Potential Estrogenic Effects of Phosphorus-Containing Flame

May 21, 2014 - College of Biological and Environmental Engineering, Zhejiang .... of Precision Engineering and Manufacturing-Green Technology 2017 4 (...
3 downloads 0 Views 635KB Size
Article pubs.acs.org/est

Potential Estrogenic Effects of Phosphorus-Containing Flame Retardants Quan Zhang,†,§ Meiya Lu,† Xiaowu Dong,‡ Cui Wang,§ Chunlong Zhang,∥ Weiping Liu,§ and Meirong Zhao*,† †

College of Biological and Environmental Engineering, Zhejiang University of Technology, Hangzhou 310032, China College of Pharmaceutical Sciences and §Ministry of Education Key Laboratory of Environmental Remediation & Ecosystem Health, Zhejiang University, Hangzhou 310058, China ∥ Department of Environmental Science, University of Houston-Clear Lake, 2700 Bay Area Blvd., Houston, Texas 77058, United States ‡

S Supporting Information *

ABSTRACT: As the substitute of polybrominated diphenyl ethers (PBDEs), further assessments about the potential ecological safety and health risks of phosphorus-containing flame retardants (PFRs) are required because the worldwide demand for PFRs has been increasing every year. In this study, we examined the agonistic/antagonistic activity of a group of PFRs by three in vitro models (luciferase reporter gene assay, yeast two-hybrid assay, and Escreen assay). Molecule docking was used to further explain the interactions between ERα and PFRs. Data from luciferase reporter gene analysis showed three members of the nine tested PFRs significantly induced estrogenic effects, with the order of TPP > TCP > TDCPP, while TCEP and TEHP have remarkable antiestrogenic properties with calculated REC20 and RIC20 values of 10−6 M or lower. Results from the luciferase reporter gene method are generally consistent with results obtained from the yeast two-hybrid assay and E-screen, except for the positive estrogenic activity of TBP in E-screen testing. Docking results showed that binding between ligands and ERα was stabilized by hydrophobic interactions. As a proposed alternative for brominated flame retardant, PFRs may have anti/estrogenic activity via ERα at the low dose typical of residue in environmental matrix or animals. PFRs with a short chain, halogen, and benzene ring in the substituent group tend to be estrogenic. Our research suggests that comprehensive evaluations, including health and ecological assessments, are required in determining whether PFRs are preferable as an emerging industrial substitute.



respectively.1,9 TCPP was detected in Norwegian bird species and Great Lakes herring gull eggs at concentrations of 10 and 4.1 ng/g wet weight, respectively.10,11 Thus, studies regarding the ecological safety and the potential health effects of PFRs are warranted. The early assessment on the environmental safety of PFRs originated from the acute toxicity effect, showing the association between high doses of PFRs and neurological toxicity.12,13 Recently, chronic toxicity effects, especially the endocrine disruption, have attracted researchers’ concerns. The hypothalamic−pituitary−gonad (HPG) axis, a significant neuroendocrine system in mammalians, is vulnerable to xenoestrogenic chemicals via hormone synthesis, metabolites, and ligand binding.14 Epidemiological results indicated that PFRs in house dust altered hormone levels and decreased the sperm

INTRODUCTION

Flame retardants (FRs) are chemicals used in thermoplastics, thermosets, textiles, and coatings that inhibit or resist the spread of fire.1 In the past few years, world usage of the common brominated flame retardants (BFRs) was subsequently forbidden by the European Union and the United States, because BFRs (such as PBDE 99, PBDE 203, PBDE 209) are persistent, bioaccumulative, and toxic to animals, humans, and the environment.2−6 Consequently, phosphoruscontaining flame retardants (PFRs) are expected to be the alternatives for BFRs7 for the high flame-retardant efficiency. The annual production of PFRs was estimated to reach nearly 100 thousand tons in 2011.7 Demands for PFRs increase every year with a growth rate of 15% in China. Large consumption of PFRs resulted in high detection frequency in various environmental media and biotas. Organisms are proved to be exposed to PFRs through different routes, such as surface water, sediment, and dust.1,8 Aquatic organisms, including mussels and fishes, accumulated TCPP and TDCPP up to 1300 and 140 ng/g lipid weight, © 2014 American Chemical Society

Received: Revised: Accepted: Published: 6995

February 15, 2014 May 19, 2014 May 21, 2014 May 21, 2014 dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology

Article

Table 1. Information on Nine Organic Phosphorus Flame Retardants chemicals and abbreviation

CAS no.

purity (%)

tributyl phosphate (TBP) tricresyl phosphate (TCP) triphenyl phosphate (TPP) tris(2-butoxyethyl)phosphate (TBEP) tris(2-chloroethyl)phosphate (TCEP) tris(2-chloro-1-(chloromethyl)ethyl)phosphate (TDCPP) tris(2-chloroisopropyl)phosphate (TCPP) tris(2,3-dibromopropyl)phosphate (TBPP) tris(2-ethylhexyl)phosphate (TEHP)

126-73-8 1330-78-5 115-86-6 78-51-3 115-96-8 13674-87-8 13674-84-5 126-72-7 78-42-2

99.5 99.4 99.5 95.8 98.5 95.5 96.0 93.0 98.0

The standard solution of E2 was prepared from its stock concentration of 10−2 M in DMSO. The nine PFRs were also dissolved in DMSO with the initial concentrations of 10−1 M, and stored at −20 °C. Before the experiment, all chemicals were immediately diluted to the experimental concentrations in RPMI-1640 medium (Hyclone; Logan, UT, U.S.A.) with 10% fetal bovine serum (FBS; Hyclone). To avoid any cytotoxic effect of DMSO, its final concentration in the culture medium was kept below 0.1% (V/V), according to a previous study.21 Cell Lines and Culture Condition. CHO-K1 cells (Chinese Hamster Ovary cell line) and MCF-7 cells (human breast adenocarcinoma cell line) were both obtained from the cell bank of the Chinese Academy of Sciences (Shanghai, China; the original source was from ATCC, Manassas, U.S.A.). CHO-K1and MCF-7 cells were grown in RPMI-1640 medium with 10% FBS under the incubation conditions of 37 °C, 5% CO2, and saturating humidity. Every 3 or 5 days, they were separately passaged by trypsinization with 0.25% EDTA disodium salt solution (Gibco, MD, U.S.A.). Plasmids. The luciferase reporter plasmid pERE-AUG-Luc+ with the construction of three copiesthe Xenopus laevis vitellogenin A estrogen responsive element, rat α2u globulin promoter, and rERα/pCI containing the full open reading frame of rat ERα cDNA, used in the ER reporter gene assay, were kindly provided by Dr M. Takeyoshi (Chemicals Assessment Center, Chemicals Evaluation and Research Institute, Oita, Japan). The plasmid phRL-tk used as an internal control for transfection efficiency, containing the Renilla luciferase gene, was purchased from Promega (Madison, WI, U.S.A.). MTS Assay. The cytotoxicity induced by the nine tested PFRs was assessed by 3-(4,5-dimethylthiazol-2-yl)-5-(3carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay on CHO-K1 cells before the estrogenic estimation. The negative control in the MTS assay was the CHO-K1 cells treated with DMSO (0.1%) instead of the test chemicals. The obtained minimum and maximum cell number were 1000 and 2.5 × 105 cells, respectively, according to the manufacturer’s instruction. MTS assay is a colorimetric method for determining the number of viable cells in proliferation or cytotoxicity assays. After adding the MTS reagent, the toxicity of test chemicals was determined by the number of viable cells (indirectly from the absorbance of treated cells). This assay is often described as a “one-step” MTT assay, because the reagent can be added straight to the cell culture without the intermittent steps required in the MTT assay. Cells were collected from the culture flask and seeded in 96well plates (Corning, U.S.A.) at a density of 1 × 104 cells/well, with the RPMI-1640 medium containing 10% FBS. Cells were treated with various concentrations of the test chemicals or

concentration, suggesting that HPG axis can be disrupted by PFRs through altering the levels of steroid hormone. As a group of chemicals with potential interference with the HPG axis, information available on the estrogenic effects of PFRs is scarce, let alone the agonistic/antagonistic of the chemicals. Up to date, only one study has reported the estrogenic potential of PFRs in indoor dust extracts through the estrogenic receptor α (ER α) using the human osteosarcoma (U2OS) cell-based reporter gene assay.15 However, for xeno-estrogen screening, false negative (positive) results are always observed in such a single testing model. Therefore, it is urgent to qualitatively and quantitatively screen the estrogen effects of PFRs through integrated assessments from multiple methods. Luciferase reporter gene assay, yeast two-hybrid assay, and E-screen assay are three popular in vitro models used to evaluate endocrine effects of the chemicals. The first two assays depend on chemicals interacting with the estrogen receptor, whereas the E-screen assay can be employed to assess the estrogenic activities of compounds using the proliferative effect in MCF-7 cells as an end point. In the previous study, estrogenic activities of 20 PCBs were successfully assessed in these models.16 These models have advantages of rapidity, sensitivity, and reproducibility. Molecular docking is another effective tool to study and explain the potential interactions between ligands and receptors. Here we employed three in vitro assays and in silico calculation to ascertain in vitro toxicity results. In the current study, we used these three in vitro assays to rapidly screen nine PFRs for their endocrine disruption potential and quantify the endocrine disruption capacity of suspected compounds. Furthermore, we compared the lowest observed effect levels (LOELs) determined from this study with the residues reported in previous research to evaluate the adverse effects of PFRs to humans or wild animals. Besides, molecule docking was used to further explain the interactions between hERα and PFRs. The integrated data provided here will be valuable for the clarification of binding activities on ERα and better understanding the endocrine disruption potential of each PFR.



MATERIALS AND METHODS Chemicals. Nine PFRs selected in our study are the widely used groups of PFRs which have diverse structures but a general structural moiety of phosphate ester.17 Several PFRs such as TCP, TPP, TDCPP, TCEP, TCPP, and TBEP have been frequently detected in the environment including indoor dust, air, soil, and sediment worldwide.18−20 Nine PFRs were purchased from Dr. Ehrenstorfer GmbH (Augsburg, Germany), with the highest available purity listed in Table 1. 17β-Estradiol (E2, >97%) and dimethyl sulfoxide (DMSO, as a vehicle) were both purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). 6996

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology

Article

DMSO 24 h later, and incubated at 37 °C with 5% CO2 for 24 h. Finally, the proliferation of cells was detected by MTS kit (Cell Titer 96 Aqueous One Solution, Promega). Dual-Luciferase Reporter Gene Assay for ERα. Before transfections, CHO-K1 cells were seeded in 96-well plates containing phenol red-free RPMI-1640 medium with 5% charcoal−dextran-treated fetal bovine serum (CD-FBS; Hyclone) at a density of 15 000 cells/well for 24 h. In order to detect the ERα activity, cells were transfected with 25 ng of rERα/pCI, 145 ng of pERE-AUG-Luc+, and 10 ng of phRL-tk using 0.5 μL of transfection reagent lipofectamine 2000 (Invitrogen, MD, U.S.A.) per well. After 4 h of transfection, cells were replaced with the new culture medium overnight. Then various concentrations of test chemicals or 0.1% DMSO (as negative control) were added into cells. In addition, cells were also treated with various concentrations of the test chemicals along with 10−9 M E2 for the measurement of antagonistic activity to ERα. After 24 h of exposure, cells were rinsed with phosphate-buffered saline (PBS; pH = 7.4) twice and lysed with 20 μL/well 1 × passive lysis buffer (Promega, Madison, WI, U.S.A.). Then following the instructions of the Dual-Luciferase Reporter Assay kit (Promega), both luciferase and Renilla luciferase activity were measured with a fluorescence spectrophotometer (Infinite M200, Tecan, Switzerland). The results were normalized as the ratio to the negative control to indicate the relative transcriptional activity. Yeast Two-Hybrid Assay. The yeast two-hybrid assay in the present study was slightly modified from the manufacturer’s instructions.22 Briefly, yeast cells were incubated at 30 °C with a shake of 150 rpm for 48−64 h. The test chemical or DMSO (1 μL) was added to 999 μL of yeast cells stock solution. Afterward, 200 μL of the test solution was transferred into a 96well plate for 4 h of incubation at 30 °C with a shake of 800 rpm in a thermo-shaker (MB100−4P, AoSheng, China). The cell density of the solution was measured at 600 nm with a model 680 microplate reader (Bio-Rad Laboratories, Hercules, CA, U.S.A.). Subsequently, 150 μL of the test solution was removed, and different reagents were added into the remaining solution in the order indicated in the instruction of kit (State Key Laboratory of Environmental Aquatic Chemistry, China). After vigorously shaking (1200 rpm) for 10 min, 40 μL of βgalactosidase substrate was added for another 60 min incubation at 30 °C. The reaction was terminated by adding 100 μL of stop solution with a 10 min incubation at 800 rpm. Finally, the absorbance of the solution at 420 nm was recorded. Each sample was tested in quadruplicate, and each assay was conducted more than three times. E-Screen Assay. The procedures of the E-screen assay were slightly modified from previously described methods.23−25 In brief, MCF-7 cells (2500 cells/well) were plated in 96-well plates with experimental medium (phenol red-free RPMI-1640 with 5% CD-FBS) for 24 h to allow the attachment to the bottom, after growth in the seeding medium (RPMI-1640 medium with 10% FBS) at 37 °C with 5% CO2. Then the dosing medium (experimental medium containing various concentrations of test chemicals or DMSO) was replaced for 5 days of incubation. Dosing medium was refreshed after 60 h. Finally, cell proliferation was detected by the MTS kit (Cell Titer 96 Aqueous One Solution, Promega) following the instruction of the manufacturer. Twenty microliters of test reagent was added into each well for 40−60 min incubation, and the absorbance at 490 nm was recorded by a Bio-Rad

Model 680 microplate reader (Bio-Rad Laboratories, Hercules, CA, U.S.A.). Molecular Docking and Molecular Dynamic Simulation. Docking studies were achieved using Discovery Studio 2.5/Ligandfit module (Accelrys, Inc., San Diego, CA). The docking site of hERα (PDB entry code: 3ERD) was derived using the Ligand Fit site search utility.26 For the generation of the ligand’s conformations, we used variable numbers of Monte Carlo simulations. All the calculations during the docking steps were performed under the PLP Force Field formalism. A short rigid body minimization was then performed, and 50 poses for each ligand were saved. Scoring was performed with a set of scoring functions (including Dock_score, LigScore2, PLP1, Jain, and PMF) implemented in the LigandFit module. The combination of consensus scoring method and the interaction mode was applied to select the preferable output conformation. Furthermore, the TPP-bound hERα complex proposed by LigandFit study was further refined by molecular dynamic simulation using Discovery Studio 2.5/Simulation module, with an aim to examine the reliability of the interaction mode of TPP: (a) minimization was initially run with CHARMm forcefield for 1000 iterations of steepest descents, followed by a conjugate gradient optimization until the maximum derivative of energy became less than 0.1 kcal·mol−1·Å−1; (b) the systems were heated from 50 to 300 K over 0.1 ns in the NVT ensemble; (c) molecular dynamic simulations with CHARMm force-field were then performed at a constant temperature of 300 K with a time step of 1 fs for 1 ns equilibration and 3 ns production, in an NVT ensemble. The other parameters of molecular dynamic simulation were maintained at the default configurations. Data Analysis. The statistical analysis of results was conducted by the statistical program package Origin 8.0 (OriginLab, Northampton, MA). All the results are presented as the mean ± SD (standard deviation), obtained from triplicate experiments. Significance of mean difference between groups was assessed by one-way analysis of variance (ANOVA) followed by Dunnett’s multiple-comparison test, and the level of significant difference was p < 0.05.



RESULTS Cytotoxicity of PFRs. Below the concentrations of 1 × 10−5 M for TBP, TBEP, TCEP, TDCPP, TCPP, TEHP, and 1 × 10−6 M for TCP, TPP, TBPP, none of the nine PFRs with the test concentrations affected viability or proliferation of CHO-K1 cells alone (see Figure S1 in Supporting Information (SI) ) or in the presence of 10−9 M E2. No cytotoxic effects of solvent were observed by microscopic examination throughout the transfection assay. Estrogenic/Antiestrogenic Effects of PFRs by DualLuciferase Reporter Gene Assay. Estrogenic disrupting effects of nine PFRs mediated via ERα were estimated by dualluciferase reporter gene assay. From the dose−response curve, we estimated the EC20 (the concentration of E2 with its activity achieved 20% of its maximal activity) value of E2 for ERα was 1.6 × 10−10 M. Relative potency was represented as EC20 instead of EC50, because the test chemicals did not show the maximum induction as E2.27 As shown in Table 2, we found that of nine PFRs, only TCP, TPP, and TDCPP induced estrogenic activity, and the order of relative potencies according to REC20 was TPP > TCP > TDCPP. Obviously, TPP showed the most potent estrogenic activity among the three compounds with the REC20 value of 2.7 × 10−7 M. Figure S4 6997

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology

Article

Table 2. Agonistic and Antagonistic Activities of Nine OPFRs against ERα in the in Vitro Dual-Luciferase Reporter Gene Assay compound

REC20 (M)

TBP TCP TPP TBEP TCEP TDCPP TCPP TBPP TEHP

ND 8.8 × 10−7 2.7 × 10−7 ND ND 6.4 × 10−6 ND ND ND

RLAa (%) 20 23

21

RIC20 (M)

RLAb (%)

ND ND ND ND 8.2 × 10−6 ND ND ND 1.9 × 10−6

78

Table 3. Agonistic and Antagonistic Activities of Nine OPFRs against ERα in Yeast Two-Hybrid System

52

compound

REC20 (M)

RIC20(M)

TBP TCP TPP TBEP TCEP TDCPP TCPP TBPP TEHP

ND 9.7 × 10−7 6.5 × 10−7 ND ND 7.8 × 10−6 ND ND ND

ND ND ND ND 1.0 × 10−5 ND ND ND 5.3 × 10−6

ND: not detected. REC20: concentration of the tested chemicals showing 20% activity of the maximum activity of 1 × 10−9 M E2. RIC20: concentration of the tested chemicals showing 20% activity of inhibition by 1 × 10−10 M E2.

ND: not detected. REC20: concentration of the tested chemical showing 20% activity of the 2 × 10−8 M E2 via ERα RIC20: concentration of the tested chemical showing 20% activity of inhibition of 1 × 10−9 M E2 via ERα aRLA: relative luciferase activity. Percentage response of maximum activity of organic PFRs with 100% activity defined as the activity achieved for E2 at 2 × 10−8 M bRLA: relative luciferase activity. Percentage response of maximum inhibition of organic PFRs with 100% activity defined as the activity achieved for E2 at 1 × 10−9 M

Table 4. Estrogenic Effect of Nine OPFRs Measured by the E-Screen Assay compound E2 TBP TCP TPP TBEP TCEP TDCPP TCPP TBPP TEHP

(SI) reveals the dose-dependent ERα estrogenic activity of TPP, TCP, and TDCPP. Values of the lowest observed effect level (LOEL) of TPP, TCP, and TDCPP obtained from the dual-luciferase reporter gene assay were 1.0 × 10−7 M, 1.0 × 10−7 M, and 1.0 × 10−6 M, respectively. In addition, TCEP and TEHP were found to have antiestrogenic properties induced by 10−9 M E2 (Table 2). The dose-related manners of TCEP and TEHP are shown in Figure S5. The relative potencies of their antiestrogenic activities for ERα descended in the following order: TEHP > TCEP, with the RIC20 value of 1.9 × 10−6 and 8.2 × 10−6 M, respectively. Values of the lowest observed effect level (LOEL) of TEHP and TCEP were 1.0 × 10−6 M and 1.0 × 10−5 M. No anti/estrogenic activity was observed for TBP, TBEP, TCPP, and TBPP at the test concentrations. Estrogenic Effects of PFRs by Yeast Two-Hybrid Assay and E-Screen Assay. To further investigate the estrogenic effects of nine PFRs, we employed two other in vitro assays, that is, yeast two-hybrid reporter assay (a Saccharomyces cerevisiae-based lac-Z (β-galactosidase)) and E-screen assay. For yeast two-hybrid assay, the β-glactosidase activity induced by E2 reached a plateau at 1 × 10−9 M. The EC20 value of E2 deduced from the E2 dose−response curve was 3.5 × 10−11 M (data not shown). The estrogenic and antiestrognic activities of nine PFRs in yeast two-hybrid reporter assay are listed in Table 3. TPP, TCP, and TDCPP exerted estrogenic activity (Figure S2, SI), whereas TCEP and TEHP showed the antiestrognic activity (Figure S3, SI). Similarly, TPP showed the greatest estrogenic activity among the three compounds (TPP > TCP > TDCPP) with the REC20 value of 6.5 × 10−7 M. The order of the relative potencies for antiestrogenic activity was also shown as TEHP > TCEP. Estrogenic effects of nine PFRs measured through the Escreen assay are shown in Table 4. We found that of the nine PFRs tested, TBP, TCP, TPP, and TDCPP expressed the significant estrogenic activity by inducing the proliferation of MCF-7 cells, and the other five PFRs showed no effect at the test concentrations (Figure S6, SI). TPP induced the maximum cell proliferation at 1 × 10−6 M, followed by TCP and TDCPP. Curves of MCF-7 cells proliferation at different concentrations

concentration (M) 1 1 1 1 1 1 1 1 1 1

× × × × × × × × × ×

10−9 10−7 10−7 10−6 10−5 10−6 10−6 10−5 10−7 10−5

PE

RPE (%)

2.1523 1.3049 1.4104 1.4320 1.1944 1.1701 1.3351 1.1973 1.1251 1.1066

100 26.46 35.62 37.49 16.87 14.76 29.08 17.12 10.86 9.25

PE: the proliferative effect, calculated as the ratio of the maximal cell yield of the tested chemicals to the cell yield of the DMSO control. RPE: the relative proliferative effect, calculated as the ratio (PE-1) of the tested chemicals over (PE-1) of 1 × 10−9 M E2 ( × 100%).

of TBP, TCP, TPP, and TDCPP are plotted in Figure 1. Overall, the cell proliferation increased with the increasing concentration. Molecular Docking and Molecular Dynamic Simulations. All PFRs were docked into the active pocket of hERα

Figure 1. Proliferation of MCF-7 cells grown in experimental medium (phenol-red-free RPMI-1640 medium supplemented with 5% CDFBS) exposed to various concentrations of TBP, TCP, TPP, TDCPP for 5 days. Values represent mean ± SD of three independent experiments. * p < 0.05 compared with the value of DMSO control. 6998

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology

Article

equilibrium state had been reached. Comparing the initial and lowest-energy conformations of the backbone atom of the complexes indicated that the relative RMSD fluctuations were very small. It can be noted that interactions between hERα and TPP were stable throughout the simulations.

using LigandFit program. Except for TEHP, the binding modes of all compounds were proposed. It was observed that the consensus scoring results of TPP, tri-p-tolyphosphate, tri-otolyphosphate, and tri-m-tolyphosphate are better than that of TCEP, TBP, TBEP, TDCPP, TCPP, TBPP, and TEHP, indicating that three phenyl rings of PFRs are preferable in binding to the active pocket of hERα in our docking studies. As exemplified by TPP, the hydrophobic interactions were highlighted that three phenyl rings of TPP insert into three hydrophic pockets (Pocket A: Leu349, Ala350, Leu387, Met388, and Phe404; pocket B: Ile424, Phe425, Leu428 and MET388; pocket C: Met343, Met421, and Leu525) of hERα, respectively (Figure 2). Further, the scores from DOCK_Score,



DISCUSSION To our knowledge, the present study gave a qualitative and quantitative analysis of the estrogenic and antiestrogenic effects of nine PFRs for ERα. Results showed that TCP, TPP, and TDCPP behaved as an ERα agonist, and the relative estrogenic potencies are TPP > TCP > TDCPP. Comparatively, TCEP and TEHP had antiestrogenic activities. Previous studies identified the endocrine disruption of PFRs in thyroid hormone disturbance. As reported, T4 contents were significantly decreased by 66.4%, 58.6%, 66.8%, and 77.0% in 50, 100, 300, and 600 μg/L exposure groups, respectively. Significant effects on the whole-body T3 content were observed upon exposure to 300 μg/L TDCPP.28 Data illustrated herein raise a question as to whether PFRs are the suitable substitute for BFRs. PFRs are widely distributed in the environment, but evidence about their ecological and health risks are limited. Here, we converted the molar concentration unit to m/v unit (ppb) to facilitate the comparison between the LOEL (the dualluciferase reporter gene assay) and the residue levels in previous reports, assuming negligible differences related with species, bioavailability, and metabolism. The LOEL in this study can be calculated as follows: TCP (1.0 × 10−7 M = 36.8 ppb), TPP (1.0 × 10−7 M = 32.6 ppb), TDCPP (1.0 × 10−6 M = 431 ppb), TCEP (1.0 × 10−5 M = 2853 ppb), and TEHP (1.0 × 10−6 M = 435 ppb). According to the previous study, TPP in coral grouper (adult) collected from Manila Bay was up to 350 ppb, and TEHP in yellow striped goat fish was up to 2000 ppb.29 Besides, in freshwater perch, TCP was also detected with the highest level of 137 ppb.9 It can be clearly concluded that the residue levels of TCP, TPP, and TEHP are higher than their corresponding LOELs, indicating that the existence of some PFRs in biota could potentially induce detrimental effects and possess ecological risks. Although TDCPP (max of 140 ppb) and TCEP (max of 160 ppb) in the fish samples collected from freshwater9 are lower than their LOELs in our study, risks would not be diminished due to its stable and lipophilic properties. On the basis of previous reports and our data, we concluded that the residues of PFRs in the biota, especially TCP, TPP, and TDCPP, may cause endocrine-disrupting effects to animals, even humans. That does again remind us that endocrine-disrupting effects, in addition to the physicochemical properties, should be considered to determine whether a PFR is a preferable emerging industrial substitute. The estrogenic effects were further compared between PFRs and BFRs (especially PBDEs with available data, because PFRs are the potential alternatives for PBDE. Some PBDEs are classified as potential endocrine-disrupting chemicals, due to their estrogenic and antiestrogenic activities. For example, Kojima et al. discovered that BDE-28, BDE-47, BDE-100, and three other metabolites showed estrogenic effects with the REC20 values which ranged from 1.9 × 10−7 M to 6.7 × 10−6 M, whereas BDE-99, BDE-153, and four other metabolites exhibited antiestrogenic activities against ERα with the RIC20 values ranging from 2.3 × 10−6 M to 8.8 × 10−6 M.30 In our study, REC20 values of TCP, TPP, and TDCPP are 8.8 × 10−7, 2.7 × 10−7, 6.4 × 10−6 M, and the RIC20 values of TCEP and

Figure 2. Proposed binding mode between hERα and TPP in molecular docking and molecular simulation study.

LigScore2, PLP1, and PMF of TPP (DOCK_Score: 35.1; LigScore2:4.06; −PLP1:80.9; −PMF: 98.0) are promising that is consistent with the tightly binding affinity between hERα with TPP. Furthermore, the TPP-bound hERα complex was refined in the molecular dynamic simulations using the Discovery Studio 2.5/Simulation module. After an initial decrease, the rootmean-square deviation (RMSD) for the backbone atoms of the complexes kept a stable trend (Figure 3), indicating that an

Figure 3. Time evolution of backbone atoms of TPP-bound hERα complex of the complexes during molecular dynamic (MD) simulation. 6999

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology



TEHP are 8.2 × 10−6 and 1.9 × 10−6 M, respectively, in the dual-luciferase reporter gene assay. Accordingly, the corresponding RIC20 or REC20 values for BFRs and PFRs were in the same order of magnitude (10−7 M −10−6 M), implying several PFRs, to some extent, may possess the equivalent estrogenic activities as PBDEs. New high-throughput screening technologies have recently been introduced to effectively and sensitively monitor the endocrine disruptors. However, discrepancies are usually observed among different assays. The subtle differences in the intensities of estrogenicity between the assays could be due to factors, such as different cell lines, protein synthesis pathways, as well as differences in cell uptake of the test chemicals. In our study, TBP possesses significant estrogenic activity in the Escreen assay, which is negative in the reporter gene assay or two-hybrid yeast assay. This discrepancy may be attributed to the different modes of action of this chemical. The reporter gene assay and yeast two-hybrid assay were constructed on the basis of the activation of ER via genomic signaling, whereas cell proliferation in the E-screen assay detects nongenomic estrogenic effects (e.g., via cell surface receptors). On the other hand, estrogenic-disrupting effects of PFRs were only mediated by ERα in the dual-luciferase reporter gene assay and the two-hybrid yeast assay, whereas other reporters excluding ER exist in MCF-7 cells lines. The estrogenic potential of chemicals can be mainly attributed to the actual concentration in cells through transmembrane and the affinity between ER and the exogenous concentration. Different chain lengths among various PFRs are one of the important factors controlling the biotransport, which further determine the actual concentration in cells. Previous studies have shown that chain lengths determine the biological effect of chemicals.31−34 In this case, the estrogenic potential of chemical may also attribute to the chain length. For example, TPP possesses the estrogenic activity, although TBEP exerts no estrogenic effect in our in vitro assays. It can be deduced that the diminished estrogenic effects for PFRs may attribute to the decreased solubility of PFRs in culture media with an increasing chain length. Conversely, chemicals with a hydroxylated form such as OHPBDE and OH-PCB have been reported to possess estrogenic effects.30,35 Molecular docking and molecular dynamic simulations indicated that PFRs possessing estrogenic activity fit well in the active pocket. PFRs with a short chain, halogen, and benzene ring as the substituent moiety tend to be estrogenic. However, the mechanisms behind these activities are not completely clear and need to be fully elucidated. The result also demonstrates that multiple models and integrated assessments should be applied to accurately evaluate endocrine disruption of chemicals. This study was aimed to qualitatively and quantitatively analyze the estrogenic and antiestrogenic effects of nine PFRs through ER using three in vitro assays. Of the nine PFRs tested, TCP, TPP, and TDCPP showed estrogenic effects, whereas TCEP and TEHP exhibited antiestrogenic effects. The potency of estrogenic effects for PFRs may partly attribute to the different interactions with ERα. Our research also suggests that comprehensive evaluations, including health and ecological assessments, are required in determining whether PFRs are a preferable emerging industrial substitute.

Article

ASSOCIATED CONTENT

S Supporting Information *

Additional information is included in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Fax: +86 571 88932 0265. Tel.: +86 571 8832 0265. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This study was supported by the Zhejiang Provincial Natural Science Foundation of China (LR12B07002) and the National Natural Science Foundation of China (21337005, 21307109, and 21377119).



REFERENCES

(1) Van der Veen, I.; de Boer, J. Phosphorus flame retardants: Properties, production, environmental occurrence, toxicity and analysis. Chemosphere 2012, 88, 1119−1153. (2) Cao, Z. G.; Yu, G.; Chen, Y. S.; Liu, C.; Liu, K.; Zhang, T. T.; Wang, B.; Deng, S. B.; Huang, J. Mechanisms influencing the BFR distribution patterns in office dust and implications for estimating human exposure. J. Hazard. Mater. 2013, 252−253, 11−18. (3) Hakk, H.; Szabo, D. T.; Huwe, J.; Diliberto, J.; Birnbaum, L. S. Novel and distinct metabolites identified following a single oral dose of alpha- or gamma-hexabromocyclododecane in mice. Environ. Sci. Technol. 2012, 46, 13494−13503. (4) Herrick, R. F.; Meeker, J. D.; Hauser, R.; Altshul, L.; Weymouth, G. A. Serum PCB levels and congener profiles among US construction workers. Environ. Health. 2007, 6, 25 DOI: 10.1186/1476-069X-6-25. (5) Decastro, B. R.; Korrick, S. A.; Spengler, J. D.; Soto, A. M. Estrogenic activity of polychlorinated biphenyls present in human tissue and the environment. Environ. Sci. Technol. 2006, 40, 2819− 2825. (6) Hloušková, V.; Lanková, D.; Kalachová, K.; Hrádková, P.; Poustka, J.; Hajšlová, J.; Pulkrabová, J. Occurrence of brominatedflame retardants and perfluoroalkyl substances in fish from the Czech aquatic ecosystem. Sci. Total Environ. 2013, 461−462, 88−98. (7) Ou, Y. X. Review on polyurethane foams flame retardants marketing. Plast. Addit. 2011, 87, 8−11 (in Chinese). (8) Meeker, J. D.; Stapleton, H. M. House dust concentrations of organophosphate flame retardants in relation to hormone levels and semen quality parameters. Environ. Health Perspect. 2010, 118, 318− 323. (9) Sundkvist, A. M.; Olofsson, U.; Haglund, P. Organophosphorus flame retardants and plasticizers in marine and fresh water biota and in human milk. J. Environ. Monit. 2010, 12, 943−951. (10) Chen, D.; Letcher, R. J.; Chu, S. G. Determination of nonhalogenated, chlorinated and brominated organophosphate flame retardants in herring gull eggs based on liquid chromatographytandem quadrupole mass specteometry. J. Chromatogr. 2012, 1220, 169−174. (11) Leonards, P.; Steindal, E. H.; van der Veen, I.; Berg, V.; Bustnes, J. O.; van Leeuwen, S. Screening of Organophosphor Flame Retardants 2010. SPFO-Report 1091/2011. TA-2786/2011. (12) World Health Organization (WHO). Report No. EHC 209: Flame Retardants: Tris-(Chloropropyl)Phosphate and Tris-(2-Chloroethyl)phosphate; WHO: Geneva, Switzerland, 1998. (13) World Health Organization (WHO). Report No. EHC 218: Flame Retardants: Tris(2-butoxyethyl) phosphate, Tris(2-Ethylhexyl) Phosphate and Tetrakis (Hydroxymethyl) Phosphonium Salts; WHO: Geneva, Switzerland, 2000. 7000

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001

Environmental Science & Technology

Article

(14) Maruska, K. P.; Fernald, R. D. Social regulation of gene expression in the hypothalamic−pituitary−gonadal axis. Physiology 2011, 26, 412−423. (15) Suzuki, G.; Tue, N. M.; Malarvannan, G.; Sudaryanto, A.; Takahashi, S.; Tanabe, S.; Sakai, S.; Brouwer, A.; Uramaru, N.; Kitamura, S.; Takigami, H. Similarities in the endocrine-disrupting potencies of indoor dust and flame retardants by using human osteosarcoma (U2OS) cell-based reporter gene assays. Environ. Sci. Technol. 2013, 47, 2898−2980. (16) Zhang, Q.; Lu, M. Y.; Wang, C.; Du, J.; Zhou, P. X.; Zhao, M. R. Characterization of estrogen receptor α activities in polychlorinated biphenyls by in vitro dual-luciferase reporter gene assay. Environ. Pollut. 2014, 189, 169−175. (17) Fisk, P. R.; Girling, A. E.; Wildey, R. J. Prioritisation of Flame Retardants for Environmental Risk Assessment; Environment Agency: Wallingford, U.K., 2003. (18) Garcia-Lopez, M.; Rodriquez, I.; Cela, R. Pressurized liquid extraction of organophosphate triesters from sediment samples using aqueous solutions. J. Chromatogr. A 2009, 1216, 6986−6993. (19) Stapleton, H. M.; Klosterhaus, S.; Eagle, S.; Fuh, J.; Meeker, J. D.; Blum, A.; Webster, T. F. Detection of organophosphate flame retardants in furniture foam and U.S. house dust. Environ. Sci. Technol. 2009, 43, 7490−7495. (20) Takigami, H.; Suzuki, G.; Hirai, Y.; Ishikawa, Y.; Sunami, M.; Sakai, S. Flame retardants in indoor dust and air of a hotel in Japan. Environ. Int. 2009, 35, 688−693. (21) Du, G. Z.; Shen, O. X.; Sun, H.; Fei, J.; Lu, C. C.; Song, L.; Xia, Y.; Wang, X. Assessing hormone receptor activities of pyrethroid insecticides and their metabolites in reporter gene assays. Toxicol. Sci. 2010, 116, 58−66. (22) Wang, C.; Zhang, Q.; Zhang, X. F.; Liu, J.; Liu, W. P. Understanding the endocrine disruption of chiral pesticides: the enantioselectivity in estrogenic activity of synthetic pyrethroids. Sci. China-Chem. 2010, 53, 1003−1009. (23) Chen, H. Y.; Xiao, J. G.; Hu, G.; Zhou, J. W.; Xiao, H.; Wang, X. R. Estrogenicity of organophosphorus and pyrethroid pesticides. J. Toxicol. Environ. Health Part A 2002, 65, 1419−1435. (24) Soto, A. M.; Sonnenschein, C.; Chung, K. L.; Fernandez, M. F.; Olea, N.; Serrano, F. O. The E-SREEN assay as a tool to identify estrogens: an update on estrogenic environmental pollutants. Environ. Health Perspect. 1995, 103, 113−122. (25) Payne, J.; Jones, C.; Lakhani, S.; Kortenkamp, A. Improving the reproducibility of the MCF-7 cell proliferation assay for the detection of xenoestrogens. Sci. Total Environ. 2000, 248, 51−62. (26) Venkatachalam, C. M.; Jiang, X.; Oldfield, T.; Waldman, M. Ligand Fit: a novel method for the shape-directed rapid docking of ligands to protein active sites. J. Mol. Graph Model 2003, 21, 289−307. (27) Villeneuve, D. L.; Blankenship, A. L.; Giesy, J. P. Derivation and application of relative potency estimates based on in vitro bioassay results. Environ. Toxicol. Chem. 2000, 19, 2835−2843. (28) Wang, Q. W.; Liang, K.; Liu, J. F.; Yang, L. H.; Guo, Y. Y.; Liu, C. S.; Zhou, B. S. Exposure of zebrafish embryos/larvae to TDCPP alters concentrations of thyroid hormones and transcriptions of genes involved in the hypothalamic−pituitary−thyroid axis. Aquatic Toxicol. 2013, 126, 207−213. (29) Kim, J. M.; Isobe, T.; Chang, K. H.; Amano, A.; Maneia, R. H.; Zamora, P. B.; Siringan, F. P.; Tanabe, S. Levels and distribution of organophosphorus flame retardants and plasticizers in fishes from Manila Bay, the Philippines. Environ. Pollut. 2011, 159, 3653−3659. (30) Kojima, H.; Takeuchi, S.; Uramaru, N.; Sugihara, K.; Yoshida, T.; Kitamura, S. Nuclear hormone receptor activity of polybrominated diphenyl ethers and their hydroxylated and methoxylated metabolites in transactivation assays using Chinese Hamster ovary cells. Environ. Health Perspect. 2009, 117, 1210−1218. (31) Wang, C.; Wang, T.; Liu, W.; Ruan, T.; Zhou, Q. F.; Liu, J. Y.; Zhang, A. Q.; Zhao, B.; Jiang, G. B. The in vitro estrogenic activities of polyfluorinated iodine alkanes. Environ. Health Perspect. 2012, 120, 119−125.

(32) Hu, W. Y.; Jones, P. D.; Upham, B. L.; Trosko, J. E.; Lau, C.; Giesy, J. P. Inhibition of gap junctional intercellular communication by perfluorinated compounds in rat liver and dolphin kidney epithelial cell lines in vitro and Sprague-Dawley rats in vivo. Toxicol. Sci. 2002, 68, 429−436. (33) Liao, C. Y.; Wang, T.; Cui, L.; Zhou, Q. F.; Duan, S. M.; Jiang, G. B. Changes in synaptic transmission, calcium current, and neurite growth by perfluorinated compounds are dependent on the chain length and functional group. Environ. Sci. Technol. 2009, 43, 2099− 2104. (34) Upham, B. L.; Deocampo, N. D.; Wurl, B.; Trosko, J. E. Inhibition of gap junctional intercellular communication by perfluorinated fatty acids is dependent on the chain length of the fluorinated tail. Int. J. Cancer 1998, 78, 491−495. (35) Nomiyama, K.; Nomura, Y.; Takahashi, T.; Uchiyama, Y.; Arizono, K.; Shinohara, R. Hydroxylated polychlorinated biphenyls (OH-PCBs) induce vitellogenin through estrogenicactivity in primarycultured hepatocytes of the Xenopus laevis. Chemosphere 2010, 28, 800−806.

7001

dx.doi.org/10.1021/es5007862 | Environ. Sci. Technol. 2014, 48, 6995−7001