Article pubs.acs.org/est
Highly Selective Screening of Estrogenic Compounds in ConsumerElectronics Plastics by Liquid Chromatography in Parallel Combined with Nanofractionation-Bioactivity Detection and Mass Spectrometry Willem Jonker,† Ana Ballesteros-Gómez,*,‡ Timo Hamers,‡ Govert W. Somsen,† Marja H. Lamoree,‡ and Jeroen Kool† †
Division of Bioanalytical Chemistry, Vrije Universiteit Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands Institute for Environmental Studies, Vrij Universiteit Amsterdam, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands
‡
S Supporting Information *
ABSTRACT: The chemical safety of consumer products is an issue of emerging concern. Plastics are widely used, e.g. as casings of consumer electronics (TVs, computers, routers, etc.), which are present in houses and offices in continuously increasing numbers. In this study, we investigate the estrogenic activity of components of plastics coming from electronics’ casings. A recently developed fractionation platform for effect-directed analysis (EDA) was used. This platform combines reversed-phase liquid chromatography in parallel with bioassay detection via nanofractionation and with online high-resolution time-of-flight mass spectrometry (TOFMS) for the identification of bioactives. Four out of eight of the analyzed plastics samples showed the presence of estrogenic compounds. Based on the MS results these were assigned to bisphenol A (BPA), 2,4-di-tert-butylphenol, and a possible bisphenol A analog. All samples contained flame retardants, but these did not show any estrogenicity. The observed BPA, however, could be an impurity of tetrabromo-BPA (TBBPA) or TBBPA-based flame retardants. Due to the plausible migration of additives from plastics into the environment, plastics from consumer electronics likely constitute a source of estrogenic compound contamination in the indoor environment.
■
INTRODUCTION A growing body of scientific evidence indicates that the indoor environment (homes and other buildings) can be more polluted with organic contaminants than the outdoor. Since we spend most of our time indoors, the study of potential sources of contamination of harmful chemicals, such as endocrine disruptors, becomes essential to improve the safety of our indoor environment. Consumer products are important sources of indoor contamination,1,2 and, among them, plastics are widely used. Plastics contain additives, such as flame retardants, phthalates, antioxidants, or light stabilizers, to improve the physicochemical properties of the material. Exposure to these additives can take place directly by dermal contact and ingestion (e.g., mouthing of products by toddlers and hand-to-mouth contact).3 Indirect exposure through indoor dust and air can also occur, since additives easily leach into the environment from products through volatilization or abrasion.4,5 High levels of phthalates, flame retardants, and bisphenols, among other contaminants, have been reported in indoor dust.6−9 Furthermore, employees working at electronic waste recycling sites are under additional risk of exposure, especially when this occurs under primitive conditions.10,11 Although the plastic polymers often are considered inert, some of these additives can cause health problems such as cancer, reproductive disorders, neurotoxic effects, and severe allergies.12,13 © 2016 American Chemical Society
Plastic leachates have been reported to exert aquatic toxicity which varied with the type of plastic and the weathering conditions.14,15 More specifically, estrogenic activity has been reported in leachates from polycarbonate and epoxy resins used as food contact materials that are known to contain bisphenol A (BPA) as an unpolymerized residue.16−18 More recently, also plastic leachates made up of other polymers (the so-called “BPA-free” plastics) have been reported to exert estrogenic activity too, although individual chemicals have not yet been identified.19,20 In a recent study, Tritan, as a replacement of polycarbonate plastics, was found to leach low levels of the estrogens BPA and dimethyl isophthalate (DMIP) under standard protocols for food contact materials, although levels produced negligible endocrine disrupting effects in in vivo and in vitro assays.21 A recent study of our research group reported that plastic casings of electronic equipment contain a variety of phthalates, UV filters, antioxidants, flame retardants, and related compounds.22 Among these chemicals, some phthalates (e.g., butyl benzyl phthalate, dibutyl phthalate),23 antioxidants (e.g., 2-hydroxy- 4-methoxybenzophenone, 2,2′-dihydroxy-4-methoxReceived: Revised: Accepted: Published: 12385
July 27, 2016 October 12, 2016 October 24, 2016 October 24, 2016 DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology ybenzophenone),24 and flame retardants (e.g., triphenyl phosphate)25,26 have been reported to be estrogenic. Due to the increasing number of consumer electronics in homes and offices and their complex composition, they could constitute an important source of organic contaminants in the indoor environment. As mentioned before, estrogenic activity has been reported in plastics but, to the best of our knowledge, never in plastics from electronics. Furthermore, studies scarcely aimed to elucidate the identity of the individual bioactive chemicals (other than BPA). The aim of this study was to identify estrogenic compounds in plastic casings from electronics that could leach into the environment. For this purpose, we have used a recently developed fractionation platform for effect-directed analysis (EDA).27 The platform combines liquid chromatography (LC) with high resolution time-of-flight mass spectrometry (TOFMS) in parallel with a human cell (VM7Luc4E2) gene reporter assay via nanofractionation for the detection of estrogenic compounds. The platform allows reconstruction of bioassay chromatograms that can directly be correlated to MS chromatograms recorded in parallel, allowing straightforward pinpointing of accurate masses of estrogenic compounds. Results were obtained within a single fractionation cycle, resulting in a drastic decrease of total analysis time and thus an increase in throughput compared to traditional EDA studies.
described previously.27 A Multidrop was used for microliter dispensing in microtiter plates and a VarioSkan plate reader for bioassay readout (ThermoFisher Scientific, Breda, The Netherlands). FTIR measurements were done on a selection of the plastics with a FTIR 8400S spectrometer from Shimadzu (Breda,The Netherlands) for polymer identification (scan range 400−4000 cm−1). Sample Collection and Preparation. Eight samples of electrical/electronic devices were bought in supermarkets in The Netherlands in 2015. A detailed list of the samples is given in Table S-1 (Supporting Information). Samples were preselected from a total of 30 samples by a fast ambient mass spectrometry screening method28 to select those ones that contained flame retardants as a compound class with potential estrogenic properties. Sample preparation and extraction was adapted from Ballesteros-Gómez et al.,22 a method previously optimized for covering additives with a wide range of polarity. Plastic samples from the hard plastics casings of electronic/electrical devices were taken using a surgical cutter. Samples (approximately 50 mg) were extracted with 10 mL of a mixture of THF−MeOH (50:50, v/v) by sonicating (30 min) and stirring (200 rpm) for 12 h. Extracts were evaporated (40 °C, N2) and reconstituted in 1 mL of MeOH, ultracentrifuged (10.000 rpm, 5 min), and further filtrated if required for removal of remaining particles in suspension (with 0.2 μm microcentrifuge filters). Aliquots of 7 μL were analyzed by LC-MS/nanofractionation. In order to cover a wide range of compounds, no further cleanup was performed before analysis. LC-High Resolution MS-Nanofractionation. For LC separation, Milli-Q water and MeOH were used in the following gradient: 55% MeOH (v/v) for 1 min, linear gradient to 72% MeOH (v/v) in 4 min, linear gradient to 85% MeOH (v/v) in 10 min, linear gradient to 95% in 0.5 min, and hold at isocratic conditions for 5 min. The flow rate was 0.25 mL/min, the column temperature was set at 35 °C, and the injection volume was 7 μL. The column eluate was split 9:1 toward, respectively, the fraction collector and the microTOF II mass spectrometer. For TOFMS calibration using ESI, a solution forming sodium formate adducts after ionization was prepared according to the instructions of the instrument supplier (9.9 mL isopropyl alchohol:water 1:1 v/v containing 0.2% of acetic acid %, v/v and 100 μL of 1 M NaOH). For calibration in APCI mode, a commercial calibration solution (APCI-LC low concentration tuning mix from Agilent Technologies) was employed instead. All samples were analyzed in positive and negative mode with both ESI and APCI sources. Calibration was done before each batch of experiments or in each sample (within the first minute of each chromatogram) by means of a syringe pump. TOFMS was optimized for m/z 100−1000. Capillary exit and skimmer 1 were set at 90 and 30 V, respectively, hexapole RF at 150 Vpp, transfer time at 50 μs, and pulse storage time at 5 μs. The heater temperature was set at 200 °C (APCI) and the vaporizer temperature at 280 °C (ESI, APCI). The corona voltage value was +5000 and −9000 nA in APCI(+) and APCI(−), respectively. Additional measurements were done for a higher mass range of m/z 1000−2000 with the aim of identifying polymer flame retardants. For this purpose the hexapole RF was set at 250 Vpp, transfer time at 70 μs, and pulse storage time at 15 μs. Regarding the fraction collection, a solvent keeper was first added to the well plates. In detail, 5 μL of a 10% DMSO/H2O
■
MATERIAL AND METHODS Chemical and Reagents. Methanol (MeOH) was from J.T. Baker (Center Valley, USA). Tetrahydrofuran (THF) was acquired from Biosolve (Valkenswaard, The Netherlands). Milli-Q water was obtained from an ultrapure water purification Q-Pod system (Millipore, Bedford, USA). Bisphenol A, 2,4-ditert-butylphenol, and diphenolic acid were obtained from Sigma-Aldrich. All solvents and reagents were of analytical grade and used as supplied. For sample treatment, microcentrifuge filters (0.2 μm, nylon) from Costar Spin-X (Sigma-Aldrich) were used for removing microparticles from sample extracts when necessary. For bioassay testing the following reagents were used: Dulbecco’s modified eagle medium (DMEM) containing F12 glutamax, charcoal stripped fetal bovine serum (FBS), and low glucose, phenol free DMEM (Thermofisher, Landsmeer, The Netherlands). D-Luciferin, streptomycin, penicillin, fetal bovine serum, dimethyl sulfoxide, estradiol, tris(hydroxymethyl)aminomethane (TRIS), dithiotreitol (DTT), 1,2-cyclohexylenedinitrilo-tetraacetic acid (CDTA), glycerol, Triton-X100, G418, phosphate buffered saline (PBS), adenosine triphosphate (ATP), coenzyme A, and ethylenediaminetetraacetic acid (EDTA) were obtained from Sigma-Aldrich (Zwijndrecht, The Netherlands). Apparatus. An Agilent Technologies 1290 autosampler and 1290 binary pump (Amstelveen, The Netherlands) equipped with a Luna C18 column (3 μm particle size, 2 mm I.D., 10 mm length) from Phenomenex (Torrance, CA, USA) were used for separation. MS detection was performed with a Bruker microTOF II mass spectrometer equipped with a LC-APCI II and electrospray ionization (ESI) source (Bruker Daltonics, Bremen, Germany). MS/MS analysis was conducted with a maXis high resolution quadrupole time-of-flight mass spectrometer from Bruker Daltonics equipped with an ESI source. The fraction collector used for high resolution fractionation in 96-well plates consisted of a modified Gilson 235p autosampler and was controlled with in-house developed software as 12386
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology
Figure 1. LC-nanofractionation-MS workflow to screen for the presence of estrogenic compounds in plastics from electronic products.
transparent 96 well cell culture plates (100 μL/well). The outer wells were filled with 200 μL of PBS to prevent edge effects. The next day, aliquots of 5 μL of the reconstituted fractions were transferred to the cell containing plates via a multichannel pipet. To each plate, 3 controls were added in duplicate by adding 5 μL of 3.0, 0.1, and 0 nM estradiol to the 100 μL/well cell suspension. After 24 h incubation at 37 °C and 5% CO2, cells were visually inspected and lysed. Bioassay readout was performed with a plate reader. A bioassay chromatogram was reconstructed by plotting the bioassay response of each fraction against the corresponding fraction time for comparison with the MS signal. MS/MS High Resolution QTOF-MS. Further MS/MS experiments for compound identification were performed with a high resolution QTOF mass spectrometer and the ESI source operating in negative mode. The source parameters were as follows: capillary, 1 kV; end plate offset, 400 V; charging voltage, 500 V; nebulizer gas, 4.1 bar; dry gas, 3.0 L/min; dry temperature, 200 °C; and vaporizer temperature, 300 °C. The mass analyzer settings were as follows: funnel 1 RF, 200 Vpp (peak-to-peak voltage); multipole RF, 400 Vpp; quadrupole ion energy, 5.0 eV; collision RF, 330 Vpp; transfer time, 50 μs; and prepulse storage, 10 μs. Collision energy for MS/MS experiments was set at 30 eV. Data Processing. Bioactive peaks were identified by generating molecular formulas using the tool smart formula of the Data Analysis program based on mass accuracy and isotopic pattern fit. General parameters for formulas generation were as follows: mass error tolerance was set to 5 ppm; H/C ratio from 0 to 3; number of rings and double bonds was restricted from 0.5 to 40. In general, the following criteria were taken into account to identify a positive peak: a) the mass accuracy (ppm error) was below 5 ppm and the isotopic pattern fit (mSigma value) below 100; b) the signal was at least of 5,000 counts of intensity or 3 times higher than the background noise in blanks; the ionization mode giving a higher response was consistent
mixture was dispensed in each well with a multidrop microdispenser and functioned as a keeper during the drying step. The fraction collection time was set at 10 s/fraction and was started 1 min after injection by turning the switch valve from fraction collector waste to fraction collector tip. A total of 108 fractions were collected per sample in two 96-well plates leaving the outer rows empty to prevent edge effects during bioassay testing. Additionally, 6 wells were used for bioassay controls (3.0, 0.1, and 0 nM estradiol duplicates). Fractionation was performed in serpentine fashion starting in the first well B2 and moving in steps of 10 s/fraction toward the last well in this row (B10). Subsequently, the tip was moved down 1 row and continued the movement in the opposite direction. This serpentine motion was continued until the final fraction of the well plate (G2) after which the fraction collection tip moved to the second 96-well plate where the process was repeated. Bioassay Testing for the Detection of Estrogenic Compounds. After fraction collection with parallel MS detection, well plates were dried via vacuum centrifugation to remove the organic modifier prior to bioassay analysis. Fractions were reconstituted by addition of DMSO (20 μL/ well) and shaken for 10 min at 500 rpm. Subsequently, 200 μL of water was added after which the well plates were shaken at 300 rpm. Detection of estrogenic compounds was performed with a reporter gene assay using human VM7Luc4E2 cells which were kindly provided by Michael Denison (University of California, Davis, CA). The applied assay was selected because of its relatively high sensitivity and selectivity compared to other commonly applied assays, such as proliferation, ligand binding, and yeast based assays. Cell culturing was performed according to Rogers and Denison,29 and the bioassay analysis was performed as described before by our group with minor modifications.25 In brief, cells were subcultured in assay medium containing stripped FBS. After 3 days, the medium was refreshed, and after 7 days from the subculture the cells were seeded at a concentration of 200,000 cells/mL in 12387
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology
Figure 2. Reconstructed bioassay chromatograms aligned with the MS chromatograms of (A) sample 4 and (B) sample 8. MS base-peak chromatograms (BPCs), extracted-ion chromatograms (EICs), and mass spectra of the most abundant ion in the bioactive peak are shown.
estrogenic compounds) and subjected to LC-MS/nanofractionation. The platform was previously validated for measuring low levels of estrogens in environmental waters by injection of standards and spiked samples (e.g., LODs of 3.2 nM and 320 pM were reported for estriol and estradiol, respectively).27 The workflow of the LC-MS/nanofractionation study is shown in Figure 1. Extracts of the plastic were analyzed by LC, and the column eluate was split toward the mass spectrometer and fraction collector. Subsequently, fractions were subjected to bioassay testing, and a reconstructed chromatogram was created for pinpointing bioactive peaks in the MS base-peak chromatogram. Molecular formulas were generated with high resolution MS software and ranked on their ppm error and isotope pattern fit. Next, a database and literature search was performed to retrieve the identity of the estrogenic compounds for the given formulas. Those matches that were most frequently cited in the literature (e.g., Chemspider database, number of references) and/or those with reported estrogenic activity were selected as the most suitable candidates. MS/MS experiments were carried out for structure confirmation on the basis of fragmentation. Finally, an authentic standard of the potential estrogenic compound candidate was injected for confirmation if available. The reconstructed chromatograms obtained after EDA on the eight samples tested showed bioactive peaks in four of them. As an example, Figure 2 shows the results obtained for
with the chemical structure of the compound (negative and/or positive and ESI or APCI); c) the matches were not false by visual inspection (e.g., they do not come from in-source fragmentation peaks of larger molecules). The bioactive peaks were further confirmed by MS/MS experiments and by the injection of pure standards (when available) to match both retention time and spectra. The presence of less abundant flame retardants and related degradation products or impurities was further studied in the bioactive extracts for the discussion of the results. They were identified by both target screening (in-house database) and by nontarget approaches. For generating formulas, the ions [M − H]−, [M + H]+, and [M + Na]+ were considered the most probable together with [M − Br + O]−, [M − Cl + O]−, [M + O2]− for brominated compounds in APCI(−). For the detection of highly halogenated compounds (>5 halogenated atoms), not easily recognized by the target analysis/database approach, the tool isotope cluster analysis of the software Data Analysis (Bruker Daltonics) was employed.30
■
RESULTS AND DISCUSSION Identification of Estrogenic Compounds in Plastic Casings from Electronics. Eight samples of hard plastic casings of electronic devices and previously screened by direct probe-TOF-MS were selected for this study based on their relatively high content of flame retardants (as potential 12388
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology
diphenolic acid (DPA) as the most probable match (taken into account the number of references). DPA is widely used as additive in paints, lubricating oil additives, cosmetics, surfactants, plasticizers, and textiles and has been proposed to replace BPA as a plasticizer in the production of industrial and household plastics.36 DPA is a weak ERα agonist, just as BPA, and a structural analog.37 However, the injection of the authentic standard in acidified mobile phase (0.5% v/v acetic acid in water) in order to keep the protonated form of the compound revealed that DPA eluted much earlier than the unknown peak (2.5 min instead of 7.5 min). The third identified compound, BPA, is a well-known weak estrogenic compound used as monomer (unpolymerized residue) and additive in a variety of plastics.18 While its presence in food contact materials has been widely reported,16−18 plastic casings from electronics have not been reported as potential sources of BPA contamination thus far. Since polycarbonate is commonly used in casings of consumer electronics and BPA could be present as a residue in this type of plastic, the three plastics showing BPA as a bioactive peak were analyzed by FTIR in order to identify the polymer. The three plastics showed the same main FTIR spectra peaks that were related to HIPS (High Impact Polystyrene) with some characteristic peaks at 966 cm−1 (trans-2-butene-1,4-diyl moieties), 1603 cm−1 (styrene moieties), and 1453 cm−1 (C−H bonds from butadiene). So, the presence of BPA was not related to the plastic polymer itself. Confirmation of the Identified Compounds. The identities of both BPA and 2,4-DTBP were confirmed by both MS/MS experiments (see Figure 3) and by injection of
sample 4 (A) and 8 (B) depicting the base peak MS chromatogram, the bioactivity chromatogram, the extractedion chromatogram of the observed estrogenic compounds, and their corresponding mass spectra. A complete overview of the detected estrogenic compounds per sample can be found in Table 1 in which the identified compounds, their molecular Table 1. Estrogenic Compounds Identified in Plastic Casings of Consumer Electronicsa retention time (min) Sample 1 Bisphenol A (BPA) 2,4-di-tertbutylphenol (2,4-DTBP)
C15H16O2
5.5
C14H22O
13.5
[M − H]− ESI (−)
[M − H]− APCI (−)
−1.3 ppm
−2.0 ppm
1.6 mSigma 1.3 ppm
5.0 mSigma 2.1 ppm
3.5 mSigma
5.8 mSigma
Sample 4 Bisphenol A (BPA)
C15H16O2
5.5
0.6 ppm
−2.0 ppm
Suspected BPA analog
C17H18O4
7.5
2.0 mSigma 2.9 ppm
5.0 mSigma 0.1 ppm
2.0 mSigma
3.2 mSigma
Sample 5 Bisphenol A (BPA)
C15H16O2
5.5
−2.0 ppm
−0.3 ppm
suspected BPA analog
C17H18O4
7.5
4 mSigma 2.0 ppm
1.3 mSigma −4.0 ppm
1.8 mSigma
6.1 mSigma
0.3 ppm
3.1 ppm
4.9 mSigma
1.3 mSigma
Sample 8 2,4-di-tertbutylphenol (2,4-DTBP)
C14H22O
13.5
Obtained mass accuracy (ppm error) and isotopic pattern fit (mSigma value) are shown for negative mode ESI and APCI detection. a
formula, retention time, ppm error, and mSigma value (isotopic pattern fit) are given for ESI and APCI in both positive and negative modes. Two estrogenic compounds observed in both samples 4 and 5 were assigned to BPA and a compound with the formula C17H18O4. BPA was also detected, but at lower intensity, in the bioactivity trace of sample 1. In addition to BPA and C17H18O4, samples 1 and 8 (Figure 2B) showed a bioactive component that was identified as 2,4-di-tertbutylphenol (2,4-DTBP). 2,4-DTBP is used as an intermediate in a variety of chemical syntheses, such as the manufacture of UV stabilizers and antioxidants, that are common additives in plastics.31 It has also been reported as an antifungal agent.32 This compound is classified as very toxic to aquatic life with long lasting effects.33 Alkylphenols are known to be endocrine disruptors, and 2,4DTBP has been recently described as an estrogenic compound.34,35 To the best of our knowledge the presence of 2,4-DTBP in plastics has not been reported yet and could be a source of introduction of this compound into the environment. For the compound with formula C17H18O4, an unequivocal chemical match could not be proposed. This formula gave 1780 possible results in the online database Chemspider, with
Figure 3. MS/MS spectra of the estrogenic compounds present in samples 4 and 8. A) 2,4-DTBP, B) suspected BPA analog (C17H18O4), and C) BPA. Main ion formulas of the parent compound and fragment ions are given together with mass error (ppm) and isotopic pattern fit (mSigma). 12389
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology Table 2. Flame Retardants Present in the Bioactive Samples
a Color red: compounds for which estrogenic activity has been reported. bColor dark gray: main flame retardants in terms of relative abundance; color light gray: flame retardants present as trace levels (estimated concentration below 1% w/w).
spectrometry method (that was used for preselection of samples). The main flame retardants detected with ambient mass spectrometry are shown in Table S-2, and all the detected flame retardants (main compounds and trace levels compounds) are given in Table 2. The presence of mixtures of flame retardants in the same plastic has been reported before28 and could originate from the use of synergistic flame retardant mixtures used to enhance the flame retarding properties of the material. The presence of flame retardant mixtures in plastics can, however, also be caused by cross-contamination from a previously produced plastic and/or the use of recycled plastic during manufacturing. Some flame retardants and related products, such as impurities, that were present in the samples, have been reported to exert some estrogenic activity (see Table S-2). This is the case of e.g., triphenyl phosphate (TPHP), that could be present as flame retardant, plasticizer, or impurity of other common flame retardants, such as resorcinol diphenyl phosphate or RDP.26,38 Nevertheless, none of the flame retardants correlated with the bioassay peaks indicating that bioactivity was not induced by these compounds in the tested extracts. This may result from the limited solubility of these compounds in aqueous media thereby preventing uptake into the cells and because of their relatively low bioactivity. In addition, the detected flame retardants may require bioactiva-
the authentic standards. Their retention times (±0.1 min) and spectra (main fragments and relative abundances) were identical to those of the authentic standards. For the compound with formula C17H18O4, the later retention time compared to BPA (7.5 min instead of 5 min) suggests a slightly lower polarity than the latter. The MS/MS spectra of C17H18O4 showed characteristic BPA fragments, such as the loss of a methyl group [BPA−CH3−H]− at m/z 211.07 and the loss of a phenol group [BPA−C6H5O−H]− at m/z 133.06. Furthermore, an ion fragment with the same formula as BPA appeared as an in-source fragment of C17H18O4 in APCI(−). This suggests a structural similarity with BPA. Presence of Flame Retardants and Estrogenicity. Flame retardants and related compounds, such as impurities, have been reported to exert a certain degree of estrogenicity (see Table S-2). Although flame retardants are present in the selected plastics, bioactivity could not be linked directly to these compounds. Nevertheless, the detected estrogenic compounds may originate from flame retardants in the form of impurities, byproducts, or degradation products during the plastic manufacturing process. To this end an in-depth target and nontarget screening was performed on the four bioactive samples to detect also low abundant flame retardants not initially detected with the direct probe ambient mass 12390
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology
Figure 4. Flame retardants and related products that are structurally similar to Bisphenol A and, therefore, could lead to the formation/presence of this compound as an impurity, byproduct, or degradation product.
tert-butyl phenol)40 may be also used in the manufacturing of phosphorus flame retardants. Indeed, sample 8, that showed an intense bioactive peak caused by 2,4-DTBP, contained resorcinol bis(diphenyl phosphate) (RDP) as main flame retardant. Sample 1 also showed a minor bioactive peak (10 times less intense in the bioassay) identified as 2,4-DTBP. Although in sample 1 a brominated oligomeric epoxy flame retardant and TBBPA were the main flame retardants detected, traces of PFRs were also present in this extract (see Table 2). To summarize, the LC-MS/nanofractionation platform allowed to detect three estrogenic compounds in 4 of the 8 plastic casings analyzed, namely, BPA, 2,4-DTBP, and a suspected BPA analog. Flame retardants that were present in the analyzed samples did not show any estrogenic response in the human cell-based bioassay. However, bisphenol A and the suspected analog could be present as an impurity of BDP, TBBPA, and TBBPA-based polymers. In general, we could conclude that plastic casings from consumer electronics contained estrogenic compounds. Consequently, these common consumer products could constitute a source of estrogenic contamination for human exposure indoors but also at primitive electronic waste recycling sites.
tion in order to become estrogenic, as it was reported for polybrominated biphenyls,39 and, therefore, will show limited or no bioactivity without metabolization. It is worth mentioning that given the fact that we studied the activity of individual compounds present in the plastics, which were separated via nanofractionation, we did not examine possible mixture effects. In this sense, although no activity was observed for flame retardants, their contribution on the overall estrogenicity of the product should not be completely ruled out, as possible mixture toxicity effects could take place. Nevertheless, some of the detected flame retardants and products are structurally related to BPA, sharing the same basic structure. Therefore, BPA and maybe the unidentified analog could also be present in the samples as an impurity (or degradation product) of these structurally related flame retardants. In fact, BPA has been reported to be a common impurity of TBBPA,16,18 which was present as the main flame retardant in the three samples showing BPA as a bioactive peak (1, 4, and 5). Other FRs and related products with a BPA-based structure were also present in the samples and could give rise to BPA (and to the suspect analog) as impurity, byproduct, or degradation product. BPA-related compounds present in the samples are summarized in Figure 4 and were di- and tribromobisphenol A, the organophosphate bisphenol A bis(diphenyl phosphate) (BDP), and a TBBPA-based polymer (CAS 135229-48-0). The latter is a brominated oligomeric epoxy flame retardant, considered a suitable replacement of the banned polybrominated diphenyl ethers with trade names F3100 and Pratherm EC20. Table S-3 shows the detected ions and structures of the polymer FR, that for the best of our knowledge has not been reported before. Finally, the presence of 2,4-DTBP is most probably related to other types of plastic additives (antioxidants, UV filters).31 However, other compounds from the same chemical class (e.g.,
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b03762. Samples of plastic casings from electronic products and the main flame retardant present and detected by ambient mass spectrometry, flame retardants present in the bioactive samples and estrogenicity (if reported), and MS detected ions and structures of the polymer FR (135229-48-0) (PDF) 12391
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology
■
(14) Bejgarn, S.; MacLeod, M.; Bogdal, C.; Breitholtz, M. Toxicity of leachate from weathering plastics: An exploratory screening study with Nitocra spinipes. Chemosphere 2015, 132, 114−119. (15) Lithner, D.; Nordensvan, I.; Dave, G. Comparative acute toxicity of leachates from plastic products made of polypropylene, polyethylene, PVC, acrylonitrile-butadiene-styrene, and epoxy to Daphnia magna. Environ. Sci. Pollut. Res. 2012, 19, 1763−1772. (16) Kang, J. H.; Kondo, F.; Katayama, Y. Human exposure to bisphenol A. Toxicology 2006, 226, 79−89. (17) Vandenberg, L. N.; Hauser, R.; Marcus, M.; Olea, N.; Welshons, W. V. Human exposure to bisphenol A (BPA). Reprod. Toxicol. 2007, 24, 139−177. (18) Geens, T.; Goeyens, K.; Covaci, A. Are potential sources for human exposure to bisphenol-A overlooked? Int. J. Hyg. Environ. Health 2011, 214, 339−347. (19) Yang, C. Z.; Yaniger, S. I.; Jordan, V. C.; Klein, D. J.; Bittner, G. D. Most plastic products release estrogenic chemicals: a potential health problem that can be solved. Environ. Health Perspect. 2011, 119, 989−996. (20) Bittner, G. D.; Yang, C. Z.; Stoner, M. A. Estrogenic chemicals often leach from BPA-free plastic products that are replacements for BPA-containing polycarbonate products. Environ. Health 2014, 13, 41. (21) Guart, A.; Wagner, M.; Mezquida, A.; Lacorte, S.; Oehlmann, J.; Borrell, A. Migration of plasticisers from Tritan and polycarbonate bottles and toxicological evaluation. Food Chem. 2013, 141, 373−380. (22) Ballesteros-Gómez, A.; Jonkers, T.; Covaci, A.; de Boer, J. Screening of additives in plastics with high resolution time-of-flight mass spectrometry and different ionization sources: direct probe injection (DIP)-APCI, LC-APCI, and LC-ion booster ESI. Anal. Bioanal. Chem. 2016, 408, 2945−53. (23) Harris, C. A.; Henttu, P.; Parker, M. G.; Sumpter, J. P. The Estrogenic Activity of Phthalate Esters In Vitro. Environ. Health Perspect 1997, 105, 802−811. (24) Kawamura, Y.; Ogawa, Y.; Nishimura, T.; Kikuchi, Y.; Nishikawa, J.; Nishihara, T.; Tanamoto, K. Estrogenic activities of UV stabilizers used in food contact plastics and benzophenone derivatives tested by the yeast two-hybrid assay. J. Health Sci. 2003, 49, 205−212. (25) Kitamura, S.; Suzuki, T.; Sanoh, S.; Kohta, R.; Jinno, N.; Sugihara, K.; Yoshihara, S.; Fujimoto, N.; Watanabe, H.; Ohta, S. Comparative Study of the Endocrine-Disrupting Activity of Bisphenol A and 19 Related Compounds. Toxicol. Sci. 2005, 84, 249−259. (26) Zhang, Q.; Lu, M.; Dong, X.; Wang, C.; Zhang, C.; Liu, W.; Zhao, M. Potential Estrogenic Effects of Phosphorus-Containing Flame Retardants. Environ. Sci. Technol. 2014, 48 (12), 6995−7001. (27) Jonker, W.; Lamoree, M. H.; Houtman, C. J.; Hamers, T.; Somsen, G. W.; Kool, J. Rapid activity-directed screening of estrogens by parallel coupling of liquid chromatography with a functional gene reporter assay and mass spectrometry. J. Chrom. A 2015, 1406, 165− 174. (28) Ballesteros-Gómez, A.; Brandsma, S. H.; de Boer, J.; Leonards, P. E. Direct probe atmospheric pressure photoionization/atmospheric pressure chemical ionization high-resolution mass spectrometry for fast screening of flame retardants and plasticizers in products and waste. Anal. Bioanal. Chem. 2014, 406 (11), 2503−12. (29) Rogers, J. M.; Denison, M. S. Recombinant cell bioassays for endocrine disruptors: development of a satably tranfected human ovarian cell line for the detectionof estrogenic and anti-estrogenic chemicals. In Vitro Mol. Toxicol. 2000, 13, 67−82. (30) Ionas, A. C.; Ballesteros-Gómez, A.; Leonards, P. E. G.; Covaci, A. Identification strategies for flame retardants employing time-offlight mass spectrometric detectors along with spectral and spectra-less databases. J. Mass Spectrom. 2015, 50, 1031−1038. (31) McRobb, F. M.; Kufareva, I.; Abagyan, R. In Silico Identification and Pharmacological Evaluation of Novel Endocrine Disrupting Chemicals That Act via the Ligand-Binding Domain of the Estrogen Receptor α. Toxicol. Sci. 2014, 141, 188−197. (32) Varsha, K. K.; Devendra, L.; Shilpa, G.; Priya, S.; Pandey, A.; Nampoothiri, K. M. 2,4-Di-tert-butyl phenol as the antifungal,
AUTHOR INFORMATION
Corresponding Author
*Phone: +31 20 59 83193. E-mail: a.m.ballesterosgomez@ gmail.com. Author Contributions
W.J. and A.B.-G. contributed equally to this work. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Ana Ballesteros-Gómez acknowledges the funding by The Netherlands Organization for Scientific Research (NWO) (VENI2014-722.014.003). Willem Jonker acknowledges the Dutch Technology Foundation STW for financial support, which is part of NWO, and which is partly funded by the Ministry of Economic Affairs. Project number: 12936. Art Kruithof (Department of Organic Chemistry, Vrije Universiteit Amsterdam) is thanked for his help with the FTIR measurements.
■
REFERENCES
(1) Allen, J. G.; McClean, M. D.; Stapleton, H. M.; Webster, T. F. Linking PBDEs in House Dust to Consumer Products using X-ray Fluorescence. Environ. Sci. Technol. 2008, 42, 4222−4228. (2) Abbasi, G.; Saini, A.; Goosey, E.; Diamond, M. L. Product screening for sources of halogenated flame retardants in Canadian house and office dust. Sci. Total Environ. 2016, 545-546, 299−307. (3) Ionas, A. C.; Ulevicus, J.; Ballesteros-Gómez, A.; Brandsma, S. H.; Leonards, P. E. G.; van de Bor, M.; Covaci, A. Children’s Exposure to Polybrominated Diphenyl Ethers (PBDEs) through Mouthing Toys. Environ. Int. 2016, 87, 101−107. (4) Schettler, T. Human exposure to phthalates via consumer products. Int. J. Androl. 2006, 29, 134−139. (5) Covaci, A.; et al. Human Exposure and Health Risks to Emerging Organic Contaminants. In The Handbook of Environmental Chemistry; Barceló, D., Ed.; Springer: 2011; Vol. 20, pp 243−305. (6) Ma, W. L.; Subedi, B.; Kannan, K. The Occurrence of Bisphenol A, Phthalates, Parabens and Other Environmental Phenolic Compounds in House Dust: A Review. Curr. Org. Chem. 2014, 18, 2182− 2199. (7) Coelho, S. D.; Sousa, A. C. A.; Isobe, T.; Tanabe, S.; Nogueira, A. J. A. Flame Retardants in Indoor Dust - A Review on the Levels of Polybrominated Diphenyl Ethers and Hexabromocyclododecanes. Curr. Org. Chem. 2014, 18, 2218−2230. (8) de Boer, J.; Ballesteros-Gomez, A.; Leslie, H. A.; Brandsma, S. H.; Leonards, P. E. G. Flame retardants: Dust - And not food - Might be the risk. Chemosphere 2016, 150, 461−464. (9) Ali, N.; Van den Eede, N.; Dirtu, A. C.; Neels, H.; Covaci, A. Assessment of human exposure to indoor organic contaminants via dust ingestion in Pakistan. Indoor air 2012, 22, 200−211. (10) Awasthi, A. K.; Zeng, X.; Li, J. Relationship between e-waste recycling and human health risk in India: a critical review. Environ. Sci. Pollut. Res. 2016, 23, 11509−11532. (11) Cuadra, S. A. N.; Linderholm, L.; Athanasiadou, M.; Jakobsson, K. Persistent organochlorine pollutants in children working at a wastedisposal site and in young females with high fish consumption in Managua, Nicaragua. Ambio 2006, 35, 109−116. (12) Lyche, J. L.; Gutleb, A. C.; Bergman, A.; Eriksen, G. S.; Murk, A. J.; Ropstad, E.; Saunders, M.; Skaare, J. U. Reproductive And Developmental Toxicity of Phthalates. J. Toxicol. Environ. Health, Part B 2009, 12, 225−49. (13) Kim, Y. R.; Harden, F. A.; Toms, L. L. L.; Norman, R. E. Health consequences of exposure to brominated flame retardants: A systematic review. Chemosphere 2014, 106, 1−19. 12392
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
Article
Environmental Science & Technology antioxidant bioactive purified from a newly isolated Lactococcus sp. Int. J. Food Microbiol. 2015, 211, 44−50. (33) ECHA European database. Available at http://echa.europa.eu/ information-on-chemicals/registered-substances (accessed 28-042016). (34) Tollefsen, K.-E.; Nilsen, A. J. Binding of alkylphenols and alkylated non-phenolics to rainbow trout (Oncorhynchus mykiss) hepatic estrogen receptors. Ecotoxicol. Environ. Saf. 2008, 69, 163−172. (35) Creusot, N.; Budzinski, H.; Balaguer, P.; Kinani, S.; Porcher, J.M.; Aït-Aïssa, S. Effect-directed analysis of endocrine-disrupting compounds in multi-contaminated sediment: identification of novel ligands of estrogen and pregnane X receptors. Anal. Bioanal. Chem. 2013, 405, 2553−2566. (36) Summary of data for chemical selection-Diphenolic acid. National Toxicology Program. U.S. Department of Health and Human Services. Available at https://ntp.niehs.nih.gov/ntp/htdocs/ chem_background/exsumpdf/diphenolicacid_508.pdf (accessed 2504-2016). (37) Blair, R. M.; Fang, H.; Branham, W. S.; Hass, B. S.; Dial, S. L.; Moland, C. L.; Tong, W.; Shi, L.; Perkins, R.; Sheehan, D. M. The Estrogen Receptor Relative Binding Affinities of 188 Natural and Xenochemicals: Structural Diversity of Ligands. Toxicol. Sci. 2000, 54, 138−153. (38) Ballesteros-Gomez, A.; Aragon, A.; Van den Eede, N.; de Boer, J.; Covaci, A. Impurities of Resorcinol Bis(diphenyl phosphate) in Plastics and Dust Collected on Electric/Electronic Material. Environ. Sci. Technol. 2016, 50, 1934−1940. (39) van Lipzig, M. M. H.; Commandeur, J. N.; de Kanter, F. J. J.; Damsten, M. C.; Vermeulen, N. P. E.; Maat, E.; Groot, E. J.; Brouwer, A.; Kester, M. H. A.; Visser, T. J.; Meerman, J. H. N. Bioactivation of Dibrominated Biphenyls by Cytochrome P450 Activity to Metabolites with Estrogenic Activity and Estrogen Sulfotransferase Inhibition Capacity Chem. Chem. Res. Toxicol. 2005, 18, 1691−1700. (40) P-TERT-BUTYLPHENOL CAS No: 98-54-4 EINECS No: 202-679-0. SUMMARY RISK ASSESSMENT REPORT. Norwegian Pollution Control Authority (2008). Available at http://echa.europa. eu/documents/10162/296c0932-cf7a-4d90-9051-ba7b224a32b3 (accessed 06.06.2016).
12393
DOI: 10.1021/acs.est.6b03762 Environ. Sci. Technol. 2016, 50, 12385−12393
