Article pubs.acs.org/est
Occurrence of Thyroid Hormone Activities in Drinking Water from Eastern China: Contributions of Phthalate Esters Wei Shi,† Xinxin Hu,† Fengxian Zhang,† Guanjiu Hu,‡ Yingqun Hao,‡ Xiaowei Zhang,*,† Hongling Liu,† Si Wei,† Xinru Wang,§ John P. Giesy,†,⊥,∥,#,▼,¶ and Hongxia Yu*,† †
State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing, People’s Republic of China ‡ State Environmental Protection Key Laboratory of Monitoring and Analysis for Organic Pollutants in Surface Water, Jiangsu Provincial Environmental Monitoring Center, Nanjing, People’s Republic of China § Key Laboratory of Reproductive Medicine and Institute of Toxicology, Nanjing Medical University, Nanjing, People’s Republic of China ⊥ Department of Veterinary Biomedical Sciences and Toxicology Centre, University of Saskatchewan, Saskatoon, Saskatchewan, Canada ∥ Department of Zoology, and Center for Integrative Toxicology, Michigan State University, East Lansing, Michigan, United States # Zoology Department, College of Science, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia ▼ School of Biological Sciences, University of Hong Kong, Hong Kong, SAR, China ¶ State Key Laboratory of Marine Environmental Science, College of Oceanography and Environmental Science, Xiamen University, Xiamen 361005, China S Supporting Information *
ABSTRACT: Thyroid hormone is essential for the development of humans. However, some synthetic chemicals with thyroid disrupting potentials are detectable in drinking water. This study investigated the presence of thyroid active chemicals and their toxicity potential in drinking water from five cities in eastern China by use of an in vitro CV-1 cellbased reporter gene assay. Waters were examined from several phases of drinking water processing, including source water, finished water from waterworks, tap water, and boiled tap water. To identify the responsible compounds, concentrations and toxic equivalents of a list of phthalate esters were quantitatively determined. None of the extracts exhibited thyroid receptor (TR) agonist activity. Most of the water samples exhibited TR antagonistic activities. None of the boiled water displayed the TR antagonistic activity. Dibutyl phthalate accounted for 84.0−98.1% of the antagonist equivalents in water sources, while diisobutyl phthalate, di-n-octyl phthalate and di-2-ethylhexyl phthalate also contributed. Approximately 90% of phthalate esters and TR antagonistic activities were removable by waterworks treatment processes, including filtration, coagulation, aerobic biodegradation, chlorination, and ozonation. Boiling water effectively removed phthalate esters from tap water. Thus, this process was recommended to local residents to reduce certain potential thyroid related risks through drinking water.
■
maturation and development of the brain.5 Disruption of TH homeostasis during development of the central nervous system of children might cause irreversible mental retardation and neurological deficits and might affect behavior.4 Anti/thyroid hormone effects have been observed in industrial effluents, sediment extracts, water sources in eastern
INTRODUCTION Increasing attention has been given to the effects of environmental contaminants that can interfere with the endocrine systems of humans or wildlife.1 Most of the efforts have especially been focused on androgen and estrogen homeostasis.2 Limited information is available regarding disruption of thyroid hormone (TH) function by environmental contaminants.3 Thyroid hormone is important for normal development of the brain in higher vertebrates and postembryonic development of lower vertebrates.4 Thyroid hormone regulates genes, such as TSH-β and ChAT, that are involved in © 2011 American Chemical Society
Received: Revised: Accepted: Published: 1811
July 29, 2011 December 8, 2011 December 15, 2011 December 15, 2011 dx.doi.org/10.1021/es202625r | Environ. Sci. Technol. 2012, 46, 1811−1818
Environmental Science & Technology
Article
China, and even in finished drinking water in Beijing.6,7 Thyroid-disrupting activities that pose a risk to human health through drinking water and the responsible contaminants are of concern.8 TH function can be disrupted by synthetic chemicals from agriculture and industry, such as organochlorine (OC) pesticides, 4-aminophenol, and phthalate esters.9,10 These chemicals have been reported to interact with the thyroid receptor (TR) by inhibiting binding to its endogenous ligands or by providing additional ligands that can bind to the TR.11,7,12 Previously, concentrations of OC pesticides, polychlorinated biphenyls, and some phenols have been determined to be relatively small in drinking water in eastern China, but concentrations of phthalate esters were 100- to 1000-fold greater than the other chemicals.6 Phthalates are commonly used in a variety of products, including building materials, cables, wires, food packaging, toys, medical devices, bags, skin care products, clothing, insect repellent, and medication coating.13 There are several different phthalates, with di-2-ethylhexyl phthalate (DEHP), dibutyl phthalate (DNBP), diisobutyl phthalate (DIBP), and din-octyl phthalate (DNOP) being the most commonly produced phthalates (Table 1). DEHP and DNBP are commonly used
assay was used to determine anti/thyroid hormone effects in source water, finished water from waterworks, tap water, and the related boiled water in five cities in eastern China (see also Figure S1, Supporting Information). The three objectives of the study were to (1) detect TR agonist and antagonist activities in water during treatment, including source water, finished water from waterworks, tap water, and the related boiled water; (2) identify the compounds responsible for thyroid-active potency of extracts by use of a combination of instrumental analysis and bioassays; and (3) examine the potential adverse effects posed to the residents through drinking water.
■
EXPERIMENTAL SECTION Chemicals and Materials. All phthalate esters were purchased from Labor Dr. Ehrenstorfer-Schäfers (Augsburg, Germany). Purity and abbreviations of individual chemicals are listed in Table 1. L-3,5′,3′-triiodothyronine (T3) was obtained from Sigma Chemical Co., St. Louis, MO, USA. Compounds were dissolved in dimethylsulfoxide (DMSO, BDH Laboratory Supplies, UK) and were diluted with appropriate culture medium before use to give less than 0.5% (v/v) solvent. Sampling Locations. Changzhou (CZ), Suzhou (SZ), Wuxi (WX), Xuzhou (XZ), and Yancheng (YC) are five main cities in the Yangtze River Delta which get source water from the Yangtze River, Eastern Taihu Lake, Northern Taihu Lake, groundwater, and Huaihe River, respectively. In August 2010, 20 L (10 L for bioassay and 10 L for chemical analysis) of water was collected for each sample in the water treatment processes, including source water, finished water from waterworks, tap water, and the related boiled water from each of the 5 cities (see also Figure S1, Supporting Information). Sample Preparation and Determination. Samples and procedure blanks (Milli-Q water) were extracted by use of a previously published protocol with modification.6 Samples were not filtered. Solid phase extraction (SPE) was performed using 500-mg Oasis HLB cartridges (Waters, USA). Cartridges were activated and conditioned with 10 mL of high-purity hexane, dichloromethane, acetone, methanol, and Milli-Q water sequentially. Water was extracted under vacuum at a flow rate of 5−8 mL/min. Samples were separated into two aliquots and stored in brown glass bottles for respective instrumental and biological analysis. Approximately 2 L of sample was passed through each column to avoid over filtration. Then the column was dried completely under a gentle stream of nitrogen gas (99.999% pure). Analytes were eluted with 10 mL of hexane, 10 mL of hexane/dichloromethane (1:1), followed by 10 mL of acetone/ methanol (1:1). SPE extracts were filtered through anhydrous sodium sulfate to remove water. Samples were dehydrated and reduced to dryness under gentle nitrogen flow and reconstituted in 0.1 mL of dichloromethane for quantification. Samples used in bioassays were prepared by reduction to dryness under gentle nitrogen flow and reconstituted in 0.2 mL of dimethyl sulfoxide (DMSO). Extracts in DMSO were diluted 12.5-, 25-, 50-, 100-, and 200-fold relative to the original concentrations in water for use in bioassays. The laboratory blank consisted of Milli-Q water to exclude endocrine disrupting chemicals during the working procedure. The external standard was Milli-Q water spiked with each target analyte, and analyzed using the complete procedure. Plasticizers were quantified using a Thermo TSQ Quantum Discovery triple-quadrupole mass spectrometer (San Jose, CA, USA) in multiple-reaction monitoring (SRM) mode. The precision of
Table 1. CAS Number and Purities of Tested Chemicals chemical
abbreviation
CAS No.
purity (%)
dibutyl phthalate di-2-ethylhexyl phthalate dimethyl phthalate diethyl phthalate benzyl butyl phthalate diisodecyl phthalate bis(2-ethylhexyl) adipate di-n-octyl phthalate diisononyl phthalate diisobutyl phthalate
DNBP DEHP DMP DEP BBP DIDP DEHA DNOP DINP DIBP
84-74-2 117-81-7 131-11-3 84-66-2 85-68-7 68515-48-0 103-23-1 117-84-0 28553-12-0 84-69-5
99.0 99.0 99.5 99.5 99.5 99.5 99.5 99.5 99.0 99.0
in most polyvinyl chloride (PVC) consumer products (toys, shower curtains, etc.).14 DNBP is also used in some personal care products.15 DIBP is a special plasticizer, used as a substitute for DNBP.16 DNOP is commonly used in the manufacture of flexible vinyl.17 Detectable concentrations of DNBP and DEHP as great as 2.9 × 101 μg/L were found in finished water from waterworks in Beijing.7 DNBP has also been found in finished drinking water in Chongqing and Hangzhou, at concentrations as great as 1.7 × 101 and 7.6 × 101 μg/L, respectively.7,18 DNBP has been demonstrated to be a TR antagonist.6 Activated carbon has been reported to be effective for removing organic contaminants,19 but the use of activated carbon is not common in drinking water treatment in eastern China. Thus, more attention should be given to removal of phthalates and the related toxicity during water treatment or locally at the tap. Promoter−reporter gene assays that were rapid, sensitive, and reproducible in in vitro methods were employed in this study to assess the ligand−receptor interaction by receptor agonists and antagonists. Previously, green monkey kidney fibroblast cells (CV-1) have been used as the basis of a TH reporter gene assay for the screening of chemicals with TR ant/agonistic properties.20 This system has been used to characterize both anti/ thyroid hormone effects of water extracts and some related chemicals, including bisphenol A (BPA), tetrachlorobisphenol A, carbaryl, 1-naphthol, 2-naphthol, DNBP, mono-n-butyl phthalate (MBP), and DEHP.20 In this study, the CV-1 TH 1812
dx.doi.org/10.1021/es202625r | Environ. Sci. Technol. 2012, 46, 1811−1818
Environmental Science & Technology
Article
Relative potencies (REPs-DNBP) for antagonist activity of the standard compounds were calculated by dividing the EC20 of DNBP by the EC20 of the test compound. Activities of DNBP equivalents (DNBP-EQs) were calculated by summing the product of concentrations of individual congeners multiplied by their respective relative potencies (REPs-DNBP) (see Supporting Information). Values reported were mean ± SD (n = 3). In the bioassays triplicate wells were done for each treatment. Data were analyzed by use of one-way ANOVA, followed by Duncan’s multiple comparisons test when appropriate using SPSS statistical software (version 11, SPSS Inc., Chicago, IL, USA). Curvefitting analyses were carried out with GraphPad 5.4 (San Diego, CA, USA). The level of significance was set at *p < 0.05 and **p < 0.01. For agonistic activities, treatments were compared to the vehicle control groups. For antagonistic activities, treatments were compared to 5.0 × 10−9 mol/L T3 positive control groups.
the method quantified by relative standard deviation (RSD), was determined by replicate extractions (n = 3) of a single sample. A signal-to-noise ratio of 3:l was used as the criteria for the analytical limit of detection (LOD). Limit of quantification (LOQ) was defined as 10 times the noise level. Plasmids and Cell Culture Condition. Plasmids pGal4L-TRβ and pUAS-tk-Luc were kindly provided by Dr. Ronald M. Evans (Gene Expression Laboratory, Howard Hughes Medical Institute, San Diego, CA, USA). The ligand binding domain (LBD) of TRβ was fused to the DNA binding domain (DBD) of Gal4 in plasmid pGal4-L-TRβ. Green monkey kidney fibroblast (CV-1) cell line without endogenous receptors (TR) was purchased from the Institute of Biochemistry and Cell Biology in Shanghai, Chinese Academy of Science. CV-1 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Gibco, Invitrogen Corporation, Carlsbad, CA, USA), 100 U/mL penicillin (Sigma),and 100 μg/mL streptomycin (Sigma) at 37 °C in an atmosphere containing 5% CO2. Cytotoxicity and Reporter Gene Assay. The cytotoxicity induced by the tested chemicals and extracts was assessed using the MTT assay according to the protocol described previously.20 For the reporter gene assay, cells were plated and transfected as described previously.6 After an additional 12 h of incubation, the cells were exposed to various concentrations of chemicals and extracts for 24 h. DMSO concentrations in wells never exceed 0.5% (v/v) (see Supporting Information). Data Analysis. The antagonist equivalency of the extracts (Ant-TR-EQ) was derived by comparing the observed activity to the concentration of a known TR antagonist DNBP (standard) that produced an equivalent reduction in TR antagonist activity.21−23 The assumptions of equal potency and efficacy were addressed as proposed by Villeneuve et al.23 TR antagonist activity was reported as the concentration of DNBP divided by the sample concentration factor that produced an equivalent (20%) depression in the bioassay response to 5.0 × 10−9 mol T3/L. Thyroid receptor antagonistic activities of tested chemicals were reported as the EC20 (20% relative inhibitory concentration),24,25 which refers to the concentration at which the tested chemicals showing a 20% reduction in the activity of 5.0 × 10−9 mol/L T3 via TRβ. By using these magnitudes of response it was possible to avoid artifacts and biases associated with violation of the assumptions necessary to determine relative potencies and determine concentrations of equivalents in extracts of water.
■
RESULTS AND DISCUSSION Cell Viability and Assay Validation. None of the tested concentrations of individual chemicals or extracts affected viability or proliferation of CV-1 cells alone or in the presence of 5.0 × 10−9 mol T3/L. No cytotoxic effects of solvent or water extracts were observed by microscopic examination throughout the transfection assay. The natural TR ligand T3 induced luciferase activity in a concentration-dependent manner in the CV-1 reporter assay (see also Figure S2, Supporting Information). T3 induced luciferase activity in the range of 1.0 × 10−10 mol/L to 1.0 × 10−6 mol/L, with maximal induction of 240-fold relative to that of the vehicle control achieved at a concentration of 1.0 × 10−6 mol T3/L. The positive control, DNBP caused a typical dose− response curve in 5.0 × 10−9 mol T3/L (Figure 1). No significant induction of luciferase was observed in any of the solvent controls (data not shown). Recoveries of plasticizers in instrumental analysis were between 90% and 120%. RSDs ranged from 3% to 12%. TR Antagonistic Activity. In the presence of 5.0 × 10−9 mol T3/L, DIDP caused no effect and DEHA caused only a slight increase in expression of the reporter gene. However, all the other phthalate esters exhibited greater TR antagonistic potencies that resulted in concentration-dependent expression of the reporter gene under control of the TR (Figure 1). The antagonistic potency (EC20) reached a maximum from 6.9 × 10−7 to 3.6 × 10−4 mol/L
Figure 1. Concentration-dependent thyroid receptor antagonist activities of phthalates as measured by the CV-1 cell line TR reporter gene assay. Results are expressed as mean ± SD (n = 3). Significant differences relative to control are indicated by asterisks (*p < 0.05 and **p < 0.01). The dashed line depicts the control. 1813
dx.doi.org/10.1021/es202625r | Environ. Sci. Technol. 2012, 46, 1811−1818
Environmental Science & Technology
Article
extrapolation estimates of REP for BBP and DMP. The detected TR antagonistic potency and the LOEC values of DIDP, DEHA, DOP, and DINP in the present study were similar to those reported by Ghisari et al.26 It has been previously reported that agonistic compounds may appear as antagonistic once in presence of a natural biological nuclear receptor depending on concentration of ligand, ligand binding affinity, and the presence of competing natural ligands.32,33 However the reason for the mixed agonist/antagonist activities remains inconclusive. Mixed ligand dimers may be required for antagonism, whereas sameligand dimers may promote gene activation, depending on the ligand bound receptor induced conformational changes.34−36 None of the tested extracts displayed any TR agonistic activity. However, 4 of the 5 tested source water extracts exhibited TR antagonistic potencies in a concentration-dependent manner (Figure 2). Neither the procedure blank nor XZ city source water caused TR antagonistic responses. We recalculated the relative antagonistic potency to DNBP, because DNBP was identified to have the most potent TR antagonist activity among the commonly used synthetic chemicals.7 Moreover, DNBP has been studied a lot for its potential toxicity induced by its thyroid hormone disrupting effects. The Ant-TR-EQs for the water sources ranged from 3.2 to 8.2 μg DNBP-EQs/L (Table 3). Previous studies have studied the effect of DNBP on T3dependent activation of TRβ gene in T3-induced metamorphosing tadpoles and the TR antagonist response was detected at 1.1 × 103 μg DNBP/L.37 Although this concentration was more than 100-fold greater than the TR antagonist equivalents detected in this study, thyroid hormone modulating potential was observed in source waters. Significant TR antagonistic activities in waterworks and the related tap water from SZ and WX were detected at the greatest concentration tested (200 times the original concentration). Because of the limited sample volumes, extracts could not be tested at more dilutions. In these cases, the concentration of DNBP which caused the same response as the greatest concentration tested was divided by the enrichment factor 200 to estimate the single point Ant-TR-EQs. For the other samples that exhibited significant potency but did not reach a maximal/ minimal response, extrapolation was used to generate Ant-TREQs from the dose−response curves.23 Almost 90% of TR antagonistic activities were removable by treatment in waterworks
(Table 2). The measured TR antagonistic potencies of DNBP and DEHP observed in the present study were similar to those reported Table 2. TR Antagonistic Potency (EC20) and the Related Potencies (REPs-DNBP) of Phthalate Esters TR antagonistic 95% confidence limits (mol/L) DMP DEP DIBP DNBP BBP DEHA DEHP DNOP DIDP DINP a
EC20 (mol/L)
REPs-DNBP
lower
upper
1.6 × 10−4 5.0 × 10−5 1.5 × 10−5 6.9 × 10−7 3.6 × 10−4 9.5 × 10−5b 3.3 × 10−5 9.1 × 10−6 1.4 × 10−5
6.22 × 10−3 1.73 × 10−2 4.48 × 10−2 1.00 1.71 × 10−3 1.48 × 10−2 5.42 × 10−2 3.29 × 10−2
-a 3.0 × 10−5 1.0 × 10−5 3.4 × 10−7 2.3 × 10−6b 1.6 × 10−5 9.0 × 10−6 8.5 × 10−6
9.7 × 10−5 2.2 × 10−5 1.3 × 10−6 3.4 × 10−5b 8.0 × 10−5 1.1 × 10−5 1.8 × 10−5
- No effect. bSynergetic effects.
previously.7,20,26 The least concentrations at which significant effects (p < 0.05) were detected (LOEC) were similar to those reported by Ghisari et al. based on GH3 cell growth assays,26 but were greater than those based on yeast assays.7,27 The sensitivity of the yeast system depends on the number of receptors, the response elements and the choice of reporter gene used.28,29 The difference might be due to various factors, such as different cell lines, protein synthesis pathways, receptor affinities, and also differences in TR expression level and cell uptake of the tested chemicals.30 In previous studies DEP exhibited no TR antagonism,7 which is quite different from the results in the present study. This difference might be due to the impermeability of the cell wall of yeast cells to the test substances.24,31 Yeast does not contain the complex generegulating network that is normally involved in responses of mammalian cells to hormones, which can result in different results.30 Yeast was reported to have some degree of metabolic competence, and it is possible that the metabolites of some chemicals would not produce these effects rather than the parent compound.29 BBP and DMP exhibited weak antagonist activities without reaching 20% potency. For further use of the determined relative potencies, nonlinear regression methods were used for
Figure 2. Concentration-dependent TR antagonist activities in extracts of water as measured by the CV-1 cell line TR reporter gene assay. Extracts of water in cell culture media were tested at 12.5, 25, 50, 100, and 200 times of the original concentration in water samples. Cells were exposed to extracts in parallel with 5 nmol/L T3. The TR antagonist activity was expressed as relative expression versus the untreated cells (control) (mean ± SD, *p < 0.05 and **p < 0.01). 1814
dx.doi.org/10.1021/es202625r | Environ. Sci. Technol. 2012, 46, 1811−1818
Environmental Science & Technology
Article
10−2 to 8.0 μg DNBP/L. When concentrations of DNBP-EQs measured by the bioassay were compared with those calculated from the concentrations of phthalates, 86−99% of the total Ant-TR-EQ in water sources was contributed by phthalate esters (Figure 3). With the greatest concentration and the strong TR antagonist potency, DNBP accounted for 84.0− 98.1% of the Ant-TR-EQs. DIBP, DNOP, and DEHP accounted for 0.1−1.2%, 0−0.7%, and 0.1−0.3% of the AntTR-EQs, respectively. The contribution of DEHP was in agreement with the results of a previous study conducted in Beijing, in Northern China.7 It could be speculated that DNBP might be the major TR antagonist potency in water sources, while DIBP, DNOP, and DEHP also contributed. Concentrations of DNBP-EQs in finished drinking water from waterworks and tap water ranged from less than 1.0 × 10−2 to 9.4 × 10−1 μg DNBP/L. Phthalate esters contributed almost all of the TR antagonist potency. Greater than 90% of the total concentrations of Ant-TR-EQs were contributed by DNBP in finished drinking water and tap water. Humans and animals are exposed to phthalate esters from surface water, soil, air and even in bottled water and water from drinking fountains.41−44 In this study, waterworks were found to effectively remove most of the phthalate esters, although detectable concentrations of these chemicals still existed in the tap water, but none of the target chemicals were detected in boiled water. As semivolatile organic compounds (SVOC), phthalate esters result in significant volatilization during the boiling.45 In China, boiling or heating is the most widely used means of treating water. Boiling is effective at removing phthalate esters, and the risk posed by these chemicals to residents through drinking is thus small. However, all available data considering these phthalate esters showed that they are rapidly metabolized and their metabolites were considered to be more noxious than the parent products.20,46,47 Recently, metabolites of phthalates have been detected in marine biota, consumer milk, and even in human milk.48−50 The health risk posed by the long-term human exposure to low dose of phthalate ethers and their metabolites still warrants further careful evaluation. Comparisons of the toxicity equivalents from instrumental analysis and bioassays suggested that DNBP could play a major role in the TR antagonistic activities in water sources, while diisobutyl phthalate (DIBP), di-n-octyl phthalate (DNOP), and di-2-ethylhexyl phthalate (DEHP) also showed minor contributions. The waterworks treatment processes including filtration, coagulation, aerobic biodegradation, chlorination, and ozonation seems effective for the removal of phthalate esters and the related TR suppression. The boiling process is effective enough to remove the phthalate esters in tap water, and the risk posed by these chemicals to residents through drinking is small. However, the volatilized phthalate esters in indoor air might enhance the risk through respiratory and skin exposure. More attentions should be paid to the high concentrations of DNBP, DIBP, DNOP, and DEHP in water sources and the related surface water.
Table 3. Thyroid Receptor Antagonist Equivalents Derived from Instrumental Analysis (DNBP-EQs) and Reporter Gene Assays (Ant-TR-EQs) for the Water Samples (μg/L) 95% confidence limits for Ant-TR-EQs locations CZ
SZ
WX
XZ
YC
a
source waterworks tap source waterworks tap source waterworks tap source waterworks tap source waterworks tap
DNBP-EQs 4.2 1.0 × 10−1 9.0 × 10−2 8.0 4.7 × 10−1 9.4 × 10−1 7.6 3.2 × 10−1 4.3 × 10−1 4.0 × 10−2
