell growth compared to 2OH or IsoBuP at the fixed concentrations indicated above. Data are normalized from maximum growth (100%) to minimum growth (0%) for each treatment. Points on dose response curve represent Points represent 6-day growth versus proliferation with EC50 of indicated paraben without ICI ± SE (n = 6 technical replicates). 3OH and 2OH Metabolites Induce Expression of the Estrogen-Regulated Gene, GREB1 To confirm the estrogen agonist action of the paraben metabolites we tested the ability of these compounds to induce the expression of an estrogen responsive gene, GREB1. We measured the effects of paraben-treated ER-positive MCF-7 cell lines on GREB1 mRNA levels. GREB1 is a critical downstream target of ERα signaling the expression of which is induced by exposure to E2, which can be suppressed by ICI 182, 780 as previously described (Rae et al., 2005). GREB1 has also been described as a critical estrogen-specific ER-interacting protein, an interaction that has been shown to be highly enriched upon estrogen exposure (Mohammed et al., 2013). ER-positive MCF-7 cells were treated at 10 µM with BuP, IsoBuP, 3OH, and 2OH as described in Materials and Methods section. E2 was used as a positive control; it induced GREB1 expression (∼29-fold at 6 h; p < .001) at all time-points compared with the vehicle control (Figure 4). Treatment with BuP, IsoBuP, 2OH, and 3OH induced GREB1 expression ∼30, ∼36, ∼20, and ∼10-fold at 6 h, respectively (p < .001) compared to vehicle control. Neither ethanol nor DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Co-treatment with the paraben compounds and the pure anti-estrogen ICI 182, 780 at 6 h blocked these effects (Figure 5). Figure 4. View largeDownload slide Time course induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), or 100 pM E2, for 2, 4, and 6 h. Neither ethanol or DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Cells were also treated at 10 µM with butylparaben (BuP), IsoBuP (IsoBuP), 3OH, and 2OH. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance within each treatment group was determined by one-way ANOVA followed by Bonferroni post-hoc analysis. # = not significant; a, b, c, d, e = p ≤ .005. Figure 4. View largeDownload slide Time course induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), or 100 pM E2, for 2, 4, and 6 h. Neither ethanol or DMSO significantly affected expression levels compared to media (CCS) alone (data not shown). Cells were also treated at 10 µM with butylparaben (BuP), IsoBuP (IsoBuP), 3OH, and 2OH. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance within each treatment group was determined by one-way ANOVA followed by Bonferroni post-hoc analysis. # = not significant; a, b, c, d, e = p ≤ .005. Figure 5. View largeDownload slide Induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites is blocked in the presence of ICI 182, 780. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), 100 pM E2, or 100 nM ICI for 6 h. Cells were also treated at 10 µM with BuP, IsoBuP (IsoBuP), 3OH, and 2OH alone or in combination with 100 nM ICI. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance between each lettered pair was determined by t-test and significant at p < .001. Figure 5. View largeDownload slide Induction of GREB1 expression in MCF-7 cells by the 3OH and 2OH paraben metabolites is blocked in the presence of ICI 182, 780. MCF-7 cells were assayed in E2-free conditions. Cells were treated with a vehicle control (0.001% ethanol or 0.1% DMSO), 100 pM E2, or 100 nM ICI for 6 h. Cells were also treated at 10 µM with BuP, IsoBuP (IsoBuP), 3OH, and 2OH alone or in combination with 100 nM ICI. Bars represent GREB1 expression versus vehicle-treated control using the ΔΔCT method. Bars represent the mean from 3 technical replicates ± SE. Statistical significance between each lettered pair was determined by t-test and significant at p < .001. 3OH and 2OH Metabolites Induce Luciferase Expression in MCF-7 Cells Transfected With an Estrogen-Responsive Luciferase Reporter Construct To further confirm that the estrogenic activity of the paraben metabolites is mediated by classical ER mediated signaling, we tested the ability of the compounds to induce ER-dependent transcription using an ERE-luciferase reporter construct (El-Ashry et al., 1996). The 2OH metabolite significantly (p < .05) induced transcriptional activity of the ERE-luciferase reporter construct at 10 and 20 µM (Figure 6). Induced transcriptional activity by the 3OH metabolite was significant at 20 µM (p < .05) but not at 10 µM. BuP and IsoBuP also demonstrated increased ERE transcriptional activity consistent with previous reports (Vo et al., 2011; Wielogorska et al., 2015) (data not shown). Figure 6. View largeDownload slide 3OH and 2OH paraben metabolites promote significant ERE-luciferase transcriptional activity. The 2OH and 3OH paraben metabolites significantly induce ERE-luciferase activity. MCF-7 cells were co-transfected with the ERE-luciferase construct and renilla reporter plasmid in estrogen-free conditions and treated at the indicated concentrations for each compound. Relative firefly luciferase activity was plotted over renilla luciferase activity induced by treatment from each specified compound versus ethanol vehicle control for E2 and DMSO vehicle control for each paraben compound. Statistical significance between each indicated pair was determined by t-test. Bars represent the mean from 3 technical replicates ± SE. ns = not significant; *p ≤ .05; **p ≤ .001. Figure 6. View largeDownload slide 3OH and 2OH paraben metabolites promote significant ERE-luciferase transcriptional activity. The 2OH and 3OH paraben metabolites significantly induce ERE-luciferase activity. MCF-7 cells were co-transfected with the ERE-luciferase construct and renilla reporter plasmid in estrogen-free conditions and treated at the indicated concentrations for each compound. Relative firefly luciferase activity was plotted over renilla luciferase activity induced by treatment from each specified compound versus ethanol vehicle control for E2 and DMSO vehicle control for each paraben compound. Statistical significance between each indicated pair was determined by t-test. Bars represent the mean from 3 technical replicates ± SE. ns = not significant; *p ≤ .05; **p ≤ .001. 3OH and 2OH Metabolites Can Be Docked Within the Ligand-Binding Pocket of ERα To provide additional support that the hydroxylated metabolites were promoting ERE-mediated transcription and stimulated cellular proliferation, molecular docking was used to characterize the interactions of the metabolites as flexible ligands within the known active conformation of the ligand-binding site of human ERα. The E2 ligand in the crystal structure of ERα (PDB ID: 3UUD) was removed and re-docked into the active site. The RMSD among all carbon and oxygen atoms for the binding pose of the docked E2 ligand and the pose identified in the crystal structure of ERα was 0.819 Å, indicating that our prepared ERα structure could accurately reproduce experimentally determined binding poses. The high resolution of the 3UUD ERα structure (1.6 Å), compared to PDB entries 4TV1 (1.85 Å) and 4MG9 (2.0 Å) ERα structures complexed with either propyl- or BuP, respectively, was a significant contributing factor in choosing this structure for docking the hydroxylated paraben metabolites. Docking of either propyl- and or BuP to 4TV1 and 4MG9 resulted in relatively poorer RMSD values compared to 3UUD which was another consideration for proceeding with the 3UUD structure for this study. The binding poses of the hydroxylated paraben metabolites and their respective parent compound were analyzed, and each of them displayed a preference to dock to the reported active site of ERα (Delfosse et al., 2014). Hydrogen bonding interactions of the paraben compounds with key side-chain residues in the active site of ERα are shown in Figures 7A–C and Supplementary Figures 4A–C. The 2OH compound displayed hydrogen bonding capabilities with amino acid residues Arg394 and Glu353 (Figure 7A) which have been shown to hydrogen bond to the phenol ring of E2 (Delfosse et al., 2012, 2015). 2D interaction diagrams generated with Discovery Studio Visualizer indicated the potential for 2 additional hydrogen bonds to be formed with L387 and G521 (Supplementary Figure 4A). The calculated RMSD for E2 and the 2OH metabolite was determined as described in the Materials and Methods section and indicated a high degree of binding similarity with E2 (0.658 Å). Due to the presence of a chiral center located at the secondary alcohol group of the 3OH compound and unknown proportions of a potential racemic mixture of our 3OH test substance, it was necessary to dock both the R and S isomer of the 3OH compound. The identified binding poses of the R & S isomers both suggested favorable hydrogen bonding interactions with amino acids Arg394 and Glu353, but the S isomer was additionally capable of forming hydrogen bonds with His524 and Gly521 (Figs. 7B and C). 2D interaction diagrams showed that the R & S isomer may both form additional hydrogen bonds with L387 but only the S isomer may form an additional hydrogen bond with L525 (Supplementary Figs. 4B and 4C). The calculated RMSD for E2 compared with the 3OH R & S isomers both indicated a high degree of binding similarity with E2 (0.905 and 0.562 Å, respectively) and determined as described in the Materials and Methods section. The calculated RMSD for E2 between each of the parent compounds BuP and IsoBuP demonstrated comparable binding similarity as the metabolites (1.001 and 0.976 Å, respectively). The calculated RMSD for each of the metabolite and parent compound pairs also indicated; high binding similarity with each other (2OH & IsoBuP: 0.123 Å; 3OH R & BuP: 0.293 Å; 3OH S & BuP: 0.555 Å). Figure 7. View largeDownload slide 3OH and 2OH paraben metabolites dock to ERα. Human ERα ligand-binding domain docked with (A) 2OH isomer (B) 3OH R isomer or (C) 3OH S isomer. 2OH and 3OH are colored orange and magenta, respectively. 17ß-estradiol has been overlaid with each ligand in grey for comparison. Hydrogen bonds are represented as yellow dashes. Oxygen and nitrogen atoms are colored in red and blue, respectively. The key structural features and hydrogen bonding residues are displayed and labeled. Figure 7. View largeDownload slide 3OH and 2OH paraben metabolites dock to ERα. Human ERα ligand-binding domain docked with (A) 2OH isomer (B) 3OH R isomer or (C) 3OH S isomer. 2OH and 3OH are colored orange and magenta, respectively. 17ß-estradiol has been overlaid with each ligand in grey for comparison. Hydrogen bonds are represented as yellow dashes. Oxygen and nitrogen atoms are colored in red and blue, respectively. The key structural features and hydrogen bonding residues are displayed and labeled. DISCUSSION In addition to the numerous urine biomonitoring studies of parabens in diverse populations around the world (Artacho-Cordón et al., 2017; Calafat et al., 2010; Kang et al., 2016; Moos et al., 2016b), in vivo studies using male rats suggest potential disruption of testosterone and sperm production from propyl- and BuP exposures (Oishi, 2001, 2002). In contrast, human epidemiological studies have not conclusively demonstrated that paraben concentrations in urine are associated with altered male steroid hormone levels or with indicators of sperm DNA damage (Meeker, 2010; Nishihama et al., 2017). However, a common finding among studies measuring paraben compounds in urine is the several fold higher paraben exposure among females compared to males (Guo and Kannan, 2013; Meeker, 2010; Moos et al., 2016b). Urine studies have reported median concentrations of paraben compounds in urine of females generally at low nanomolar concentrations including 1.0–1.9 µM for methylparaben. 41.0–250.0 nM for propylparaben, and 3.1–3.6 nM for butylparaben (Nishihama et al., 2016; Watkins et al., 2015). One report found significantly higher urine BuP levels (as a marker of higher exposure) being associated with shorter menstrual cycle length among female Japanese women age 19–22 despite only reporting a median concentration of 3.6 nM for this compound (Nishihama et al., 2016) which is 2–3 orders of magnitude lower than the EC50 for butylparaben as determined in our study. A larger study showed an association between higher levels of estrogen metabolites in urine, such as estrone sulfate and estrone glucuronide, and shorter menstrual cycles in a group of women that were part of the Women’s Reproductive Health Study (Windham et al., 2002). Therefore, the presence of estrogen mimicking compounds in systemic circulation, such as parabens, may have the potential to alter normal hormone or estrogen regulated processes such as those observed in ER-positive breast cancer. Similarly, it has been well documented that the paraben parent compounds, BuP and IsoBuP, induce estrogen-dependent breast cancer cell proliferation by ERα signaling and behave as ER agonists as observed in reporter gene and ER binding assays (Okubo et al., 2001; Vo et al., 2011; Watanabe et al., 2013; Wielogorska et al., 2015). Two novel oxidized paraben metabolites were recently shown to be present in human urine samples which structurally appeared to meet the key amino acid hydrogen bonding requirements for ERα in complex with E2 (Delfosse et al., 2014, 2015). We also determined the chemical similarity of the oxidized metabolites, indicated by their Tanimoto coefficient (Tc) (Supplementary Table 1), which were found to be comparable to known agonists, BuP and IsoBuP, relative to E2. Given these structural similarities of the oxidized metabolites and their parent compounds with E2, we characterized the potential estrogenic behavior of these novel paraben metabolites and provide evidence as to whether they can promote cell proliferation by an estrogen signaling mechanism. Using in vitro growth assays, we were able to demonstrate that the novel paraben metabolites 2OH and 3OH display characteristics of estrogenic behavior in ER-positive, estrogen-dependent human breast cancer cell lines. Although the 2OH and 3OH metabolites induce the proliferation of MCF-7 breast cancer cells, they appear to be less potent in T47D cells, particularly 3OH which required concentrations above 10 µM to induce significant cell proliferation. The proliferative ability of ER-negative MDA-MB-231 cells treated with the hydroxylated metabolites was not affected further suggesting that the metabolites induce cell proliferation via an ER signaling mechanism. Despite the paraben metabolite induced growth of MCF-7 cells, 2OH and 3OH were found to exhibit estrogenic behavior that was less potent in comparison to their respective parent compounds. A possible explanation for the observed lower potency of the metabolites may be the result of lower membrane permeability and increased hydrophilic characteristics among the metabolites compared to BuP and IsoBuP indicated by their computationally derived log p values (Supplementary Table 1). The presence of the second hydroxyl group on the 2OH and 3OH metabolites likely contributes to an increase in polarity making it more difficult to diffuse across the cell membrane compared their more lipophilic parent compounds. The correlation between membrane permeability and physiochemical properties of small molecules has been widely discussed elsewhere (Guimaraes et al., 2012; Lipinski et al., 2001). We also investigated whether the cell proliferation of the estrogen-dependent breast cancer cells induced by the 3OH and 2OH paraben metabolites was due to ERα signaling. MCF-7 cell proliferation induced by each of the paraben compounds was blocked in the presence of a pure-antiestrogen suggesting that the metabolites promote cell proliferation via ERα. We also examined the effect of the 3OH and 2OH metabolites on the transcriptional activity of an estrogen-inducible promoter by ERE-luciferase assay. We found that 2OH could significantly promote increased transcriptional activity at the concentrations tested; however, the transcriptional activity induced by 3OH was statistically significant only at the highest dose tested. In addition, we observed significant metabolite-induced expression of the estrogen regulated gene, GREB1, which is a well characterized downstream target of ERα signaling. Increased expression of GREB1 was found to be time-dependent upon exposure to the metabolites as observed with the E2 positive-control. GREB1 expression induced by the compounds was blocked in the presence of a pure anti-estrogen for all tested paraben compounds which is consistent with the results from the growth assays as previously discussed. Lastly, our in silico modeling data suggest that the paraben metabolites display a preference for docking to the ligand binding domain of ERα and demonstrate favorable interactions with key amino acid residues as seen in reported crystal structures of ERα with either E2, propylparaben, or BuP (Delfosse et al., 2012, 2014, 2015). Despite the favorable ligand-binding domain interactions predicted from the docking experiments, the computationally derived partition coefficients and greater hydrophilic characteristics of the metabolites suggest that poor bioavailability might explain why the metabolites were not more potent than the parent compounds. It is important to note that our in silico approach cannot be used to make a distinction of whether the oxidized metabolites are indeed true agonists or antagonists. However, future work will focus on elucidating other potential binding modes or allosteric interactions among other hormone receptors and the oxidized metabolites. Collectively, these data, suggest that the novel 3OH and 2OH metabolites demonstrate behavior consistent with their being weak estrogens. Although the paraben metabolites were found to be generally less potent than their parent compounds, their calculated EC50 values were still within a similar order of magnitude as their parent compounds according to previous reports. Furthermore, the extent of oxidative modification for the oxidized metabolites have been shown to be present ∼2.3 or ∼1.1-fold higher than their parent compounds in human urine for the 2OH and 3OH metabolites, respectively (Moos et al., 2016a). This is especially important due to the presence of parabens in breast tissue (Charles and Darbre, 2013) where the tissue concentrations of the oxidized metabolites is currently unknown. Charles and Darbre (2013) have previously shown that 27% of breast tissue samples taken from patients with ER + PR+ primary tumors contained at least 1 measureable paraben compound that was above its lowest-observed-effect concentration in MCF-7 cells (Charles and Darbre, 2013). Although the breast tissue concentrations analyzed by Charles and Darbre (2013) were found to have median concentrations in the low nanomolar range, some of the tissue samples were observed to have measurable paraben compounds in the micromolar range within 1–3 orders of magnitude of their experimental EC50 values. This same study also reported a few breast tissue concentrations that were at or above our experimentally determined EC50 for IsoBuP or near the EC30 for BuP (Barr et al., 2012; Charles and Darbre, 2013). Despite these findings, healthy control tissue was not examined for paraben content making it difficult to interpret what potential biological effect the presence of these paraben compounds in breast tissue might have on ER-positive breast cancer. Furthermore, it is not well understood what potential effect these estrogen-mimicking compounds might have in the context of ER-positive breast cancer patients who are on antiestrogen therapy, such as AIs, and whether or not total paraben exposure could contribute as possible mechanism of resistance for these patients. Therefore, combined exposure from the metabolites and their parent compounds should not be overlooked when being assessed in biomonitoring studies due to the risk of underestimating human exposure. Future work is needed to establish whether the metabolites would have a combined estrogenic effect in the presence of relevant concentrations of other paraben compounds that have been previously measured in tissue samples from patients with ER+ PR+ primary breast cancer (Barr et al., 2012; Charles and Darbre, 2013). To our knowledge, this is the first report to characterize the estrogenic behavior of the novel paraben metabolites, 3OH and 2OH. We have demonstrated that the oxidized 3OH and 2OH paraben metabolites induce breast cancer cell proliferation by estrogen signaling on the same order of magnitude as their parent compounds. However, the derived EC50 for the metabolites suggest that they are relatively less potent than their parent compounds. Given the lower potency of the metabolites relative to their respective parent compounds, existing regulatory standards and industry trends toward safer alternatives may be adequate in limiting human exposure to paraben compounds. However, future biomonitoring studies should attempt to account for the metabolites when determining total daily intake averages in order to prevent an underestimation of an equally important component of potential endocrine disrupting effects upon paraben exposure. SUPPLEMENTARY DATA Supplementary data are available at Toxicological Sciences online. ACKNOWLEDGMENTS We would to like to thank Dr. Vladimir Belov, Max Planck Institute for Biophysical Chemistry, Germany, for synthesizing the hydroxylated metabolites used in this study. We acknowledge the assistance provided by the GUMC Tissue Culture Shared Resource, which is supported in part by the Lombardi Comprehensive Cancer Center support grant (NIH/NCI Grant P30-CA051008). FUNDING National Institute of Environmental Health Sciences of the National Institutes of Health under Award Number T32ES007062 (T.L.G.) and the Breast Cancer Research Foundation (BCRF) (N003173 to J.M.R.). REFERENCES Artacho-Cordón F., Arrebola J. P., Nielsen O., Hernández P., Skakkebaek N. E., Fernández M. F., Andersson A. M., Olea N., Frederiksen H. ( 2017). Assumed non-persistent environmental chemicals in human adipose tissue; matrix stability and correlation with levels measured in urine and serum. Environ. Res.  156, 120– 127. Google Scholar CrossRef Search ADS   Azzouz A., Rascon A. J., Ballesteros E. ( 2016). Simultaneous determination of parabens, alkylphenols, phenylphenols, bisphenol A and triclosan in human urine, blood and breast milk by continuous solid-phase extraction and gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal.  119, 16– 26. Google Scholar CrossRef Search ADS   Barr L., Metaxas G., Harbach C. A., Savoy L. A., Darbre P. D. ( 2012). Measurement of paraben concentrations in human breast tissue at serial locations across the breast from axilla to sternum. J. Appl. Toxicol.  32, 219– 232. Google Scholar CrossRef Search ADS   Calafat A. M., Ye X., Wong L. Y., Bishop A. M., Needham L. L. ( 2010). Urinary concentrations of four parabens in the U.S. population: NHANES 2005–2006. Environ. Health Perspect.  118, 679– 685. Google Scholar CrossRef Search ADS   Canutescu A. A., Dunbrack R. L. ( 2003). Cyclic coordinate descent: A robotics algorithm for protein loop closure. Protein Sci.  12, 963– 972. Google Scholar CrossRef Search ADS   Charles A. K., Darbre P. D. ( 2013). Combinations of parabens at concentrations measured in human breast tissue can increase proliferation of MCF-7 human breast cancer cells. J. Appl. Toxicol.  33, 390– 398. Google Scholar CrossRef Search ADS   Chen J., Ahn K. C., Gee N. A., Gee S. J., Hammock B. D., Lasley B. L. ( 2007). Antiandrogenic properties of parabens and other phenolic containing small molecules in personal care products. Toxicol. Appl. Pharmacol.  221, 278– 284. Google Scholar CrossRef Search ADS   Csiszar S. A., Ernstoff A. S., Fantke P., Meyer D. E., Jolliet O. ( 2016). High-throughput exposure modeling to support prioritization of chemicals in personal care products. Chemosphere  163, 490– 498. Google Scholar CrossRef Search ADS   Delfosse V., Grimaldi M., Cavailles V., Balaguer P., Bourguet W. ( 2014). Structural and functional profiling of environmental ligands for estrogen receptors. Environ. Health Perspect.  122, 1306– 1313. Delfosse V., Grimaldi M., Pons J. L., Boulahtouf A., le Maire A., Cavailles V., Labesse G., Bourguet W., Balaguer P. ( 2012). Structural and mechanistic insights into bisphenols action provide guidelines for risk assessment and discovery of bisphenol A substitutes. Proc. Natl Acad. Sci . 109, 14930– 14935. Google Scholar CrossRef Search ADS   Delfosse V., Maire A. l., Balaguer P., Bourguet W. ( 2015). A structural perspective on nuclear receptors as targets of environmental compounds. Acta Pharmacol. Sin  36, 88– 104. Google Scholar CrossRef Search ADS   Dewalque L., Pirard C., Charlier C. ( 2014). Measurement of urinary biomarkers of parabens, benzophenone-3, and phthalates in a Belgian population. Biomed. Res. Int.  2014, 1. Google Scholar CrossRef Search ADS   Dodson R. E., Nishioka M., Standley L. J., Perovich L. J., Brody J. G., Rudel R. A. ( 2012). Endocrine disruptors and asthma-associated chemicals in consumer products. Environ. Health Perspect.  120, 935– 943. Google Scholar CrossRef Search ADS   El-Ashry D., Chrysogelos S. A., Lippman M. E., Kern F. G. ( 1996). Estrogen induction of TGF-alpha is mediated by an estrogen response element composed of two imperfect palindromes. J. Steroid Biochem. Mol. Biol.  59, 261– 269. Google Scholar CrossRef Search ADS   Evans R. M., Scholze M., Kortenkamp A. ( 2012). Additive mixture effects of estrogenic chemicals in human cell-based assays can be influenced by inclusion of chemicals with differing effect profiles. PLoS ONE  7, e43606. Google Scholar CrossRef Search ADS   Frederiksen H., Jorgensen N., Andersson A. M. ( 2011). Parabens in urine, serum and seminal plasma from healthy Danish men determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS). J. Expo. Sci. Environ. Epidemiol.  21, 262– 271. Google Scholar CrossRef Search ADS   Guimaraes C. R., Mathiowetz A. M., Shalaeva M., Goetz G., Liras S. ( 2012). Use of 3D properties to characterize beyond rule-of-5 property space for passive permeation. J. Chem. Inf. Model.  52, 882– 890. Google Scholar CrossRef Search ADS   Guo Y., Kannan K. ( 2013). A survey of phthalates and parabens in personal care products from the United States and its implications for human exposure. Environ. Sci. Technol.  47, 14442– 14449. Google Scholar CrossRef Search ADS   Hines E. P., Mendola P., von Ehrenstein O. S., Ye X., Calafat A. M., Fenton S. E. ( 2015). Concentrations of environmental phenols and parabens in milk, urine and serum of lactating North Carolina women. Reprod. Toxicol.  54, 120– 128. Google Scholar CrossRef Search ADS   Johnson M. D., Zuo H., Lee K. H., Trebley J. P., Rae J. M., Weatherman R. V., Desta Z., Flockhart D. A., Skaar T. C. ( 2004). Pharmacological characterization of 4-hydroxy-N-desmethyl tamoxifen, a novel active metabolite of tamoxifen. Breast Cancer Res. Treat.  85, 151– 159. Google Scholar CrossRef Search ADS   Kang H. S., Kyung M. S., Ko A., Park J. H., Hwang M. S., Kwon J. E., Suh J. H., Lee H. S., Moon G. I., Hong J. H.et al.  , ( 2016). Urinary concentrations of parabens and their association with demographic factors: A population-based cross-sectional study. Environ. Res.  146, 245– 251. Google Scholar CrossRef Search ADS   Krieger E., Dunbrack R. L.Jr, Hooft R. W., Krieger B. ( 2012). Assignment of protonation states in proteins and ligands: Combining pKa prediction with hydrogen bonding network optimization. Methods Mol. Biol.  819, 405– 421. Google Scholar CrossRef Search ADS   Krieger E., Koraimann G., Vriend G. ( 2002). Increasing the precision of comparative models with YASARA NOVA—A self-parameterizing force field. Proteins  47, 393– 402. Google Scholar CrossRef Search ADS   Krieger E., Vriend G. ( 2014). YASARA View—Molecular graphics for all devices—From smartphones to workstations. Bioinformatics  30, 2981– 2982. Google Scholar CrossRef Search ADS   Liao C., Liu F., Kannan K. ( 2013). Occurrence of and dietary exposure to parabens in foodstuffs from the United States. Environ. Sci. Technol.  47, 3918– 3925. Google Scholar CrossRef Search ADS   Lipinski C. A., Lombardo F., Dominy B. W., Feeney P. J. ( 2001). Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev.  46, 3– 26. Google Scholar CrossRef Search ADS   Lippman M. E., Bolan G., Huff K. ( 1976). The effects of estrogens and antiestrogens on hormone-responsive breast cancer in long-term tissue culture. Cancer Res.  36, 4595– 4601. Lippman M. E., Bolan G. ( 1975). Oestrogen-responsive human breast cancer in long term tissue culture. Nature  256, 592– 593. Google Scholar CrossRef Search ADS   Lippman M. E., Krueger K. A., Eckert S., Sashegyi A., Walls E. L., Jamal S., Cauley J. A., Cummings S. R. ( 2001). Indicators of lifetime estrogen exposure: effect on breast cancer incidence and interaction with raloxifene therapy in the multiple outcomes of raloxifene evaluation study participants. J. Clin. Oncol.  19, 3111– 3116. Google Scholar CrossRef Search ADS   Livak K. J., Schmittgen T. D. ( 2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods  25, 402– 408. Google Scholar CrossRef Search ADS   Lu R., Hu X., Zhou J., Sun J., Zhu A. Z., Xu X., Zheng H., Gao X., Wang X., Jin H. ( 2016). COPS5 amplification and overexpression confers tamoxifen-resistance in ERalpha-positive breast cancer by degradation of NCoR. Nat. Commun.  7, 12044. Google Scholar CrossRef Search ADS   Ma D., Chen L., Zhu X., Li F., Liu C., Liu R. ( 2014). Assessment of combined antiandrogenic effects of binary parabens mixtures in a yeast-based reporter assay. Environ. Sci. Pollut. Res. Int . 21, 6482– 6494. Google Scholar CrossRef Search ADS   Meeker J. D., Yang T., Ye X., Calafat A. M., Hauser R. ( 2010). Urinary concentrations of parabens and serum hormone levels, semen quality parameters, and sperm DNA damage. Environ. Health Perspect.  119, 252– 257. Google Scholar CrossRef Search ADS   Mohammed H., D’Santos C., Serandour A. A., Ali H. R., Brown G. D., Atkins A., Rueda O. M., Holmes K. A., Theodorou V., Robinson J. L.et al.  , ( 2013). Endogenous purification reveals GREB1 as a key estrogen receptor regulatory factor. Cell Rep.  3, 342– 349. Google Scholar CrossRef Search ADS   Moos R. K., Angerer J., Dierkes G., Bruning T., Koch H. M. ( 2016a). Metabolism and elimination of methyl, iso- and n-butyl paraben in human urine after single oral dosage. Arch. Toxicol.  90, 2699– 2709. Google Scholar CrossRef Search ADS   Moos R. K., Apel P., Schroter-Kermani C., Kolossa-Gehring M., Bruning T., Koch H. M. ( 2016b). Daily intake and hazard index of parabens based upon 24 h urine samples of the German Environmental Specimen Bank from 1995 to 2012. J. Expo. Sci. Environ. Epidemiol . 27, 591– 600. Google Scholar CrossRef Search ADS   Nishihama Y., Toshima H., Yoshinaga J., Mizumoto Y., Yoneyama M., Nakajima D., Shiraishi H., Tokuoka S. ( 2017). Paraben exposure and semen quality of Japanese male partners of subfertile couples. Environ. Health Prev. Med.  22, 5. Google Scholar CrossRef Search ADS   Nishihama Y., Yoshinaga J., Iida A., Konishi S., Imai H., Yoneyama M., Nakajima D., Shiraishi H. ( 2016). Association between paraben exposure and menstrual cycle in female university students in Japan. Reprod. Toxicol.  63, 107– 113. Google Scholar CrossRef Search ADS   Oishi S. ( 2001). Effects of butylparaben on the male reproductive system in rats. Toxicol. Ind. Health  17, 31– 39. Google Scholar CrossRef Search ADS   Oishi S. ( 2002). Effects of propyl paraben on the male reproductive system. Food Chem. Toxicol.  40, 1807– 1813. Google Scholar CrossRef Search ADS   Okubo T., Yokoyama Y., Kano K., Kano I. ( 2001). ER-dependent estrogenic activity of parabens assessed by proliferation of human breast cancer MCF-7 cells and expression of ERa and PR. Food Chem. Toxicol.  39, 1225– 1232. Google Scholar CrossRef Search ADS   Pastor-Barriuso R., Fernandez M. F., Castano-Vinyals G., Whelan D., Perez-Gomez B., Llorca J., Villanueva C. M., Guevara M., Molina-Molina J. M., Artacho C. F. ( 2016). Total effective xenoestrogen burden in serum samples and risk for breast cancer in a population-based multicase-control study in Spain. Environ. Health Perspect.  124, 1575– 1582. Google Scholar CrossRef Search ADS   Rae J. M., Johnson M. D., Scheys J. O., Cordero K. E., Larios J. M., Lippman M. E. ( 2005). GREB 1 is a critical regulator of hormone dependent breast cancer growth. Breast Cancer Res. Treat.  92, 141– 149. Google Scholar CrossRef Search ADS   Sasseville D., Alfalah M., Lacroix J. P. ( 2015). “Parabenoia” debunked, or “Who's Afraid of Parabens?”. Dermatitis  26, 254– 259. Google Scholar CrossRef Search ADS   Schwartz G. F., Hughes K. S., Lynch H. T., Fabian C. J., Fentiman I. S., Robson M. E., Domchek S. M., Hartmann L. C., Holland R., Winchester D. J. ( 2008). Proceedings of the international consensus conference on breast cancer risk, genetics, & risk management, April, 2007. Cancer  113, 2627– 2637. Google Scholar CrossRef Search ADS   Siegel R. L., Miller K. D., Jemal A. ( 2017). Cancer statistics, 2017. CA Cancer J. Clin.  67, 7– 30. Google Scholar CrossRef Search ADS   Sikora M. J., Cordero K. E., Larios J. M., Johnson M. D., Lippman M. E., Rae J. M. ( 2009). The androgen metabolite 5alpha-androstane-3beta, 17beta-diol (3betaAdiol) induces breast cancer growth via estrogen receptor: implications for aromatase inhibitor resistance. Breast Cancer Res. Treat.  115, 289– 296. Google Scholar CrossRef Search ADS   Sikora M. J., Strumba V., Lippman M. E., Johnson M. D., Rae J. M. ( 2012). Mechanisms of estrogen-independent breast cancer growth driven by low estrogen concentrations are unique versus complete estrogen deprivation. Breast Cancer Res. Treat.  134, 1027– 1039. Google Scholar CrossRef Search ADS   Trott O., Olson A. J. ( 2010). AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem.  31, 455– 461. Valle-Sistac J., Molins-Delgado D., Diaz M., Ibanez L., Barcelo D., Silvia D. iaz., Cruz M. ( 2016). Determination of parabens and benzophenone-type UV filters in human placenta. First description of the existence of benzyl paraben and benzophenone-4. Environ. Int.  88, 243– 249. Google Scholar CrossRef Search ADS   Vo T. T., Jung E. M., Choi K. C., Yu F. H., Jeung E. B. ( 2011). Estrogen receptor alpha is involved in the induction of Calbindin-D(9k) and progesterone receptor by parabens in GH3 cells: A biomarker gene for screening xenoestrogens. Steroids  76, 675– 681. Google Scholar CrossRef Search ADS   Watanabe Y., Kojima H., Takeuchi S., Uramaru N., Sanoh S., Sugihara K., Kitamura S., Ohta S., Kitamura S. ( 2013). Comparative study on transcriptional activity of 17 parabens mediated by estrogen receptor alpha and beta and androgen receptor. Food Chem. Toxicol.  57, 227– 234. Google Scholar CrossRef Search ADS   Watkins D. J., Ferguson K. K., Anzalota Del Toro L. V., Alshawabkeh A. N., Cordero J. F., Meeker J. D. ( 2015). Associations between urinary phenol and paraben concentrations and markers of oxidative stress and inflammation among pregnant women in Puerto Rico. Int. J. Hyg. Environ. Health  218, 212– 219. Google Scholar CrossRef Search ADS   Wielogorska E., Elliott C. T., Danaher M., Connolly L. ( 2015). Endocrine disruptor activity of multiple environmental food chain contaminants. Toxicol. In Vitro  29, 211– 220. Google Scholar CrossRef Search ADS   Windham G. C., Elkin E., Fenster L., Waller K., Anderson M., Mitchell P. R., Lasley B., Swan S. H. ( 2002). Ovarian hormones in premenopausal women: variation by demographic, reproductive and menstrual cycle characteristics. Epidemiology  13, 675– 684. Google Scholar CrossRef Search ADS   Wróbel A., Gregoraszczuk E. Ł. ( 2013). Effects of single and repeated in vitro exposure of three forms of parabens, methyl-, butyl- and propylparabens on the proliferation and estradiol secretion in MCF-7 and MCF-10A cells. Pharmacol. Rep.  65, 484– 493. Google Scholar CrossRef Search ADS   Wrobel A. M., Gregoraszczuk E. L. ( 2014). Actions of methyl-, propyl- and butylparaben on estrogen receptor-alpha and -beta and the progesterone receptor in MCF-7 cancer cells and non-cancerous MCF-10A cells. Toxicol. Lett.  230, 375– 381. Google Scholar CrossRef Search ADS   © 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 7, 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