Estrogen agonistic/antagonistic activity of brominated parabens

Kohei Sasaki1 & Masanori Terasaki1

Abstract

The estrogen agonistic/antagonistic activity of 16 brominated by-products of parabens was assessed by using a yeast two-hybrid assay transfected with the human estrogen receptor α. Characterization of synthetic compounds including novel brominated parabens was performed using 1H-NMR spectroscopy and high-resolution mass spectrometry. For the agonist assay, five C3–C4 alkylparabens exhibited significant activity (P < 0.05) relative to that of 17β-estradiol, ranging from 3.7 × 10−5 to 7.1 × 10−4. In contrast, none of the brominated alkyl parabens exhibited agonistic activity. In the antagonist assay, 12 brominated alkylparabens and butylparaben exhibited significant antagonistic activity (P < 0.05). Their antagonistic activity relative to 4-hydroxytamoxifen ranged from 0.11 to 2.5. The antagonist activity of C1–C4 alkylparabens increased with the number of bromine substitutions. Benzylparaben exhibited both agonistic and antagonistic activity, and these activities dissipated or were weakened with increased bromination. Thus, increased bromination appeared to attenuate the estrogen agonistic activity of most parabens such that it resulted in increased antagonistic activity, a feature of parabens that had not been previously described.

Keywords Personal care products . Disinfection by-products . Endocrine disruption . Agonist . Antagonist . Yeast two-hybrid assay

Introduction

Parabens are a group of alkyl or benzyl esters of 4hydroxybenzoic acid that are used as preservatives in pharmaceuticals and personal care products (PPCPs). Many PPCPs are used daily in various human activities. Because they are widely used, parabens are continuously released into aquatic environments via domestic wastewater; total concentrations of up to 966 ng L have been detected in river water receiving effluents from sewage treatment (Kusk et al. 2011). The parabens present in treated waters are easily chlorinated by residual chlorine in the form of hypochlorous acid (HOCl) released from tap water or bleaching agents, which leads to the formation of chlorinated parabens (Canosa et al. 2006). Several monitoringstudies havedetected chlorinated parabens in effluents, pool water, and river water (Terasaki et al. 2012, 2015). In the presence of bromide ion, HOCl can rapidly oxidize to hypobromous acid (HOBr). Subsequently, HOBr and parabens can react to form brominated parabens. In fact, brominated methylparaben was identified in raw wastewater and river water (González-Marinõ et al. 2011). Furthermore, brominated parabens have been identified in experiments in which paraben-containing PPCPs were mixed with tap water (Canosa et al. 2006).
Parabens given orally and parenterally are practically nontoxic to mammals, with high 50% lethal dose (LD50) values of 125–8000 mg kg−1 and 640–8000 mg kg−1, respectively (Soni et al. 2005). However, some studies have shown that parabens exhibit estrogen agonistic activity. The recombinant yeast estrogen screen transfected with the human estrogen receptor α (hERα) gene together with expression plasmids has demonstrated agonistic activity of C1–C4 alkyl and benzyl paraben (Miller et al. 2001). Similarly, in vitro studies using estrogen-dependent MCF7 human breast cancer cells (Vanparys et al. 2006), and in vivo studies using rat (Routledge et al. 1998) or mice uterotrophic assays (Shaw and de Catanzaro 2009), have revealed that parabens exert varying degrees of estrogen agonistic activity. In all studies, paraben compounds possessed weak estrogen agonistic activity, with a potency ranging from 10−4- to 10−7-fold compared to the activity of 17 β-estradiol (E2). Byford et al. (2002) researched the antagonistic effect of parabens by studying the competitive inhibition of [3H]-labeled E2 binding to estrogen receptor α (ERα) in MCF7 cells. For paraben concentrations up to 105 M, no significant antagonistic effect against E2induced cell proliferation (10−10 M E2) was observed.
The estrogenagonistic activity ofhalogenatedparabenshas also been assessed. In yeast two-hybrid assays that incorporated hERα or ERα derived from the freshwater fish Oryzias latipes, chlorination masked the apparent estrogen agonistic activity of the parent compounds (Terasaki et al. 2009). However, no studies have reported on the estrogen agonistic and antagonistic activities of brominated parabens. The main reason for this is the lack of chemical standards. In view of the potential toxicological significance of brominated parabens, the objectives of this study was to characterized novel synthetic brominated parabens and investigate their estrogen agonistic/antagonistic activities using a yeast two-hybrid assay incorporating hERα.

Materials and methods

Chemicals and instruments

The structures of the chemicals tested in this study are shown in Table 1. Methyl- (MP, 99% purity), ethyl- (EP, 99% purity), butylparaben (BP, 98% purity), and tetramethylsilane (TMS, 99.5%) were purchased from Wako Pure Chemical Industries, Osaka, Japan. Propyl- (PP, 99% purity), isopropyl- (iPP, 99% purity), isobutyl- (iBP, 99% purity), secondarybutyl- (sBP, 98% purity), benzylparaben (BnP, 98% purity), 4hydroxybenzoic acid (HBA, 98% purity), 3-bromo-4hydroxybenzoic acid (BrHBA, 97% purity), and 3,5dibromo-4-hydroxybenzoic acid (Br2HBA, 98% purity) were purchased from Tokyo Chemical Industry, Tokyo, Japan. These test chemicals were special grade reagents that were used without purification. Other compounds used in this study were synthesized as described below.
Table 1 Structuresand abbreviations ofparabens, mono-anddibrominated parabens, and brominated hydroxybenzoicacids. Thevalues inparentheses are the logarithm of the octanol–water partition coefficients (log Kow) values calculated using US EPA EPI Suite software Proton nuclear magnetic resonance (1H-NMR) spectra were recorded in chloroform-d1 using TMS as an internal standard on an AVANCE III 500 spectrometer (Bruker BioSpin, Rheinstetten, Germany) operating at 500 MHz. Chemical shifts are recorded in δ (ppm) values relative to  DMSO control in the yeast assay. Each point represents the mean ± standard deviation (n = 3)
TMS δ 0.00. High-resolution fast atom bombardment (HR FAB) MS was measured on a JEOL JMS-700 (JEOL, Tokyo, Japan). Gaschromatograph-flameionizationdetector(GC-FID)wasperformed on a GC353 (GL Sciences, Tokyo, Japan) with a HP-5 fused silica capillary column (30 m × 0.32 mm i.d., 0.25 mm film, J&W Scientific, Folsom, CA, USA). Sample (10 mg) was dissolved in ethyl acetate (1 mL) and 1–2 μL was injected into the GC for purity measurement. The purity is determined as the percentage of peak area of the sample relative to the total area of all peaks on the GC-FID chromatogram. Microplate reading was performed on a Multiskan GO (Thermo Fisher Scientific, Yokohama, Japan). The partition coefficients for octanol and water was calculated using the KOWWIN programs (U.S. Environmental Protection Agency 2012).

Synthesis of monobrominated alkylparabens

Each paraben (1.0 g) dissolved in a carbon disulfide (3.0 mL) was added to bromine (1.3 equivalent: Eq) dropwise and capped, and the mixture was heated at 80 °C for 4 h. After cooling, the crude product was transferred to water and extracted using ethyl acetate. The organic solution was washed witha saturated NaCl solution(6M)and dried overanhydrous sodium sulfate. The solvents were evaporated under reduced pressure to yield monobrominated paraben, which was crystallized from hexane/ethyl acetate.

Synthesis of dibrominated alkylparabens

Each paraben (500 mg) dissolved in a carbon disulfide (3.0 mL) was added to bromine (4.0Eq) dropwiseand capped, and the mixture was heated to 50 °C for 1 h. After cooling, the crude product was transferred to water and extracted using ethyl acetate. The organic solution was washed with a saturated NaCl solution and dried over anhydrous sodium sulfate. The solvents were evaporated under reduced pressure to yield dibrominated paraben, which was crystallized from hexane/ ethyl acetate.

Synthesis of mono- and dibrominated benzylparaben

Mono- or dibromo-4-hydroxybenzoic acid (1.0 g) was dissolved in 10 mL of N,N-dimethylformamide and KHCO3 (0.46 g) and added to benzyl bromide (1.2 g) dropwise and capped, and then the mixture was heatedat 40°C for 3 h.After cooling, the reaction mixture was added to 1 mL of methanol solution of K2CO3 (0.1 g) and water (10 mL) and extracted using ethyl acetate. The organic solution was washed with a saturated NaCl solution and dried over anhydrous sodium sulfate. After removal of the solvent under reduced pressure, the residue was chromatographed on a silica gel column (250 × 35 mm) and eluted with hexane/ethyl acetate (7/3, v/ v) to give brominated benzylparaben, which was crystallized from hexane/ethyl acetate.

Yeast two-hybrid assay

The agonistic activity of brominated paraben was examined by using a two-hybrid assay with the yeast cells (Saccharomyces cerevisiae Y190) transfected with hERα and the coactivator transcriptional intermediary factor 2 (TIF2). The assay procedure was performed as described by Kawagoshi et al. (2002) with minor modifications. Briefly, yeast cells were preincubated in a 2% glucose medium lacking tryptophan and leucine for 18 h at 30 °C under shaking conditions. The preincubated cell suspension was then inoculated into a 0.1% glucose medium lacking tryptophan and leucine. Then, 100 μL of the mixture was transferred into each well of a 96-well microplate. A solution (1.25 μL) of the test compound in dimethyl sulfoxide (DMSO) was mixed in the first row of a 96-well microplate (final DMSO concentration, 0.5% v/v). The test solution was serially diluted 5-fold serially from rows 1 to 7. The plate was sealed with an adhesive sheet, shaken mildly, and then incubated at 30 °C for 18 h without shaking. After 18 h of incubation, the cells were digested by addition of 50 μL of Z buffer containing 1.5 mg mL−1 Zymolyase 20 T (Nacalai Tesque, Kyoto, Japan) and incubated at 30 °C for 1 h. Then, 50 μL of 0.5 M phosphate buffer (pH 7.0) containing 0.5 mg mL−1 chlorophenol red-βgalactopyranoside (CPRG) (Wako Pure Chemical Industries, Osaka, Japan) was added, and the samples were reacted at 30 °C for 1 h; the colorimetric reaction was stopped by addition of 40 μL of 2 M Na2CO3. β-Galactosidase activity was assessed by measuring the absorbance at 540 nm (A540) and 690 nm (A690) calculated as follows (Eq. 1): β‐Galactosidase activity ¼ A540–1:2 A690 ð1Þ
E2 was used as a positive control for agonist activity. E2 −4 concentrations ranging from 3.2 × 10 to 5.0 nM were tested. E2 yielded a typical dose-dependent response curves (Fig. 1); the E2 maximum response for β-galactosidase activity was 1.6 at 5.0 nM and was set as 100%. As most compounds did not reach 50% of the E2 maximum response (i.e., EC50), their agonist activities were defined as REC10 (10% relative effective concentration), that is, the concentration of the test compound showing 10% of the E2 maximum response. The dose– response curve of each compound was derived using logistic regression analysis with the following equation: where X is the sample concentration (nM), Y is the response (from bottom to top, with a sigmoid shape), Ks is the substrate constant, and slope is the Hill coefficient. Ks and slope were determined by using the Solver function in Microsoft Excel 2016.
An antagonist activity test was designed to examine the ability of a test chemical to compete with E2 for binding to hERα and/or to inhibit the receptor functions. This assay was performed using the same procedure as for the agonist test, except that the competition was between a test chemical and E2. E2 solution was added to the initial culture and transferred into 96-well platesto attain a final E2 concentration of200 pM after the addition of yeast cells. 4-Hydroxytamoxifen (4HT) was used as a positive control for antagonist activity. 4HT concentrations ranging from 0.64 to 50,000 nM were tested. 4HT yielded typical dose-dependent inhibition curves for βgalactosidase activity (Fig. 2). The maximum inhibitory response was 0.055 (bottom) at 50,000 nM, whereas the DMSO control produced a response of 0.41 (top) in the presence of 200 pM E2; therefore, the 4HT curve-height (top– bottom) was set as 100%. Antagonist activities were defined as the REC60 (60% relative effective concentration), that is, the concentration of the test compound showing 60% of the maximum inhibitory activity of 4HT. The viability was measured by using the optical density at 620 nm (OD620) to determine cell concentration. All parabens did not show any significant absorbance at 620 nm (within 3.0% to the absorbance of control cells).The results ofthe cytotoxicity test were used to determine the concentration of test chemical that reduced the OD620 value of control cells by 20% or more (Fig. 1S in Supplemental); this concentration was defined as cytotoxic and was omitted from data analysis for estrogenic activities. The DMSO concentration in the antagonist assay mixture was 2%, and we confirmed in a preliminary study that DMSO at this concentration did not substantially affect the viability of the cells or the agonist or antagonist activity.

Statistical analysis

Data are expressed as the mean values ± standard deviation (SD) from assays performed in triplicate wells and the intraand interassay coefficients of variation were in the range of 0.84 to 26% and 4.0 to 28%, respectively. Multiple comparisons were made by one-way analysis of variance (ANOVA) followed by Dunnett’s test. A p value of less than 0.05 indicated statistical significance. R software, version 3.0.0 (R core team; Vienna, Austria) for Windows software program, was used for statistical analyses. The REC10 or REC60 values were normalized in comparison to a dose–response curve for E2 or 4HT run in parallel in the same plate (Rivas Ibáñez et al. 2017).

Results and discussion

Synthesis of the compounds used in this study

Brominated parabens were synthesized and characterized by H NMR spectroscopy and high-resolution FAB-MS, and their purities were confirmed to be 98% or greater (Table 2).

Estrogen agonistic and antagonistic activity of brominated parabens

In the agonist assay, 7 compounds exhibited estrogenic activity, with over 10% of the E2 maximum response (Fig. 3). The activity relative to E2 (RAE2) for these compounds ranged from 9.0 × 10−6 to 7.1 × 10−4 (Table 3). PP, iPP, iBP, and sBP yielded steep dose-dependent response curves (Fig. 3a). Br2BnPyielded a shallow dose-dependent responsecurve that plateaued at high concentrations (Fig. 3c). One explanation might be the increased turbidity that was observed in this sample, which may be explained by precipitation of the poorly soluble compounds during the incubation period. The RAE2 values of the active compounds yielded a linear relationship with log Kow (R2 = 0.637), indicating that agonist activity increased progressively with increasing hydrophobicity of the paraben (Fig. 4). These structural features required for optimal agonistic activity matched with previous results (Routledge et al. 1998; Byford et al. 2002). Notably, the RAE2 values of all active parabens dissipated or weakened with bromination, indicating attenuation of estrogen agonistic activity. These results are analogous to those observed with bromination of steroid hormones such as mono- and dibrominated estrone (Nakamura et al. 2006) and ethinylestradiol (Flores and Hill 2008) in the yeast estrogen receptor transcription assay. A similar tendency with bromination has been reported for xeno-estrogens, tert-butylphenol (Olsen et al. 2002), and nonylphenol (García-Reyero et al. 2004; Hill and Smith 2006) in the recombinant yeast assay and the human breast cancer cell MCF7 proliferation assay. Polybrominated diphenyl ethers (PBDEs), a class of brominated flame retardants, have shown estrogenic effects, and one of the possible mechanisms is binding of hydroxylated PBDEs to the estrogen receptor (Meerts et al. 2001; Li et al. 2013). The structural requirements for such hydroxylated PBDE have also been investigated. For 3- or 4-hydroxylated PBDEs, adjacent bromo substitution of the hydroxy group has been shown to markedly decrease the estrogenic activity in a reporter assay using MCF7 cells (Kitamura et al. 2008). The above evidence supports our results that adjacent bromine substitution of the phenol in paraben compounds reduced agonist activity.
A previous study using a yeast two-hybrid assay found that chlorination reduces the apparent estrogen agonist activity of the parent parabens (Terasaki et al. 2009). Taken together, these findings suggest that halogenation masks the apparent estrogen agonistic activity of parabens. This observation is also analogous to those made with mono- and dichlorinated derivatives of nonylphenol using a GFP-MCF7 cell expression system (Kuruto-Niwa et al. 2005). However, the degree of masking appears to differ. Although a previous study found that 3 of the 14 chlorinated parabens exhibited agonist activity with RAE2 values ranging from 2.0 × 10−5 to 9.2 × 10−5 (Terasaki et al. 2009), only 1 (dibrominated benzylparaben; Br2BnP) of the 18 brominated parabens in the present study exhibited agonistic activity. Thus, bromination may have a greater effect on the masking of estrogen agonistic activity than chlorination. We noted that the primary degradation products of parabens (HBA, BrHBA, and Br2HBA) did not elicit significant agonistic responses at the concentration ranges tested in this study. Thus, hydrolysis of the ester group in agonist-active paraben molecules appears to eliminate the activity of these compounds.
In the antagonist assay, 2 non-brominated (BP and BnP), 5 monobrominated (BrEP, BrPP, BrBP, BriBP, and BrsBP), and 7 dibrominated parabens (Br2MP, Br2EP, Br2PP, Br2iPP, Br2BP, Br2iBP, and Br2sBP) produced a significant reduction in β-galactosidase activity in a dose-dependent manner, with over a 60% of the maximum inhibitory activity of 4HT (Fig. 5). The antagonist activity relative to 4HT (RA4HT) for the 14 active compounds ranged from 0.11 to 2.5; particularly, three brominated parabens (Br2BP, Br2iBP, and Brs2BP) showed greater activity than 4HT (Table 3). Their log RA4HT values were plotted against their log Kow values and the number of bromine atoms (Fig. 6). The RA4HT values of the active compounds increased progressively with increasing hydrophobicity of the alkyl ester group for each paraben (R2 = 0.915, in Fig. 6a). Furthermore, the RA4HT values of antagonist active compounds increased with the degree of paraben bromination (Fig. 6b). Thus, brominated alkylparabens appeared to be more active than their corresponding parent parabens. The results also revealed that dibrominated parabens with C4 alkyl parabens (Br2BP, Br2iBP, and Brs2BP) had the highest antagonistic activity, which decreased with sterically larger substituents. BP and BnP with hydrophobic side chains exhibited activity in both the agonist and antagonist assays. For such compounds, benzophenone [(C6H5)2CO]-like hydrophobicity has been reported in the yeast-based reporter assay (Kolle et al. 2010). In the agonist test, BP and BnP showed a marked decline in estrogenic activity at high concentrations (Fig. 3a). At such concentrations, BP and BnP inhibited estrogenic activity demonstrating antagonistic properties (Fig. 5a). This may indicate a much higher frequency of antagonistic than agonist estrogenic activity at high concentration.
Overall, a total of 27 chemicals were tested by using the yeast two-hybrid assay in this study. These parabens can be classified into three categories according to their observed estrogenic activities: (1) pure estrogen agonistic compounds, such as PP, iPP, iBP, sBP and Br2BnP; (2) pure estrogen antagonistic compounds, including Br2MP, BrEP, Br2EP, BrPP, Br2PP, Br2iPP, BrBP, Br2BP, BriBP, Br2iBP, BrsBP, and Br2sBP; (3) compounds with both agonistic and antagonistic activity, including BP and BnP. Parabens were shown to exhibit estrogenic activity as full agonists, though their effect has been considered Bweak^ due to their relatively low estrogen receptor binding affinity in comparison with natural hormones or other endocrine-disrupting chemicals (Golden et al. 2005; Darbre and Harvey 2008). Inthe present study, it has alsobeen shown that the agonist activity of parabens is approximately 106–104 less than that of E2. In addition, van Meeuwen et al. (2008) estimated the cumulative estradiol equivalents (EEQ) for human internal systemic exposure to various cosmetic additives that act through the estrogen receptor, including parabens. They concluded that the total EEQ value is unlikely to cause adverse effects in adult humans. Compared to their estrogen agonistic activity, the estrogen antagonistic activity of parabens has received little attention. Mikula et al. (2006) reported the antiestrogenic potential of PP tested in vivo in juvenile zebrafish (Danio rerio). Zebrafish exposure to PP (0.1, 0.4, and 0.9 mg/L) for 20 days elicited a statistically significant decline in vitellogenin production. Furthermore, parabens were observed to be able to bind competitively to the estrogen-related receptor γ (ERRγ), which is known to modulate estrogen signaling in breast cancer cells (Zhang et al. 2013). MP, EP, PP, BP, and BnP exhibited inverse antagonist activities in the human ERRγ coactivator recruiting assay (REC50 range = 3.09 × 10−7− 5.88 × 10−7 M). In this study based on the yeast two-hybrid assay incorporating hERα, dibrominated derivatives (Br2BP, Br2iBP, and Br2sBP) showed strong antagonist activity (REC60 range = 4.3 × 10−7–8.1 × 10−7 M). Parabens were detected in surface waters, soils and sediments, biota, and in indoor air and dust (Błędzka et al. 2014). Moreover, parabens were detected in human urine, serum, milk, placental tissue, and breast tumor tissue (Błędzka et al. 2014). Although data on brominated parabens levels in the environment and in the human body are currently unavailable, these compounds seem to be more persistent and difficult to metabolize than their parent parabens. This speculation is supported by the fact that the metabolic rate of chlorinated propylparaben was 1/40-fold that of the propylparaben (Terasaki et al. 2016). Therefore, there may be a potential for adverse health consequences resulting from antagonist estrogenic activity of the brominated parabens studied here.
We should note that when parabens underwent electrophilic substitution, most substitutions did not occur at the meta position but rather at the ortho position of the hydroxyl group. Substitutions, such as those in tri- or tetrabrominated parabens, could not occur in aquatic environments. However, further studies are needed to evaluate the estrogen antagonistic activity of highly brominated parabens and to quantitate the structure-activity relationship.

Conclusions

The estrogen agonistic and antagonistic activities of brominated parabens were revealed for the first time using yeast assays. Although bromination attenuated the agonistic activity of most parabens, it also induced antagonist activity, a novel feature for parabens. Most notably, the estrogen antagonistic activity of the most active compounds, Br2iPP, Br2BP, Br2iBP, and Brs2BP, was higher than that of 4HT, the positive control. Thus, further studies of all the components described in our study are necessary to elucidate their activemechanisms and adverse effects.

References

Błędzka D, Gromadzińska J, Wąsowicz W (2014) Parabens. From environmental studies to human health. Environ Int 67:27–42. https:// doi.org/10.1016/j.envint.2014.02.007
Byford JR, Shaw LE, Drew MG, Pope GS, Sauer MJ, Darbre PD (2002) Oestrogenic activity of parabens in MCF7 human breast cancer cells. J Steroid Biochem Mol Biol 80:49–60. https://doi.org/10. 1016/S0960-0760(01)00174-1
Canosa P, Rodríguez I, Rubí E, Negreira N, Cela R (2006) Formation of halogenated by-products of parabens in chlorinated water. Anal Chim Acta 575:106–113. https://doi.org/10.1016/j.aca.2006.05.068
Darbre PD, Harvey PW (2008) Paraben esters: review of recent studies of endocrine toxicity, absorption, esterase and human exposure, and discussion of potential human health risks. J Appl Toxicol 28:561– 578. https://doi.org/10.1002/jat.1358
Flores A, Hill EM (2008) Formation of estrogenic brominated ethinylestradiol in drinking water: implications for aquatic toxicity testing. Chemosphere 73:1115–1120. https://doi.org/10.1016/j. chemosphere.2008.07.022
García-ReyeroN,Requena V, PetrovicM, Fischer B,Hansen PD,DíazA, Ventura F, Barceló D, Piña B (2004) Estrogenic potential of halogenated derivatives of nonylphenol ethoxylates and carboxylates. Environ Toxicol Chem 23:705–711. https://doi.org/10.1897/03-141
Golden R, Gandy J, Vollmer G (2005) A review of the endocrine activity of parabens and implications for potential risks to human health. Crit Rev Toxicol 35:435–458. https://doi.org/10.1080/ 10408440490920104
González-Marinõ I, Quintana JB, Rodríguez I, Cela R (2011) Evaluation of the 4-Hydroxytamoxifen occurrence and biodegradation of parabens and halogenated by-products in wastewater by accurate-mass liquid chromatography-quadrupole-time-of-flight-mass spectrometry (LC-QTOFMS). Water Res 45:6770–6780. https://doi.org/10.1016/j.watres. 2011.10.027
Hill EM, Smith MD (2006) Identification and steroid receptor activity of products formed from the bromination of technical nonylphenol. Chemosphere 64:1761–1768. https://doi.org/10.1016/j. chemosphere.2005.12.040
Kawagoshi Y, Tsukagoshi Y, Fukunaga I (2002) Determination of estrogenic activity in landfill leachate by simplified yeast two-hybrid assay. J Environ Monit 4:1040–1046. https://doi.org/10.1039/ B210962J
Kitamura S, Shinohara S, Iwase E, Sugihara K, Uramaru N, Shigematsu H, Fujimoto N, Ohta S (2008) Affinity for thyroid hormone and estrogen receptors of hydroxylated polybrominated diphenyl ethers. J Health Sci 54:607–614. https://doi.org/10.1248/jhs.54.607
Kolle SN, Kamp HG, Huener HA, Knickel J, Verlohner A, Woitkowiak C, Landsiedel R, van Ravenzwaay B (2010) In house validation of recombinant yeast estrogen and androgen receptor agonist and antagonist screening assays. Toxicol in Vitro 24:2030–2040. https://doi.org/10.1016/j.tiv.2010.08.008
Kuruto-Niwa R, Nozawa R, Miyakoshi T, Shiozawa T, Terao Y (2005) Estrogenic activity of alkylphenols, bisphenol S, and their chlorinated derivatives using a GFP expression system. Environ Toxicol Pharmacol 19:121–130. https://doi.org/10.1016/j.etap.2004.05.009
Kusk KO, Krüger T, Long M, Taxvig C, Lykkesfeldt AE, Frederiksen H, Andersson AM, Andersen HR, Hansen KM, Nellemann C, Bonefeld-Jørgensen EC (2011) Endocrine potency of wastewater: contents of endocrine disrupting chemicals and effects measured by in vivo and in vitro assays. Environ Toxicol Chem 30:413–426.https://doi.org/10.1002/etc.385
Li X, Gao Y, Guo LH, Jiang G (2013) Structure-dependent activities of hydroxylated polybrominated diphenyl ethers on human estrogen receptor. Toxicology 309:15–22. https://doi.org/10.1016/j.tox.2013.04.001
Meerts IATM, Letcher RJ, Hoving S, Marsh G, Bergman A, Lemmen JG, Van Der Burg B, Brouwer A (2001) In vitro estrogenicity of polybrominated diphenyl ethers, hydroxylated PBDEs, and polybrominated bisphenol A compounds. Environ Health Perspect 109:399–407. https://doi.org/10.1289/ehp.01109399
Mikula P, Dobsikova R, Svobodova Z, Jarovsky J (2006) Evaluation of xenoestrogenic potential of propylparaben in zebrafish (Danio rerio). Neuro Endocrinol Lett 27:104–107
Miller D, Wheals BB, Beresford N, Sumpter JP (2001) Estrogenic activity of phenolic additives determined by an in vitro yeast bioassay. Environ Health Perspect 109:133–138. https://doi.org/10.1289/ehp. 01109133
Nakamura H, Shiozawa T, Terao Y, Shiraishi F, Fukazawa H (2006) Byproducts produced by the reaction of estrogens with hypochlorous acid and their estrogen activities. J Health Sci 52:124–131. https://doi.org/10.1248/jhs.52.124
Olsen CM, Meussen-Elholm ET, Holme JA, Hongslo JK (2002) Brominated phenols: characterization of estrogen-like activity in the human breast cancer cell-line MCF-7. Toxicol Lett 129:55–63. https://doi.org/10.1016/S0378-4274(01)00469-6
Rivas Ibáñez G, Bittner M, Toušová Z, Campos-Mañas MC, Agüera A, Casas López JL, Sánchez Pérez JA, Hilscherová K (2017) Does micropollutant removal by solar photo-Fenton reduce ecotoxicity in municipal wastewater ? A comprehensive study at pilot scale open reactors. J Chem Technol Biotechnol 92:2114–2122. https://doi.org/10.1002/jctb.5212
Routledge EJ, Parker J, Odum J, Ashby J, Sumpter JP (1998) Some alkyl hydroxy benzoate preservatives (parabens) are estrogenic. Toxicol Appl Pharmacol 153:12–19. https://doi.org/10.1006/taap.1998. 8544
Shaw J, de Catanzaro D (2009) Estrogenicity of parabens revisited: impact of parabens on early pregnancy and an uterotrophic assay in mice. Reprod Toxicol 28:26–31. https://doi.org/10.1016/j.reprotox. 2009.03.003
Soni MG, Carabin IG, Burdock GA (2005) Safety assessment of esters of p-hydroxybenzoic acid (parabens). Food Chem Toxicol 43:985–1015. https://doi.org/10.1016/j.fct.2005.01.020
Terasaki M, Kamata R, Shiraishi F, Makino M (2009) Evaluation of estrogenic activity of paraben and their chlorinated derivatives by using the yeast two-hybrid assay the enzyme-linked immunosorbent assay. Environ Toxicol Chem 28:204–208. https://doi.org/10.1897/ 08-225.1
Terasaki M, Takemura Y, Makino M (2012) Paraben-chlorinated derivatives in river waters. Environ Chem Lett 10:401–406. https://doi. org/10.1007/s10311-012-0367-1
Terasaki M, Yasuda M, Makino M, Shimoi K (2015) Aryl hydrocarbon receptor potency ofchlorinated parabensin the aquaticenvironment. Environ Sci Water Res Technol 1:375–382. https://doi.org/10.1039/ C5EW00047E
Terasaki M, Wada T, Nagashima S, Makino M, Yasukawa H (2016) In vitro transformation of chlorinated parabens by the liver S9 fraction: kinetics, metabolite identification, and aryl hydrocarbon receptor agonist activity. Chem Pharm Bull 64:650–654. https://doi.org/ 10.1248/cpb.c15-00977.
U.S. Environmental Protection Agency (2012) EPI Suite™-Estimation Program, version 4.11, Washington, D.C.: U.S. Environmental Protection Agency. Available from: https://www.epa.gov/tscascreening-tools/download-epi-suitetm-estimation-programinterface-v411
van Meeuwen JA, van SO, Piersma AH, de Jong PC, van den BM (2008) Aromatase inhibiting and combined estrogenic effects of parabens and estrogenic effects of other additives in cosmetics. Toxicol Appl Pharmacol 230:372–382
Vanparys C, Maras M, Lenjou M, Robbens J, Van Bockstaele D, Blust R, De Coen W (2006) Flow cytometric cell cycle analysis allows for rapid screening of estrogenicity in MCF-7 breast cancer cells. Toxicol in Vitro 20:1238–1248. https://doi.org/10.1016/j.tiv.2006. 05.002
Zhang Z, Sun L, Hu Y, Jiao J, Hu J (2013) Inverse antagonist activities of parabens on human oestrogen-related receptor γ (ERRγ): in vitro and in silico studies. Toxicol Appl Pharmacol 270:16–22. https:// doi.org/10.1016/j.taap.2013.03.030