Influence of household cleaning practices on the magnitude and variability of urinary monochlorinated bisphenol A

Science of the Total Environment 490 (2014) 254–261

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Influence of household cleaning practices on the magnitude and variability of urinary monochlorinated bisphenol A H. Kalyvas a, S.S. Andra a,b, P. Charisiadis a, C. Karaolis a, K.C. Makris a,⁎ a b

Cyprus International Institute for Environmental and Public Health in association with Harvard School of Public Health, Cyprus University of Technology, Limassol, Cyprus Harvard–Cyprus Program, Department of Environmental Health, Harvard School of Public Health, Boston, MA 02115, USA

H I G H L I G H T S • Urinary levels of monochlorinated BPA and trihalomethanes were positively correlated • Exposure to monochlorinated BPA was influenced by domestic cleaning activities • BPA in household products may act as a precursor for its chlorinated derivatives

a r t i c l e

i n f o

Article history: Received 11 March 2014 Received in revised form 12 April 2014 Accepted 18 April 2014 Available online xxxx Editor: D. Barcelo Keywords: Bisphenol A Chlorinated BPA Trihalomethanes Drinking water Biomarkers of exposure Disinfection

a b s t r a c t Low-dose health effects of BPA have not been adequately explored in the presence of BPA metabolites of chlorinated structure that may exert larger estrogenic effects than those of their parent compound. We hypothesized that chlorine-containing cleaning products used in household cleaning activities could modify the magnitude of total urinary BPA concentration measurements via the production of chlorinated BPA (ClBPA) derivatives. Our objective was to investigate the influence of typical household cleaning activities (dishwashing, toilet cleaning, mopping, laundry, etc.) on the magnitude and variability of urinary total BPA and mono-ClBPA levels in the general adult population. A cross-sectional study (n = 224) included an adult (≥18 years) pool of participants from the general population of Nicosia, Cyprus. First morning urine voids were collected, and administered questionnaires included items about household cleaning habits, demographics, drinking water consumption rates and water source/usage patterns. Urinary concentrations of total BPA (range: 0.2–82 μg L−1), mono-ClBPA (16– 340 ng L−1), and total trihalomethanes (0.1–5.0 μg L−1) were measured using gas chromatography coupled with triple quadrupole mass spectrometry and large volume injection. Linear multiple regression analysis revealed that dishwashing along with age and gender (females) were able to predict urinary mono-ClBPA levels (ng g−1), even after adjusting for covariates; this was not the case for urinary total BPA levels (ng g−1). Significant (p b 0.001) association was observed between urinary mono-ClBPA and THM levels, underlying the important role of disinfectant (chlorine) in promoting formation of both ClBPA and THM. Urinary mono-ClBPA levels were measured for the first time using an appreciable sample size, highlighting the co-occurring patterns of both total BPA and mono-ClBPA. Epidemiological studies and probabilistic BPA risk assessment exercises should consider assessing daily intake estimates for chlorinated BPA compounds, as well. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Bisphenol A (BPA), 2,2-bis(4-hydroxyphenyl)propane, is widely used as a monomer in polycarbonate plastics and epoxy resins, being Abbreviations: BPA, bisphenol A; DBP, disinfection byproducts; EEU, exposure equivalence unit; Mono-ClBPA, monochlorinated bisphenol A; TTHM, total trihalomethanes; UDWDS, urban drinking water distribution system. ⁎ Corresponding author at: Water and Health Laboratory, Cyprus International Institute for Environmental and Public Health in association with Harvard School of Public Health, Cyprus University of Technology, Irenes 95, Limassol 3041, Cyprus. Tel.: + 357 25002398; fax: +357 25002676. E-mail address: [email protected] (K.C. Makris).

http://dx.doi.org/10.1016/j.scitotenv.2014.04.072 0048-9697/© 2014 Elsevier B.V. All rights reserved.

one of the world's highest production volume chemicals and as such, BPA occurrence in the environment and consumer products is ubiquitous (Staples et al., 1998; Kang et al., 2006; Vandenberg et al., 2007, 2010). Dietary (food and water) items packaged in polycarbonate plastics and/or in contact with epoxy resin coatings of food/beverage containers represent the main exposure sources of BPA (Liao and Kannan, 2013). Non-dietary BPA exposures have recently attracted attention, because certain personal care- and household-cleaning products may contain BPA, such as, bar soaps, facial/body lotions, shampoo, dishwashing and laundry detergent, and toilet bowl cleaner (Dodson et al., 2012); the BPA content of these products may range between b10 μg g−1 and ~100 μg g−1 per product. The occurrence of BPA in the aforementioned

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261

consumer products may not be obvious, because it is often not listed on the product labels, by either being (i) a necessary ingredient of product-packaging resin and material (Halden, 2010; Geens et al., 2011), (ii) a non-intentionally added substance (e.g. trace BPA levels, following its usage in the manufacturing of tetrabromobisphenol A, a widely used flame retardant in household products) (Covaci et al., 2009), and/or (iii) a product impurity (e.g. BPA as an impurity in bisphenol-A dimethacrylate used in composite resins) (Chen and Suh, 2013). Upon application, most of the aforementioned personal care or household cleaning products require their mixing with tap water that typically contains residual chlorine, which will tend to react with BPA towards the production of chlorinated BPA derivatives (ClBPAs) (Yamamoto and Yasuhara, 2002; Gallard et al., 2004; Liu et al., 2009). Reported concentrations of mono-chlorinated BPA in tap water were up to 26.7 ng L−1 versus 128 ng L−1 of BPA (Fan et al., 2013) or below detection (Dupuis et al., 2012). Among the four chlorinated forms of BPA quantified in tap water, mono-chlorinated BPA had the highest detection rate (100%) and the highest concentration (maximum: 26.7 ng L−1) compared to di- (98%, 6.3 ng L−1), tri- (60%, 7.7 ng L−1), and tetra-chlorinated BPA (50%, 4.9 ng L−1) (Fan et al., 2013). These chlorinated BPA derivatives have been detected in human urine (Liao and Kannan, 2012), adipose tissue (Fernandez et al., 2007), breast milk (Cariot et al., 2012; Migeot et al., 2013), colostrum (Migeot et al., 2013), and placental tissue (Jimenez-Diaz et al., 2010). The estrogenic activity of ClBPAs is being studied and these compounds exhibit similar activity to BPA, which depending on the receptors can be slightly lower (Molina-Molina et al., 2013; Kuruto-Niwa et al., 2002; Riu et al., 2011a,b), or higher (Terasaki et al., 2011; Riu et al., 2011a,b; Liu et al., 2005; Takemura et al., 2005; Yamauchi et al., 2003; Fukazawa et al., 2002). However, certain studies indicated that the offset of estrogenic activity of ClBPAs occurs at lower concentrations than those of BPA (Kuruto-Niwa et al., 2002; Babu et al., 2012; Viñas et al., 2013) and that biologically-relevant ClBPA concentrations triggered non-monotonic responses (Viñas et al., 2013). In addition, photodegradation of ClBPAs altered their estrogenic activity (Mutou et al., 2006; 2008; Gallart-Ayala et al., 2007; Ibuki et al., 2008), while sulfonation of ClBPAs, a metabolic process for the removal of BPA and other xenobiotics, did not eliminate their estrogenic activity, contrary to the effect of sulfonation on BPA (Riu et al., 2011a,b). Furthermore, the addition of chlorine atoms increases the lipophilicity of BPA derivatives, being evident from the higher ClBPAs to BPA concentration ratios measured in fatty tissues when compared to urine (Liao and Kannan, 2012; Fernandez et al., 2007; Cariot et al., 2012; Migeot et al., 2013; Jimenez-Diaz et al., 2010); this lipophilicity could increase dermal uptake rates of ClBPAs. Residual BPA often found in (chlorine-containing) household cleaning and personal hygiene products (Dodson et al., 2012; Liao and Kannan, 2014) could act as a precursor for ClBPA formation, when in contact with chlorinated tap water. Chlorine-containing household products often take the form of (i) cleaning products that contain sodium hypochlorite (kitchen countertop/floor/toilet cleaners, bleaching and scouring powders, stain removing sprays/gels, etc.) (Odabasi, 2008), (ii) bleach-containing laundry detergents (Nazaroff and Weschler, 2004), (iii) hypochlorite containing dishwasher detergents (Olson and Corsi, 2004), and (iv) bleached clothes and fabrics (Leri and Anthony, 2013). We hypothesized that the frequency and duration of domestic cleaning activities (dishwashing, mopping, toilet cleaning, etc.), involving the use and dilution of BPA-containing cleaning products with chlorinated tap water will dictate the magnitude and variability of exposure to ClBPAs. This hypothesis was tested in a subsample of a general population study concerning human exposures to disinfection by-products (trihalomethanes, THM), accounting for drinking water habits and household cleaning activities (Charisiadis et al., 2014). The objective of this study was to investigate whether (i) domestic activities involving

255

use of chlorine-based domestic cleaning products was proportional to ClBPA exposures (measured as urinary mono-ClBPA concentrations) and (ii) an association exists between exposures to mono-ClBPA and total THM given that they both can form in chlorinated tap water, and both are influenced by noningestion practices (dishwashing, toilet cleaning, mopping, laundry, showering, bathtub use, pool swimming, etc.).

2. Methods 2.1. Study population and experimental design The detailed description of the study design, questionnaire and data collection methods has been previously reported (Charisiadis et al., 2014) and summarized here. A cross-sectional study was conducted in Nicosia, Cyprus. Trained interviewers using door-to-door contacts and flyers given at randomly selected households recruited residents and obtained their written consent. A total of 341 adult residents (≥18 years) from 193 houses participated in this study, which focused on understanding sources and routes to THM and ClBPA exposure. A total of 15 subjects were excluded from further analyses because of: i) insufficient spot urine volume for GC– MS/MS analysis (8 participants); ii) very low urinary creatinine with a mean (SD) of 0.33 (0.44) g L−1 (6 participants) compared with the rest subjects (mean 1.24 (0.75) g L−1); and iii) an incomplete questionnaire (one participant). The chosen subset of 224 adults (≥ 18 y) was randomly selected within the 326 adults mentioned above from a recent cross-sectional study (Charisiadis et al., 2014) and it was based on sample-size calculation using JMP 7.0 software (SAS Inc., NC, USA) that yields a statistical power of 95% to differentiate participants with high and low exposures to THM (a major chemical class of DBP). No general population data exist for chlorinated BPA derivatives that could be used in our sample size calculations, thus, we used data on urinary THM. The rationale was that urinary THM concentrations were a surrogate of exposure to chlorine-based compounds in the indoor environment, and thus they could share co-occurring patterns with mono-ClBPA. The study was approved by the National Bioethics Committee of Cyprus and it was conducted in accordance with the ethical principles for medical research involving human subjects.

2.2. Questionnaire and data collection Participants were interviewed about demographics, habitual and life style factors with respect to ingestion and non-ingestion exposures to chlorine-containing domestic products. Non-ingestion exposure habits, such as frequency and duration of showering and frequency and duration of specific cleaning activities, whether cleaning the house personally or doing laundry personally, were recorded. Survey questions ‘laundry done personally’ and ‘domestic cleaning done personally’ refer to all associated activities that involve use of unprotected hands (no gloves) and/or those that enhance dermal contact of household chemicals during water collection in containers, use of laundry detergent (powder or liquid), mixing, soaking, rinsing and water draining before putting the clothes in the washer, followed by spreading clothes to dry under the sun (because dryers are in rare use in the study location), and cleaning of any sort. Information on all the household laundry and cleaning products was recorded. Questionnaire responses referring to ingestion and non-ingestion exposures were normalized using THM exposure equivalent units (Weisel and Jo, 1996). An exposure equivalence unit for THM was set equal to 1 L day− 1 tap water consumption; 15 min day−1 dish washing, mopping, toilet or other cleaning activities; 5 min day−1 shower or bathtub time; and 5 min day−1 swimming in a pool (Rivera-Nunez et al., 2012).

256

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261

2.3. Urine analyses First morning void urine samples were collected in 60 mL polypropylene vials. The samples were immediately transported to the lab and stored at−80 °C until analyzed. 2.3.1. Trihalomethane analysis A modified USEPA Method 551.1-1 was used for THM analysis in water and urine sample (US EPA, 1995). Briefly, a liquid–liquid extraction method was adopted by mixing 15 mL urine sample with 2 mL of t-butyl methyl ether, adding 6.0 g of sodium sulfate to saturate the water phase, and gently shaking for 5 min at 100 rpm. 0.5 mL of the organic phase was transferred in a GC auto-sampler vial, containing internal standard solution at a final concentration of 10 μg L−1. The detection limits (LOD) were 94 ng L−1, 46 ng L−1, 31 ng L−1, and 40 ng L−1 for chloroform, dichlorobromomethane, chlorodibromomethane and bromoform, respectively. The sum of the four urinary THM analyte concentrations was denoted as urinary total THM. The quality control consisted of pooled urine samples with negligible THM measured concentrations, fortified at a final concentration of 700 ng L−1 as a THM commercial mixture. The recoveries were 103%, 100%, 98% and 92% for chloroform, dichlorobromomethane, chlorodibromomethane and bromoform, respectively, with a mean surrogate recovery of 82%. Intra- and inter-day variabilities of analytical measurements were always b3.5%, while the average (for the 700 and 1000 ng L−1) total THM recoveries in pooled urine matrix were 95–101% with an average relative standard deviation of 4% (Charisiadis and Makris, 2014; Charisiadis et al., 2014). 2.3.2. BPA, mono-ClBPA, and creatinine analyses Total BPA (unconjugated plus conjugated BPA) and its monochlorinated BPA derivative were monitored using GC–MS/MS. Commercially available BPA (99% Sigma-Aldrich) was used for analysis while mono-ClBPA was synthesized according to published procedures (Fukazawa et al., 2001). Urinary samples (4 mL) spiked with d16-BPA at a final concentration of 300 ng L−1 as a surrogate standard were mixed with 4 mL of sodium acetate buffer solution (0.5 M, pH 4.75) and enzymatically hydrolyzed with glucuronidase. Deuterated (d16)-BPA was used at a concentration of 300 ng L−1 to assess the analyte extraction efficiency and extent of derivatization. The analytes were extracted with 3 mL of ethyl acetate/hexane 1:4 mixture, and esterified with trifluoroacetic anhydride. The solvent was evaporated and reconstituted with 0.5 mL CH2Cl2 in the presence of decafluorobiphenyl (as an internal standard to monitor inter-sample and inter-day variability) at a final concentration of 8 μg L−1, and analyzed by GC–MS/MS using a 20 μL large volume injection system. In order to minimize external BPA contamination, all solvents were distilled prior to use and only furnace-dried glass equipment (vials, syringes and pipettes) were used. Detection limits were at 95 ng L−1 for BPA, and 32 ng∙L−1 for monoClBPA. Quality control consisted of pooled urinary sample fortified with BPA and mono-ClBPA in the range of 100 to 1500 ng L−1 with satisfactory recovery (always N80%). The average surrogate recovery of all samples was 98%. Inter and intra-day variabilities for the measurements were at 5% as calculated by the surrogate response. Urinary creatinine concentrations were determined with the picric acid-based spectrophotometric method, where creatinine and alkaline picric acid reagent produce a red/orange complex after 45 minute reaction at 36 °C (Jaffe method; Clarke, 1961).

used. Categorical variables were presented as frequencies (percentages), whereas continuous variables were given as mean (standard deviation, SD). The chi-square test was used to check for gender differences in relation to different participants' qualitative characteristics. Furthermore, t-tests and analysis of variance were carried out to check for differences in the mean urinary total THM in different groups and Tukey's HSD test was used for adjustment for multiple comparisons. Bivariate linear regression analyses were performed to identify the significant factors influencing ingestion and non-ingestion exposures to creatinine-unadjusted and adjusted BPA, mono-Cl-BPA, and TTHM. The variance inflation factor was minimal (VIF ~ 1.0) among the three urinary analytes indicating lack of collinearity in a multiple regression analysis. The variables that were either significantly correlated to the urinary mono-Cl-BPA and those of exposure relevance to either BPA or TTHM were included as confounders in multiple linear regression models. The significant predictors used were gender, age (y), THM levels in tap water (μg L−1), whether domestic cleaning and laundry was done personally (yes or no), and exposure equivalence units for dishwashing, mopping, toilet and other household cleaning. Statistical analyses were performed using JMP version 10.0 (SAS Inc., USA) with α = 0.05 set as critical value for significance. Samples with analyte levels below the respective LOD were assigned half of their respective values. 3. Results 3.1. Demographics and exposure characteristics The arithmetic mean (standard deviation) age and body mass index of all participants (n = 224) were 51 (17) years and 26 (5) kg m−2; while it was 50 (17) years and 26 (6) kg m−2 for females (n = 129) and 52 (17) years and 26 (6) kg m−2 for males (n = 95), respectively (Table 1). Among the participants, 58% were females, 56% were overweight (25–30 kg m−2) or obese (N30 kg m−2), 74% had education up to secondary level, and 79% were married. Total THM concentrations in tap water (mean value of 67 μg L−1) followed a normal distribution, while this was not the case for either exposure variables to chlorine-relevant compounds in a common household environment, or for the urinary concentrations of mono-Cl-BPA and THM. The average (SD) exposure variables for chlorine-relevant compound usage from ingestion and non-ingestion sources were 0.7 (1.0) L day−1 of water consumption, 25 (44) min day−1 of personal hygiene activities in contact with water, and 22 (28) min day−1 of all domestic cleaning activities in contact with water and household cleaning products (Table 1); the corresponding exposure variables in females versus males were 0.6 (0.8) vs. 0.8 (1.1) L day−1, 26 (41) vs. 23 (47) min day−1, and 36 (30) vs. 3 (9) min day−1, respectively. The average urinary concentration of BPA, mono-ClBPA, and total THM in all the study participants was 3749 (7631) ng L−1 [creatinineadjusted: 2848 (4383) ng g− 1]; 83 (49) ng L− 1 [115 (142) ng g− 1]; and 707 (411) ng L− 1 [944 (1094) ng g− 1], respectively (Table 1). The corresponding urinary concentrations in females versus males were 2337 (2014) vs. 5667 (11,233) ng L − 1 [2446 (1936) vs. 3393 (6321) ng g − 1 ] for BPA; 84 (45) vs. 83 (55) ng L − 1 [137 (142) vs. 86 (136) ng g − 1 ] for mono-ClBPA; and 740 (475) vs. 662 (299) ng L − 1 [1207 (1318) vs. 586 (496) ng g − 1 ] for total THM, respectively. 3.2. Correlation between exposure variables and urinary biomarkers

2.4. Statistical analyses The distributions of creatinine-adjusted and unadjusted urinary BPA, ClBPA and total THM concentrations were right-skewed, and thus, were log-transformed. Associations between normally distributed variables were assessed using Pearson's correlation coefficients and for the rest a Spearman's correlation coefficient (non-parametric analysis) was

Urinary mono-ClBPA was significantly correlated with age (rS = 0.39, p b 0.001), and daily spent time performing hand dishwashing (rS = 0.33, p b 0.001), mopping (rS = 0.27, p b 0.001), and toilet cleaning (rS = 0.28, p b 0.001). Similar correlations were found between urinary total THM and age (rS = 0.31, p b 0.001), and daily spent time performing hand dishwashing (rS = 0.25, p b 0.001),

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261

257

Table 1 Frequency distribution of personal activities of exposure to chlorine-relevance chemicals in a common household environment, and the urinary concentrations of BPA, monochlorinated BPA, and total trihalomethanes in the study participants (n = 224). Characteristic

Mean (SD)

Demographics Age (years) BMI (kg m−2)

51 (17) 26 (5)

Variables of exposure to chlorine-relevance compounds Tap water trihalomethanes (μg L−1) Tap water ingestion (L day−1) Shower (min day−1) Bathtub (min day−1) Swim in pool (min day−1) All hygienic activities (min day−1) Hand dishwashing (min day−1) Toilet cleaning (min day−1) Mopping (min day−1) Other domestic cleaning (min day−1) All cleaning activities (min day−1) All non-ingestion exposure activities (min day−1)

67 (24) 0.7 (1.0) 15 (12) 1 (4) 9 (42) 25 (44) 8 (13) 4 (8) 8 (11) 2 (7) 22 (28) 47 (50)

Urinary biomarkers of exposure to chlorine-relevance compounds BPA (ng L−1)a 3749 (7631) 2848 (4383) BPA (ng g−1 creatinine) 83 (49) Mono-ClBPA (ng L−1)a 115 (142) Mono-ClBPA (ng g−1 creatinine) 707 (411) TTHM (ng L−1)b 944 (1094) TTHM (ng g−1 creatinine)

Percentiles Min.

10%

18 16

26 20

37 22

53 25

64 29

73 33

30 0.0 6.0 0.0 0.0 6.0 0.0 0.0 0.0 0.0 0.0 7.1

56 0.0 10 0.0 0.0 10 0.0 0.0 0.0 0.0 0.0 13

70 0.3 12 0.0 0.0 14 2.0 0.0 2.0 0.0 9.5 32

81 1.0 19 0.0 0.0 21 12 7.0 15 0.0 37 61

94 1.9 30 0.0 13 43 20 10 20 2.0 63 96

3 0.0 3.0 0.0 0.0 3.0 0.0 0.0 0.0 0.0 0.0 2.9

175 390 16 16 137 79

523 777 33 21 340 214

25%

1012 1168 54 37 519 333

50%

2098 1820 77 65 668 568

75%

3895 2996 102 137 840 1053

90%

6519 4842 127 237 1016 1981

Max. 87 49

125 5.7 120 63 309 319 90 60 40 60 145 319

81,817 55,131 339 1089 5163 7278

a The limit of detection, LOD for urinary BPA was 96 ng L−1 and mono-ClBPA was 32 ng L−1. Urinary concentrations below LOD for an analyte were assigned a respective half-LOD value. This corresponds to 48 ng L−1 and 16 ng L−1, for BPA and mono-ClBPA, respectively. b TTHM refers to total trihalomethanes, which is the sum of chloroform, dichlorobromomethane, chlorodibromomethane and bromoform (Charisiadis and Makris, 2014; Charisiadis et al., 2014).TTHM refers to total trihalomethanes, which is the sum of chloroform, dichlorobromomethane, chlorodibromomethane and bromoform (Charisiadis and Makris, 2014; Charisiadis et al., 2014).

mopping (rS = 0.24, p b 0.001), and toilet cleaning (rS = 0.20, p b 0.001). However, no significant correlation was found between urinary BPA and any of the participants' characteristics. Creatinine-adjusted urinary mono-ClBPA showed a significant correlation between both urinary BPA (rS = 0.24, p = 0.0003) and urinary total THM (rS = 0.69, p b 0.001), exhibiting a greater magnitude of association with the latter (data not shown).

3.3. Association between exposure variables and urinary monochlorinated bisphenol A Gender (female), age, and education (primary level) were positively associated with urinary mono-ClBPA (p b 0.05) (Table 2). The mean (SD) of creatinine-adjusted urinary mono-ClBPA in females was 137 (142) ng g− 1 versus 86 (136) ng g− 1 in males, and those in the age groups above 37 years had a significantly higher exposure (p b 0.05) (Table 2). Only 20% of the study participants consumed water from large-volume polycarbonate bottles containing BPA, while 70% consumed chlorinated tap water. Association between creatinine-adjusted urinary mono-ClBPA (ng g− 1) and ingestionrelevant exposure variables such as per capita water consumption from tap and polycarbonate bottles (L day− 1), and THM levels in water (μg L − 1 ) were non-significant (p N 0.05) (Table 2). These observations appear to corroborate that ingestion may not be the major exposure route to mono-ClBPA. Among the noningestion exposure relevant variables, level of participation in performing domestic cleaning and laundry was indicated by (i) a yes for 100%, (ii) yes-sometimes for partial (b100% to N0%), and (iii) a no for 0% involvement. Approximately, 60% of the participants reported performing household cleaning activities, and 50% did laundry by themselves; urinary mono-ClBPA was significantly higher in those subjects performing such cleaning activities when compared with the rest (Table 2).

All the classes of domestic cleaning activities monitored in this study viz., hand dishwashing, mopping, toilet and other cleaning (window glass, furniture, surfaces) were associated with urinary mono-ClBPA levels of those performing the specific cleaning activities. Study participants who reported doing any form of cleaning showed a significantly clear distinction from those who did not perform such activity (interquartile vs. Quartile 1, and Quartile 3 vs. Quartile 1) (Table 2). Urinary mono-ClBPA (ng g−1) was 1.5×higher in participants who performed any sort of domestic cleaning (EEU N 0) when compared with those who did not perform any (EEU = 0) (p b 0.001) (Table 2). By contrast, none of the exposures from individual or cumulative personal hygiene and recreational activities such as showering, bath tub use, and pool swimming were significantly associated with urinary mono-ClBPA. Similar observations were made between domestic cleaning activities and exposure to THM, but not for BPA (Table 2). In a multiple linear regression model, age, gender, and exposure equivalence units for hand dishwashing and mopping were included as predictors of both creatinine-unadjusted and adjusted urinary mono-ClBPA (Table 3). Creatinine-adjusted urinary mono-ClBPA was 1.35× higher in females compared to males (p b 0.001), 1.03× higher for 1 year increase in age (p b 0.001), and 1.22 × higher for 1 EEU increase in hand dishwashing (Table 3). The confounders could significantly explain the variability in urinary mono-ClBPA levels (R2 = 0.24, p b 0.001). An alternative testing of the model strength (R2 = 0.24, p b 0.001, Table 3) was based on the Bayes factor estimate, where its corresponding lower bound value was b0.019, indicating that the odds of the alternative hypothesis, as it was presented in the model, were higher than the odds of the null hypothesis, providing additional strength to it (Sellke et al., 2001). Residuals show a horizontal banding about the zero line indicating errors with a constant variance (Fig. SI1A). Normal distribution of the residuals was observed from the normal probability plot (Fig. SI-1B) and histograms (Fig. SI-1C). Residuals were randomly scattered around the zero line indicating a possible linear relationship between the corresponding predictors and the urinary ClBPA

258

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261

Table 2 The association between participant characteristics and practices with exposures to BPA, mono-ClBPA and trihalomethanes measured as their urinary concentrations. Mean (SD) and respective statistical analysis outcome (parametric measure such as Student's t-test or Tukey HSD or a non-parametric measure such as Wilcoxon test) were presented. Quartiles (or categories)a

Variable

n

%

Mono-ClBPA (ng g−1)

BPA (ng g−1)

Total THM (ng g−1)

Mean (SD)

Diff b

Mean (SD)

Diff b

Mean (SD)

Diff

Demographics 1 Gender

2

Female Male

129 95

58 42

2446 (1936) 3393 (6321)

ns

137 (142) 86 (136)

A B

1207 (1318) 586 (496)

A B

b37 37–64 N64

55 120 49

25 54 22

2066 (1984) 3291 (5650) 2640 (2111)

B A AB

55 (50) 120 (131) 173 (201)

B A A

527 (331) 1001 (1179) 1271 (1292)

C B A

92 (97) 121 (151)

ns

1146 (169) 1082 (81)

ns

94 (96) 132 (169) 109 (126)

ns

747 (658) 1089 (1346) 891 (908)

ns

117 (134) 128 (164) 81 (64)

ns

985 (1020) 652 (638) 1041 (1245)

A A B

Age (y)

Ingestion-relevant exposure variablesc 3 Polycarbonate bottled water consumption for PW + CB + HBd, at and away home (L day−1) 0.00 178 79 2701 (2821) ns N0.00 46 21 3394 (7970) −1 4 Tap water consumption for PW + CB + HB, at and away home (L day ) 0.00 65 29 2641 (2345) ns N0.00–1.00 107 48 3098 (5810) N1.00 52 23 2592 (2587) 5 Water total trihalomethanes (μg L−1) b56 55 25 2722 (2695) ns 56–81 121 54 3079 (5349) N81 48 21 2409 (3069) Non-ingestion relevant exposure variables 6 Domestic cleaning activities done personallye Yes 114 Yes-sometimes 18 No 92 7 Laundry done personallye Yes 110 No 113

51 8 41

2493 (2038) 3862 (3520) 3089 (6260)

B A B

139 (154) 104 (135) 88 (122)

A AB B

1218 (1327) 677 (535) 656 (706)

A AB B

49 51

2639 (2303) 2946 (5637)

ns

142 (141) 90 (139)

A B

1261 (1359) 642 (623)

A B

ns

93 (96) 129 (162) 117 (147)

ns

738 (656) 1076 (1323) 940 (1008)

ns

ns

85 (116) 125 (174) 161 (116)

C B A

687 (731) 889 (721) 1486 (1681)

B A A

ns

85 (116) 152 (181) 125 (100)

B A A

668 (724) 1126 (1087) 1296 (1601)

B A A

ns

98 (151) 142 (153) 128 (101)

B A A

720 (737) 1118 (1052) 1260 (1572)

B A A

ns

90 (121) 135 (179) 133 (108)

B A A

696 (759) 939 (779) 1367 (1668)

B A A

ns

121 (174) 121 (141) 98 (107)

ns

844 (760) 1015 (1253) 879 (979)

ns

Trihalomethanes exposure equivalence units (EEU)f 8 All ingestion EEU (tap water consumption, at and away home, PW + CB + HB) 0.00 66 30 2633 (2328) N0.00–0.58 104 46 2687 (2875) N0.58 54 24 3421 (7596) 9 Hand dishwashing EEU 0.00 105 47 2972 (5880) N0.00–0.84 63 28 2967 (2782) N0.84 56 25 2482 (1902) 10 Mopping EEU 0.00 105 47 3150 (6035) N0.00–1.00 76 34 2588 (1862) N1.00 43 19 2570 (2280) 11 Toilet cleaning EEU 0.00 118 53 3078 (5711) N0.00–0.48 50 22 2674 (2376) N0.48 56 25 2519 (1799) 12 All domestic cleaning activities EEUg 0.00 94 42 3055 (6197) N0.00–2.50 74 33 2846 (2561) N2.50 56 25 2503 (2017) 13 All personal hygienic activities EEUh b2.00 50 22 2318 (1787) 2.00–4.29 119 53 2891 (3139) N4.29 55 25 3237 (7383) a

Three classes of quartiles used in this study were bQ1 (b25th percentile), Q1–Q3 (25th–75th percentile), and NQ3 (N75th percentile). Statistical difference between the means of the quartiles was calculated by using Tukey's HSD and Student's t-test (in case of two groups). Treatment means denoted with different alphabets differ significantly (p b 0.05). ‘ns’ denotes a non-significant difference in treatment means for a given category. c 50% and 70% reduction in trihalomethane exposure was applied in case of filter use and hot beverage consumption, respectively (Forssén et al. 2007; Krasner and Wright 2005, RiveraNunez et al., 2012). d Per capita water consumption in Cyprus (Mediterranean region) includes water use not only for plain consumption (PW), but also predominantly in the form of cold (CB) and hot beverages (HB) (Makris et al., 2013a,b). e The level of involvement in the respective activity was indicated by (i) a yes for 100%, (ii) yes sometimes for partial (b100% to N0%), and (iii) a no for 0%. f An exposure equivalence unit is equal to 1 L day−1 tap water consumption; 15-min day−1 dish washing, mopping, toilet or other cleaning activities (Charisiadis et al., 2014); 5-min day−1 shower or bathtub time; and 5-min day−1 swim in pool time (Weisel and Jo, 1996; Rivera-Nunez et al., 2012). g All domestic cleaning EEU refers to sum of EEU from hand dishwashing, mopping, toilet and other cleaning activities (Charisiadis et al., 2014). h All personal hygienic activity EEU refers to sum of EEU from showering, bath tub use, and pool swimming (Charisiadis et al., 2014). b

levels (Fig. SI-1D). Multiple linear regression analysis did not reveal any significant association between household cleaning practices and urinary total BPA levels. The change in creatinine-adjusted urinary mono-

Cl-BPA per tertile increase in creatinine-adjusted urinary total BPA was highly significant in tertile 1 (p b 0.05), followed by tertile 2 (p b 0.05), and this result was non-significant in tertile 3 (p ≥ 0.05) (Table 4). The

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261 Table 3 Linear regression coefficients for study variables with log-transformed, creatinine adjusted urinary bisphenol A and mono-ClBPA (n = 224). (ln) creatinine unadjusted (ng L−1)

(ln) creatinine adjusted (ng g−1)

BPA

Mono-ClBPA

BPA

Mono-ClBPA

Model fit

R2 = 0.12; p b 0.001

R2 = 0.04; p = 0.05

R2 = 0.02; p = 0.49

R2 = 0.24; p b 0.001

Age (y) Gender [Female] Hans dishwashing EEUa Mopping EEU

−0.02⁎⁎⁎ −0.30⁎⁎⁎

0.01⁎ 0.002 0.09 0.03

0.004 −0.01 0.08 −0.05

−0.05 0.16

0.03⁎⁎⁎ 0.30⁎⁎⁎ 0.20⁎ −0.17

The critical value of statistical significance (α) was set at 0.05, 0.01, and 0.001, respectively. a An exposure equivalence unit (EEU) for trihalomethane exposure is equal to 1 L day−1 tap water consumption; 15-min day−1 dish washing, mopping, toilet or other cleaning activities; 5-min day−1 shower or bathtub time; and 5-min day−1 swim in pool time (Weisel and Jo, 1996; Rivera-Nunez et al., 2012; Charisiadis et al., 2014). ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

increment in urinary mono-ClBPA was 1.43×higher with a unit increase in urinary total BPA as a continuous variable and was highly significant (p b 0.001). In contrast, the change in creatinine-adjusted urinary mono-Cl-BPA per tertile increase in creatinine-adjusted urinary total THM was non-significant in tertile 1 (p ≥ 0.05), while it was significant in tertile 2 (p b 0.05) and tertile 3 (p b 0.001). The increment in urinary mono-ClBPA was 2.30 × higher with a unit increase in urinary total THM as a continuous variable (p b 0.001) (Table 4).

4. Discussion This study was the first to report the presence of monochlorinated BPA in an adult population of considerable size (n = 224). The frequency of occurrence (N LOD) for urinary total BPA and mono-ClBPA in our study population was 100% and 90%, respectively. Geometric mean urinary concentration of creatinine-adjusted mono-ClBPA (71 ng g−1) was comparable to that reported by Liao and Kannan (GM 55 ng g−1), which is the only other available study measuring chlorinated BPA analogs in human urine. However, the frequency of occurrence for urinary total BPA and mono-ClBPA in their study population was 97% and 16%, respectively (Liao and Kannan, 2012). Reported geometric mean (and frequency of occurrence) values of the other chlorinated BPA in urine were 49 ng g−1 (19%) for di-ClBPA and 47 ng g−1 (19%) for triClBPA (Liao and Kannan, 2012). Higher concentrations of ClBPAs, however, with lower frequency of occurrence, were measured in adipose (13 ng g−1) (Fernandez et al., 2007) and placental (44 ng g−1) tissues (Jimenez-Diaz et al., 2010).

259

This study provided insight on possible associations of noningestion exposure activities with urinary mono-ClBPA concentrations, accounting for possible confounders (age, gender and BMI). Sub-analyses on stratified potential covariates revealed that mono-ClBPA exposure was more frequent in females who performed hand dishwashing. Females reported a substantially higher proportion of domestic cleaning activities (such as hand dishwashing, mopping and toilet cleaning) compared to males in this study (p b 0.0001). Exposure from water ingestion did not significantly contribute to urinary mono-ClBPA, and neither from hygienic activities such as showering, bath tub use, and pool swimming. The main exposure sources of Cl-BPA seem to be parts of various external environmental compartments, such as tap water (Hu et al., 2002; Fan et al., 2013), water treatment plants (Dupuis et al., 2012), several types of water such as influent, effluent of wastewater treatment facility (Gallart-Ayala et al., 2010), sewage waste (Fernandez et al., 2007), and paper recycling waste (Fukazawa et al., 2002). BPA was shown to generate a suite of chlorinated transformation products in tap water when in contact with environmentally-realistic residual free chlorine concentrations (0.5–1.0 mg L−1) (Gallard et al., 2004). Moreover, BPA spiked in water disappeared quickly upon chlorination; while the generated chlorinated BPA analogs lasted for about a day (Gallard et al., 2004). Moreover, residual BPA presence in dish and laundry detergents, cleaning solutions, etc. (Dodson et al., 2012) may form Cl-BPA upon their first contact with chlorine used in any of the domestic chores. In a previous study, we found a significant positive association between domestic cleaning activities and exposures to THM suggesting the role of chlorine-based household products (Charisiadis et al., 2014). Most importantly, BPA is present in a suite of personal care products (soaps, shampoo, and sunscreen) and cosmetics (nail polish, beauty lotions) (Dodson et al., 2012). Further information on the fate and stability of Cl-BPA analogs in various environmental media is needed to explain the significant positive association between urinary ClBPA levels and the duration of domestic cleaning activities. 4.1. Study strengths and limitations

Increment in (ln) creatinine-adjusted mono-ClBPA to total BPA (95% CI)

Increment in (ln) creatinine-adjusted mono-ClBPA to total THM (95% CI)

Categorical First tertile Second tertile Third tertile

1.18 (0.28, 2.09)⁎ 1.35 (0.19, 2.52)⁎ −0.04 (−0.45, 0.37)

0.39 (−0.22, 1.00) 0.81 (0.08, 1.54)⁎ 0.85 (0.56, 1.14)⁎⁎⁎

This study using an appreciable sample size (n = 224) was the first of its nature to explore associations between urinary mono-ClBPA levels and domestic activities that involve contact of BPA-containing household products with chlorinated tap water. An advantage of our study design was that it included questionnaire items covering most of the ingestion and non-ingestion routes of exposure to chlorinated chemicals. An additional advantage of this study was analyzing for urinary total BPA, mono-ClBPA and total THM concentrations that provided hints on their co-occurring patterns. One of the limitations of this study was that quantification of chlorinated BPA analogs was reliably performed only for monochlorinated BPA. The absence of dietary specific questionnaire items pertaining to BPA sources prevented us from acquiring a more complete picture of all possible BPA exposure sources. Questionnaire outcome-based inferences, variation in seasonal water consumption and use, and variation in chemical composition of the domestic products used by the study participants added to the uncertainty of our predicted models. A single first morning void urine sample was another limitation for this study. Chlorinated analogs of BPA are considered more lipophilic compared to BPA and may not undergo glucuronidation detoxification pathway (Migeot et al., 2013), possibly exhibiting a longer urinary excretion half-life (beyond the typical 3–6 hour body clearance for BPA). This increases the likelihood of capturing Cl-BPA exposures during the previous day(s) or evening prior to the first morning void.

Continuous Per log unit change

0.36 (0.18, 0.55)⁎⁎⁎

0.83 (0.71, 0.95)⁎⁎⁎

5. Implication

Table 4 Increment change in log-transformed, creatinine-adjusted urinary mono-ClBPA per tertile, and a unit increase in urinary total BPA and total trihalomethanes. Characteristic

The critical value of statistical significance (α) was set at 0.05, 0.01, and 0.001, respectively. ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

Our cross-sectional study findings have important environmental health implications, because habitual individual characteristics, such as the frequency and duration of household cleaning activities may be

260

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261

associated with elevated human exposures to chlorinated BPA and TTHM. Similar to earlier THM trends (Charisiadis et al., 2014), elevated urinary mono-ClBPA levels were not explained by tap and polycarbonate bottled water ingestion-related sources, but rather explained by noningestion activities. Household cleaning activities like dishwashing, mopping, and toilet cleaning were mostly conducted by females in our study population and this was indeed reflected onto their monoClBPA and total THM urinary levels. The cleaning activities influencing the magnitude of both mono-ClBPA and total THM were the same, indicating that the presence of disinfectant chlorine in tap water could be triggering the formation of chlorinated BPA analogs when in contact with household cleaning or personal care products. Our results indicated that females who perform domestic cleaning activities several times in short bursts, in particular hand dishwashing, are more likely to be at higher mono-ClBPA exposure risk. In light of recent cell-based studies highlighting the equal, if not, greater endocrine disrupting activity of chlorinated BPA compounds at environmentally-relevant concentrations, this exposure assessment study presents new non-dietary exposure sources and routes of exposure (dermal uptake) to chlorinated BPA via routine cleaning household habits. The presence of BPA in personal care products such as, bar soap, body lotion, shampoo/ conditioner, shaving cream, face lotion and cleanser, sunscreens and nail polish (Dodson et al., 2012, Liao and Kannan, 2014) should be considered for inclusion in exposure assessment studies as relevant non-dietary exposure sources for chlorinated BPA. Epidemiological studies and probabilistic BPA risk assessment exercises should consider assessing daily intake estimates for chlorinated BPA compounds, as well. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.04.072. Competing financial interests The authors declare that none financial interest exists. Disclaimer The observations and speculations in this article represent those of the authors and do not necessarily reflect the views of the participating organizations, viz., Cyprus University of Technology, Limassol, Cyprus. Acknowledgments The authors would like to thank the Cyprus Research Promotion Foundation for funding this study (AEIFORIA/ASTI/0311(BIE)/20) with Structural Funds of the European Commission awarded to the corresponding author, Dr. K.C. Makris. References Babu S, Vellore NA, Kasibotla AV, Dwayne HJ, Stubblefield MA, Uppu RM. Molecular docking of bisphenol A and its nitrated and chlorinated metabolites onto human estrogen-related receptor-gamma. Biochem Biophys Res Commun 2012;426(2): 215–20. Cariot A, Dupuis A, Albouy-Llaty M, Legube B, Rabouan S, Migeot V. Reliable quantification of bisphenol A and its chlorinated derivatives in human breast milk using UPLC-MS/ MS method. Talanta 2012;100:175–82. Charisiadis P, Makris KC. A sensitive and fast method for trihalomethanes in urine using gas chromatography-triple quadrupole mass spectrometry. J Chromatogr B 2014; 947–948:17–22. Charisiadis P, Andra SS, Makris KC, Christodoulou M, Christophi CA, Kargaki S, et al. Household cleaning activities as noningestion exposure determinants of urinary trihalomethanes. Environ Sci Technol 2014;48(1):770–80. Chen L, Suh BI. Bisphenol A in dental materials: a review. JSM Dent 2013;1:1004. Clarke. Jaffe method. The MAK-Collection for Occupational Health and Safety, Part IV: Biomonitoring Methods, vol. 12. Weinheim: Wiley-VCH, Verlag, KGaA; 1961. p. 169–84. Covaci A, Voorspoels S, Abdallah MA, Geens T, Harrad S, Law RJ. Analytical and environmental aspects of the flame retardant tetrabromobisphenol-A and its derivatives. J Chromatogr A 2009;1216(3):346–63.

Dodson RE, Nishioka M, Standley LJ, Perovich LJ, Brody JG, Rudel RA. Endocrine disruptors and asthma-associated chemicals in consumer products. Environ Health Perspect 2012;120(7):935–43. Dupuis A, Migeot V, Cariot A, Albouy-Llaty M, Legube B, Rabouan S. Quantification of bisphenol A, 353-nonylphenol and their chlorinated derivatives in drinking water treatment plants. Environ Sci Pollut Res 2012;19(9):4193–205. Fan Z, Hu J, An W, Yang M. Detection and occurrence of chlorinated byproducts of bisphenol a, nonylphenol, and estrogens in drinking water of china: comparison to the parent compounds. Environ Sci Technol 2013;47:10841–50. Fernandez MF, Arrebola JP, Taoufiki J, Navalon A, Ballesteros O, Pulgar R, et al. Bisphenol-A and chlorinated derivatives in adipose tissue of women. Reprod Toxicol 2007;24(2): 259–64. Forssén UM, Herring AH, Savitz DA, Nieuwenhuijsen MJ, Murphy PA, Singer PC, et al. Predictors of use and consumption of public drinking water among pregnant women. J Expo Sci Environ Epidemiol 2007;17(2):159–69. Fukazawa H, Hoshino K, Shiozawa T, Matsushita H, Terao Y. Identification and quantification of chlorinated bisphenol A in wastewater from wastepaper recycling plants. Chemosphere 2001;44(5):973–9. Fukazawa H, Watanabe M, Shiraishi F, Shiraishi H, Shiozawa T, Matsushita H, et al. Formation of chlorinated derivatives of bisphenol A in waste paper recycling plants and their estrogenic activities. J Health Sci 2002;48(3):242–9. Gallard H, Leclercq A, Croue JP. Chlorination of bisphenol A: kinetics and by-products formation. Chemosphere 2004;56(5):465–73. Gallart-Ayala H, Moyano E, Galceran MT. Liquid chromatography/multi-stage mass spectrometry of bisphenol A and its halogenated derivatives. Rapid Commun Mass Spectrom 2007;21(24):4039–48. Gallart-Ayala H, Moyano E, Galceran MT. On-line solid phase extraction fast liquid chromatography–tandem mass spectrometry for the analysis of bisphenol A and its chlorinated derivatives in water samples. J Chromatogr A 2010;1217(21):3511–8. Geens T, Goeyens L, Covaci A. Are potential sources for human exposure to bisphenol-A overlooked? Int J Hyg Environ Health 2011;214(5):339–47. Halden RU. Plastics and health risks. Annu Rev Public Health 2010;31:179–94. Hu JY, Aizawa T, Ookubo S. Products of aqueous chlorination of bisphenol A and their estrogenic activity. Environ Sci Technol. 2002;36(9):1980–7. Ibuki Y, Tani Y, Toyooka T. UVB-exposed chlorinated bisphenol A generates phosphorylated histone H2AX in human skin cells. Chem Res Toxicol 2008;21(9):1770–6. Jimenez-Diaz I, Zafra-Gomez A, Ballesteros O, Navea N, Navalon A, Fernandez MF, et al. Determination of bisphenol A and its chlorinated derivatives in placental tissue samples by liquid chromatography–tandem mass spectrometry. J Chromatogr B Anal Technol Biomed Life Sci 2010;878(32):3363–9. Kang JH, Kondo F, Katayama Y. Human exposure to bisphenol A. Toxicology 2006; 226(2–3):79–89. Krasner SW, Wright JM. The effect of boiling water on disinfection by-product exposure. Water Res 2005;39(5):855–64. Kuruto-Niwa R, Terao Y, Nozawa R. Identification of estrogenic activity of chlorinated bisphenol A using a GFP expression system. Environ Toxicol Pharmacol 2002;12(1): 27–35. Leri AC, Anthony LN. Formation of organochlorine by-products in bleached laundry. Chemosphere 2013;90(6):2041–9. Liao C, Kannan K. Determination of free and conjugated forms of bisphenol A in human urine and serum by liquid chromatography–tandem mass spectrometry. Environ Sci Technol 2012;46(9):5003–9. Liao C, Kannan K. Concentrations and profiles of bisphenol A and other bisphenol analogues in foodstuffs from the United States and their implications for human exposure. J Agric Food Chem 2013;61:4655–62. Liao C, Kannan K. A survey of alkylphenols, bisphenols, and triclosan in personal care products from China and the United States. Arch Environ Contam Toxicol 2014. http://dx.doi.org/10.1007/s00244-014-0016-8. Liu JH, Carr S, Rinaldi K, Chandler W. Screening estrogenic oxidized by-products by combining ER binding and ultrafiltration. Environ Toxicol Pharmacol 2005;20(2):269–78. Liu H, Zhao HM, Quan X, Zhang YB, Chen S. Formation of chlorinated intermediate from bisphenol A in surface saline water under simulated solar light irradiation. Environ Sci Technol 2009;43(20):7712–7. Makris KC, Andra SS, Herrick L, Christophi CA, Snyder SA, Hauser R. Association of drinking-water source and use characteristics with urinary antimony concentrations. J Expo Sci Environ Epidemiol 2013a;23(2):120–7. Makris KC, Andra SS, Jia A, Herrick L, Christophi CA, Snyder SA, et al. Association between water consumption from polycarbonate containers and bisphenol A intake during harsh environmental conditions in summer. Environ Sci Technol 2013b;47(7): 3333–43. Migeot V, Dupuis A, Cariot A, Albouy-Llaty M, Pierre F, Rabouan S. Bisphenol A and its chlorinated derivatives in human colostrum. Environ Sci Technol 2013;47(23): 13791–7. Molina-Molina JM, Amaya E, Grimaldi M, Saenz JM, Real M, Fernandez MF, et al. In vitro study on the agonistic and antagonistic activities of bisphenol-S and other bisphenol-A congeners and derivatives via nuclear receptors. Toxicol Appl Pharmacol 2013;272(1):127–36. Mutou Y, Ibuki Y, Terao Y, Kojima S, Goto R. Change of estrogenic activity and release of chloride ion in chlorinated bisphenol A after exposure to ultraviolet B. Biol Pharm Bull 2006;29(10):2116–9. Mutou Y, Ibuki Y, Terao Y, Kojima S, Goto R. Induction of apoptosis by UV-irradiated chlorinated bisphenol A in Jurkat cells. Toxicol In Vitro 2008;22(4):864–72. Nazaroff WW, Weschler CJ. Cleaning products and air fresheners, exposure to primary and secondary air pollutants. Atmos Environ 2004;38(18):2841–65. Odabasi M. Halogenated volatile organic compounds from the use of chlorine-bleachcontaining household products. Environ Sci Technol 2008;42(5):1445–51.

H. Kalyvas et al. / Science of the Total Environment 490 (2014) 254–261 Olson DA, Corsi RL. In-home formation and emissions of trihalomethanes, the role of residential dishwashers. J Expo Anal Environ Epidemiol 2004;14(2):109–19. Riu A, Grimaldi M, le Maire A, Bey G, Phillips K, Boulahtouf A, et al. Peroxisome proliferator-activated receptor gamma is a target for halogenated analogs of bisphenol A. Environ Health Perspect 2011a;119(9):1227–32. Riu A, le Maire A, Grimaldi M, Audebert M, Hillenweck A, Bourguet W, et al. Characterization of novel ligands of ER alpha, Er beta, and PPAR gamma: the case of halogenated bisphenol A and their conjugated metabolites. Toxicol Sci 2011b;122(2):372–82. Rivera-Nunez Z, Wright JM, Blount BC, Silva LK, Jones E, Chan RL, et al. Comparison of trihalomethanes in tap water and blood: a case study in the United States. Environ Health Perspect 2012;120(5):661–7. Sellke T, Bayarri MJ, Berger JO. Calibration of ρ values for testing precise null hypotheses. Am Stat 2001;55(1):62–71. Staples CA, Dorn PB, Klecka GM, O'Block ST, Harris LR. A review of the environmental fate, effects, and exposures of bisphenol A. Chemosphere 1998;36(10):2149–73. Takemura H, Ma J, Sayama K, Terao Y, Zhu BT, Shimoi K. In vitro and in vivo estrogenic activity of chlorinated derivatives of bisphenol A. Toxicology 2005;207(2):215–21. Terasaki M, Kosaka K, Kunikane S, Makino M, Shiraishi F. Assessment of thyroid hormone activity of halogenated bisphenol A using a yeast two-hybrid assay. Chemosphere 2011;84(10):1527–30. US EPA. Determination of chlorination disinfection byproducts, chlorinated solvents, and halogenated pesticides/herbicides in drinking water by liquid–liquid extraction and

261

gas chromatography with electron-capture detection. Method 551.1. Cincinnati, OH: National Exposure Research Laboratory, Office of Research and Development, US Environmental Protection Agency; 1995 [Available, http://water.epa.gov/scitech/ drinkingwater/labcert/analyticalmethods_ogwdw.cfm. (accessed 10 March 2014)]. Vandenberg LN, Hauser R, Marcus M, Olea N, Welshons WV. Human exposure to bisphenol A (BPA). Reprod Toxicol 2007;24(2):139–77. Vandenberg LN, Chahoud I, Heindel JJ, Padmanabhan V, Paumgartten FJR, Schoenfelder G. Urinary, circulating, and tissue biomonitoring studies indicate widespread exposure to bisphenol A. Environ Health Perspect 2010;118(8):1055–70. Viñas R, Goldblum RM, Watson CS. Rapid estrogenic signaling activities of the modified (chlorinated, sulfonated, and glucuronidated) endocrine disruptor bisphenol A. Endocr Disruptors 2013;1(1):e25411. Weisel CP, Jo WK. Ingestion, inhalation, and dermal exposures to chloroform and trichloroethene from tap water. Environ Health Perspect 1996;104(1):48–51. Yamamoto T, Yasuhara A. Chlorination of bisphenol A in aqueous media: formation of chlorinated bisphenol A congeners and degradation to chlorinated phenolic compounds. Chemosphere 2002;46(8):1215–23. Yamauchi K, Ishihara A, Fukazawa H, Terao Y. Competitive interactions of chlorinated phenol compounds with 3,3′,5-triiodothyronine binding to transthyretin: detection of possible thyroid-disrupting chemicals in environmental waste water. Toxicol Appl Pharmacol 2003;187(2):110–7.