ELSEVIER
Contents lists available at ScienceDirect
Regulatory Toxicology and Pharmacology
journal homepage: www.elsevier.com/locate/yrtph
Regulatory Toxicology and Pharmacology
Endocrine activity of alternatives to BPA found in thermal paper in Switzerland
CrossMark
Daniela M. Goldingera,1, Anne-Laure Demierre a,*,1, Otmar Zoller b, Heinz Rupp b, Hans Reinhard b, Roxane Magnin b, Thomas W. Becker, Martine Bourqui-Pittetª
a Federal Office of Public Health, Division Chemical Products, 3003 Bern, Switzerland
b Federal Food Safety and Veterinary Office, 3003 Bern, Switzerland
” PhaToCon (Pharm/Tox Concept) GmbH, 82152 Martinsried, Germany
ARTICLE INFO
Article history: Received 21 October 2014 Available online 8 January 2015
Keywords:
BPA alternatives Thermal paper Endocrine activity H295R steroidogenesis assay VirtualToxLab™
ABSTRACT
Alternatives to bisphenol A (BPA) are more and more used in thermal paper receipts. To get an overview of the situation in Switzerland, 124 thermal paper receipts were collected and analyzed. Whereas BPA was detected in most samples (n = 100), some alternatives, namely bisphenol S (BPS), Pergafast® 201 and D-8 have been found in 4, 11 and 9 samples respectively. As no or few data on their endocrine activity are available, these chemicals and bisphenol F (BPF) were tested in vitro using the H295R steroidogenesis assay. 17ß-Estradiol production was induced by BPA and BPF, whereas free testosterone production was inhibited by BPA and BPS. Both non-bisphenol substances did not show significant effects. The binding affinity to 16 proteins and the toxicological potential (TP) were further calculated in silico using Virtu- alToxLab™. TP values lay between 0.269 and 0.476 and the main target was the estrogen receptor B (84.4 nM to 1.33 µM). A substitution of BPA by BPF and BPS should be thus considered with caution, since they exhibit almost a similar endocrine activity as BPA. D-8 and Pergafast® 201 could be alternatives to replace BPA, however further analyses are needed to better characterize their effects on the hormonal system.
@ 2015 Elsevier Inc. All rights reserved.
1. Introduction
Bisphenol A (BPA) is a high production volume chemical. It is widely used as monomer in the manufacturing of polymer prod- ucts such as polycarbonate, epoxy resins and also as an additive in plastics. Additionally, BPA is found in the paper industry as color developer in thermal paper (Geens et al., 2012a).
Human exposure to BPA is widespread and many data are avail- able suggesting adverse effects at low-dose. Its association with several diseases is frequently discussed (Geens et al., 2012a; Vandenberg et al., 2010, 2013), and its endocrine activity has been widely investigated, including e.g. effects on steroidogenesis (Zhang et al., 2011). However, many uncertainties remain and con- troversial discussions are still ongoing.
Due to the ubiquity of BPA, its hormonal activity and the related uncertainties, EFSA recently focused a Scientific Opinion on this
substance (EFSA, 2013, 2014). There, EFSA evaluated BPA exposure and the risks for human health. In most cases, diet was found to be the main source of exposure, whereas thermal paper was the sec- ond source (EFSA, 2013). Several reports from the USA (US EPA, 2014), Denmark (Lassen et al., 2011) and Sweden (KEMI, 2012) also identified thermal paper as a source of exposure to BPA. Moreover, some countries or states such as Japan (2001), Taiwan (2011) and Connecticut, USA (2015) prohibited the use of BPA in thermal paper following the precautionary principle. In June 2014, France submitted a restriction proposal to the European Chemicals Agency (ECHA) to ban the use of BPA in thermal paper in concentrations equal or higher than 0.02% (ECHA, 2014). Accordingly, safer alter- natives to replace BPA are required.
At the beginning of 2014 the US EPA published a final report on “Bisphenol A alternatives in thermal paper” (US EPA, 2014), identi- fying nineteen substances as potential BPA substitutes. These nine- teen substances were selected according to their physical and chemical properties and/or because they are already commercially used. No clearly safer alternative to BPA could be identified in the report, as only limited toxicological information on these replace- ment substances is available. Analyses or structural similarities of most of these alternatives led to some doubts concerning their
* Corresponding author at: Federal Office of Public Health, Division Chemical Products, Schwarzenburgstrasse 165, CH-3003 Bern, Switzerland. Fax: +41 58 464 90 34.
E-mail address: anne-laure.demierre@bag.admin.ch (A .- L. Demierre).
1 These authors contributed equally to this work.
innocuity. Indeed, it was reported that most of them have moder- ate to high probability to impact human health or aquatic toxicol- ogy endpoints.
To obtain an overview of the Swiss situation, we performed a market analysis of thermal paper receipts in the region of Bern, Switzerland. Between September 2013 and January 2014, 124 ther- mal paper receipts were randomly sampled and analyzed. We focused the follow-up studies on the alternative substances found during this Swiss market analysis. Another potential alternative bisphenol F (BPF) has been included to these analyses. The studies included BPA as control compound, bisphenol S (BPS), BPF, D-8 (also known as WinCon-8) and Pergafast® 201 (Table 1).
Data about these alternatives are scarce, particularly for the non-bisphenols, and are mostly limited to in vitro studies. BPS has been shown to bind to the estrogen receptor (ER) in vitro (Laws et al., 2006; Yamasaki et al., 2004), elicit estrogen induced gene transcription (Chen et al., 2002; Nishihara et al., 2000) and induce cell proliferation in MCF7 cancer cells (Kuruto-Niwa et al., 2005). There is only one in vivo uterotrophic study available sug- gesting a potential for estrogenic activity (Yamasaki et al., 2004).
The available in vitro and in silico assays indicate that BPF can bind to estrogen receptors (ERs) (Blair et al., 2000; Coleman et al., 2003; Yamasaki et al., 2004), trigger estrogen induced gene transcription (Chen et al., 2002; Hashimoto and Nakamura, 2000; Miller et al., 2001), induce progesterone receptors (PgR) (Kitamura et al., 2005; Perez et al., 1998), and induce cell prolif- eration in MCF7 cancer cells (Coleman et al., 2003; Stroheker et al., 2004). Additionally, BPF has been shown to exhibit in vitro androgenic and anti-androgenic effects (Cabaton et al., 2009; Kitamura et al., 2005; Stroheker et al., 2004). BPF was shown to have estrogenic and anti-estrogenic activity in some in vivo studies with female rats (Akahori et al., 2008).
For D-8 there is only limited evidence of endocrine activity. D-8 was negative for estrogenic activity in two ER binding assays and one competitive ER binding assay (Terasaki et al., 2007), and posi- tive for anti-estrogenicity in a competitive binding assay in the presence of 17ß-estradiol (Kuruto-Niwa et al., 2005).
There is only one in vitro study available suggesting that Per- gafast® 201 is non-estrogenic with a relative potency substantially low compared to 17ß-estradiol (US EPA, 2014).
The H295R steroidogenesis assay is part of the Conceptual Framework of the OECD for the testing and assessment of endo- crine disrupting chemicals (OECD, 2011). This assay allows the detection of change in the level of both estradiol and testosterone. The alteration in the concentration of hormones can result from different interactions of the chemicals with steroidogenic function, such as binding to an enzyme involved in the steroidogenesis path- way, modulating the steroid metabolism, or affecting the transcrip- tion of the enzymes, for example by binding the chemical to hormone receptor.
In this study, we evaluated in a first step which alternatives to BPA are present in thermal paper receipts on the Swiss market. Secondly the found chemicals and BPF were tested in vitro for their influence on the 17ß-estradiol and free testosterone level using the H295R steroidogenesis assay under GLP conditions. In parallel, binding affinity to 16 proteins involved in the hormonal system and the toxicological potential of the substances were determined using the in silico tool VirtualToxLab™M.
2. Materials and methods
2.1. Chemicals and materials
D-8 was obtained from Connect Chemicals GmbH (Ratingen, Germany) and Pergafast® 201 from BASF (Bradford, Great Britain). For the chemical analysis bisphenol A-propane-D6 (BPA-D6) was purchased from Cambridge Isotope Laboratories (Tewksbury, USA) and bisphenol S (BPS) was from TCI Europe (Zwijndrecht, Bel- gium). Methanol, LC-MS Chromasolv®, and the following sub- stances for the H295R steroidogenesis assay such as bisphenol A (BPA), bisphenol S (BPS), bisphenol F (BPF), forskolin and prochlo- raz were purchased from Sigma Aldrich (St. Louis, USA). Ultrapure water was obtained from an ElgaPurelab ultra water purification system (Labtec Services, Villmergen, Switzerland).
2.2. Market analysis in Switzerland
Thermal paper receipts (cashier receipts, ATM receipts, parking tickets, bus tickets etc.) were randomly sampled in Switzerland between September 2013 and January 2014, mostly in the Bern
| CAS # | Chemical name | Substance name | Molecular formula | Structure |
|---|---|---|---|---|
| 80-05-7 | 2,2-Bis(4-hydroxyphenyl)propane | Bisphenol A (BPA) | C15H16O2 | HO OH |
| 620-92-8 | Bis(4-hydroxyphenyl)methane | Bisphenol F (BPF) | C13H12O2 | HO OH |
| 80-09-1 | Bis(4-hydroxyphenyl)sulfone | Bisphenol S (BPS) | C12H10O4S | HỘ OH |
| 232938-43-1 | N-(p-Toluenesulfonyl)-N'- (3-p-toluenesulfonyloxyphenyl)urea | Pergafast® 201 | C21 H20N2O6S2 | |
| 95235-30-6 | 4-Hydroxyphenyl-4'-isopropoxyphenyl-sulfone | D-8 (WinCon-8) | C15H16O4S | OH |
area. A subsample of each receipt was tested on a hotplate at 140 ℃ to confirm it was thermal paper. Thermal paper samples (n = 124) were wrapped in aluminium foil and kept in the dark until further processing. Samples were first analyzed for their BPA content. All receipts without BPA (n = 23) and 14 randomly selected receipts containing BPA were screened for their content of possible alternative substances. The main substances detected in the screening method were quantified during a further analysis step.
2.3. Extraction of thermal paper
The extraction was carried out according to Geens et al. (2012b). About 25 mg of thermal paper were cut into small strips, accurately weighted and suspended in 2 mL of methanol. Extraction was per- formed by two cycles of vortex (30 s) followed by sonication (10 min). The solution was diluted 50 times with methanol (dilu- tion 1). This solution was diluted with water by a factor of 2 (1 +1) and used for the screening assay. Dilution 1 was further diluted by a factor of 1000 with methanol for the quantitation of Pergafast® 201, D-8 and BPS. For quantitation of BPA 0.05 mL of dilution 1 was mixed with 0.05 mL of BPA-D6 internal standard (IS) solution. If the concentration was above the calibration range, dilution 1 was further diluted by a factor of two and reanalyzed.
2.4. Screening of chemicals by liquid chromatography high resolution mass spectrometry (LC-HRMS)
The chromatographic system consisted of a Shimadzu Promi- nence binary gradient system (Shimadzu, Reinach, Switzerland), degasser, auto sampler and column heater. Chromatographic separations were performed on a Kromasil C18 (125 x 2 mm, 3.5 um particle size) analytical column (Macherey-Nagel, Düren, Germany). The flow rate was 0.2 mL/min and the column tem- perature maintained at 30 ℃. A gradient program was used start- ing with 10% methanol in water and ramped linearly over the course of 12 min to 90% methanol, held for 5 min at this condition, then re-equilibrated for 2.5 min at 90% water. Injection volume was 5 uL resulting in 500 pg of standards on column.
Mass spectrometric detection was achieved with a Bruker maXis 4G Qq-TOF mass spectrometer (Bruker, Bremen, Germany), equipped with an electrospray ionization interface operated in negative ion mode. Source parameters were: plate offset 500 V, capillary voltage 4.5 kV, dry temperature 200 ℃, nebulizer gas pressure 150 kPa and nitrogen dry gas flow rate 8 L/min. Internal calibration was achieved by incorporating sodium formate solution as calibrant at the beginning of every run with a loop injection. For instrument control, data acquisition and processing, Compass 1.5 and TargetAnalysis 1.3 were used.
Identification of compounds was accomplished by high resolu- tion mass determination (deviation ≤1 mDa allowed); positives were additionally verified by comparing retention time to the stan- dard compound.
The screening method was designed to detect 17 of the 19 alter- native substances mentioned in the report of US EPA (2014) (as no adequate information was available for 2 patented substances) and 13 additional bisphenols. Table S1 of Supplementary data lists the targeted analytes under investigation.
2.5. Quantitation of chemicals by liquid chromatography tandem mass spectrometry (LC-MS/MS)
The chromatographic system consisted of a Shimadzu UFLC binary gradient system (Shimadzu, Reinach, Switzerland) with pumps LC-30AD, vacuum degasser DGU-20A, thermostated col- umn compartment CTO-20 and autosampler SIL-30A. Separations
were performed on a Kinetex XB-C18, 100A (100 x 2.1 mm, 1.7 um particle size) with a precolumn (Phenomenex, Torrance, USA). The injection volume was 1.0 uL and the column tem- perature maintained at 50 ℃. A gradient program was used start- ing with 50% methanol in water and ramped linearly over the course of 3 min to 95% methanol, held for 1 min at this condition, then re-equilibrated for 3 min at 50% methanol. The flow rate was 0.3 mL/min for the BPA determination and 0.4 mL/min for the determination of Pergafast® 201, BPS and D-8.
MS/MS analysis was carried out on an API 5000 system (AB Sciex, Framingham, USA) equipped with a turboIon spray source (ESI). The following instrumental settings were used: source tem- perature 600 ℃, curtain gas 31, collision gas 7, gas-1 50, gas-2 70, ionspray -4500 V. Measurements were carried out using mul- tiple reaction monitoring (MRM) in negative mode. The used MRM transitions and dwell times are listed in Table S2 of Supplementary data.
BPA was quantified using the internal standard method and lin- ear regression. A five point calibration curve was used with calibra- tion points between 145 and 2900 ng/mL, corresponding to about 1.16-23.2 mg of BPA in the paper using our usual procedure. For quantification of Pergafast® 201, D-8 and BPS external calibration was used.
Individual stock solutions of 1 mg/mL for each compound were used. Mixed working solutions of 0.5, 1.0, 2.5, 5 and 10 ng/ml in methanol were used for the analysis, corresponding to a range of about 2-40 mg/g substance in the paper. As Pergafast® 201 was often unstable in methanolic solution at room temperature, stan- dards and extracts for determination of Pergafast® 201 were always freshly produced and processed in less than 20 h.
In order to confirm the concentration of Pergafast® 201 in the thermal papers, this substance was additionally quantified with an LC/UV method. This method and the performance of the quan- titation are detailed in Supplementary data.
2.6. H295R steroidogenesis assay
Steroidogenesis assay was performed under GLP conditions, fol- lowing the OECD TG 456 (H295R Steroidogenesis Assay). It has been done at MDS (Molecular Diagnostic Services) Inc., San Diego, USA, under coordination and supervision of PhaToCon GmbH, Martinsried, Germany.
The effects of the substances on the level of estradiol and testos- terone were tested in the H295R human adrenocortical carcinoma cell line (ATCC No. CRL-2128, Manassas, USA) as previously described (Hecker et al., 2011; OECD, 2011).
In brief, cells were cultured in T75 tissue culture flasks (Ther- mofisher Cat #156499, Waltham, USA) at 37 ℃ and 5% CO2. Cells were seeded in 48-well plates (Corning Cat #353078, Tewksbury, USA) at a density of 106 cells per well and incubated at 37 ℃ and 5% CO2. After 24 h, cells were exposed in triplicate for 48 h to var- ious concentrations of the test substances dissolved in 0.1% DMSO. Control wells contained the same amount of DMSO (0.1%) as exposed cells. Concentrations of BPA, BPF, BPS, D-8 and Pergafast® 201 were 0.1, 0.3, 1, 3, 10, 30 and 100 uM. Forskolin (an inducer of steroid hormone production) and prochloraz (an inhibitor) were used as positive controls at concentration of 1 and 10 uM for for- skolin, and 0.1 and 1 µM for prochloraz. After 48 h of exposure, medium was collected from each well and stored at -80 ℃ for hor- mone analysis. Complete medium was added back to each well and cell viability was analyzed using CellTiter 96® AQueous One Solu- tion Cell Proliferation Assay (Promega Cat G3581, Madison, USA).
17-Estradiol was measured using the Estradiol Ultrasensitive ELISA (ALPCO Cat 20ESTHUU-E01, Salem, USA). The analytical sensitivity of the assay was 1.4 pg/mL, calibration range was 0- 200 pg/mL. All data generated were within the validated range of
the assay and met the validity criteria. Samples were diluted in the range of 1:9-1:45 to reach the validated range of the assay. Free testosterone was measured using the Coat-a-Count Free Testos- terone solid-phase 125I RIA (Siemens Healthcare Diagnostics, Cat #TKTF2, Tarrytown, USA). The analytical sensitivity of the assay was 0.15 pg/mL, calibration range was 0.55-50 pg/mL. All data generated were within the validated range of the assay and met the validity criteria. Data processing and statistical analysis of estradiol and testosterone values was performed as described below.
The positive controls forskolin and prochloraz behaved as expected. Data can be found in Table S3 of Supplementary materials.
2.7. VirtualToxLab™M
VirtualToxLab™ is an in silico tool which was used to predict the endocrine and metabolic disruption potential of BPA, BPF, BPS, D-8 and Pergafast® 201. It calculates the toxic potential (TP) and the binding affinity (binding constant K) of any molecule to 16 pro- teins: 10 receptors (androgen, estrogen a, estrogen ß, glucocorti- coid, liver X, mineralocorticoid, progesterone, peroxisome proliferator-activated receptor y (PPARy), thyroid a and thyroid B), 4 members of the cytochrome P450 enzyme family (1A2, 2C9, 2D6 and 3A4), 1 transcription factor (aryl hydrocarbon receptor) and 1 potassium ion channel (hERG). The VirtualToxLab™ concept is described in Vedani et al. (2014).
2.8. Statistical analysis
Hormone data from the H295R steroidogenesis assay was illus- trated graphically with GraphPad® Prism 5 (GraphPad Software, San Diego, USA). Due to the small number of replicates, normality and variance were evaluated on the combined dataset for the three assays. Data distribution for normality was assessed with the Kol- mogorov-Smirnov test and the variance homogeneity with the Bartlett test. Differences between treatments were assessed by analysis of variance (ANOVA one-way) followed by Dunnett’s test to compare treatment means with respective controls. If the data was not normally distributed, differences between treatments were assessed by the Kruskal-Wallis test followed by Dunn’s mul- tiple comparison test. Results are given as mean + standard devia- tion. Differences were considered significant at p < 0.05.
3. Results
3.1. Market analysis in Switzerland
In total 124 thermal paper receipts were analyzed. All receipts contained only one single developer substance in relevant amounts. The results are summarized in Table 2. BPA was found most often (range: 5.6-30.4 mg/g), and only three alternative sub- stances namely BPS, Pergafast® 201 and D-8 were detected in the range of 3.3-13.2 mg/g. The papers containing D-8 contained also traces of BPS in the range of 0.01-0.13 mg/g.
| Chemical | Occurrence | Range conc. (mg/g) | Median conc. (mg/g) | Mean conc. (mg/g) | |
|---|---|---|---|---|---|
| n | (%) | ||||
| BPA | 100 | (81) | 5.6-30.4 | 14.5 | 13.5 |
| BPS | 4 | (3) | 8.3-12.6 | 10.0 | 10.2 |
| Pergafast® 201 | 11 | (9) | 3.3-8.2 | 4.6 | 5.4 |
| D-8 | 9 | (7) | 3.4-13.2 | 12.0 | 11.2 |
3.2. Cytotoxicity of BPA and its alternatives
Cytotoxicity was not observed in most of the tested concentra- tions. Only the highest concentration (100 uM) showed a sig- nificant decrease in viability for BPA (23% ± 8.3%), Pergafast® 201 (43% ± 6.1%) and D-8 (26% + 5.0%) (Table 3).
3.3. Effects on steroidogenesis in vitro
A significant increase of 17ß-estradiol concentration was seen for BPA and BPF (Fig. 1, Table 4). Despite a 23% + 8.3% drop in via- bility in the 100 uM treatment with BPA (Table 3), a statistically significant increase of 17ß-estradiol level was observed in a dose- dependent manner. Thus, the effect at 100 µM was taken into account when assessing the overall response of BPA. Although the lowest observed effect concentration (LOEC) for BPA and BPF were both at 30 uM, BPF seemed to be more potent than BPA, since the increase in 17ß-estradiol concentration was ~15% higher at 30 μΜ.
Overall, BPS, Pergafast® 201 and D-8 did not show any sig- nificant effects on 17ß-estradiol level. Significant effects were only seen at concentrations that did not meet the viability requirements (Fig. 1, Table 3, Table 4).
Concerning effects on free testosterone level, a decrease was seen for BPA and BPS, whereas no significant effects were seen for BPF and D-8. The LOEC for BPA was observed at 1 µM and for BPS at 30 µM, indicating that BPA is more potent than BPS (Fig. 2, Table 4). A significant effect for D-8 was only seen at the highest concentration that did not meet the viability requirements (Fig. 2, Table 4). A significant decrease of free testosterone level was observed with Pergafast® 201. This substance was not consid- ered as an inhibitor though, since these observations were not dose-dependent and were near background level (Fig. 2, Table 4).
In conclusion, BPA and BPF were found to increase the level of 17ß-estradiol, and BPA and BPS were reported to decrease the free testosterone concentration. BPS, Pergafast® 201 and D-8 were shown to have no effect on the 17-estradiol level. The free testos- terone concentration was not significantly affected by BPF, Per- gafast® 201 and D-8.
3.4. VirtualToxLab™
Using VirtualToxLab™ we calculated the toxic potential (TP) and the binding affinities of BPA, BPF, BPS, Pergafast® 201 and D- 8 to 16 proteins (Table 5). The TP was derived from the normalized binding affinities towards the 16 target proteins. The values range from 0 (none) to 1 (extreme) and could be interpreted as toxic alert. The TP values calculated for BPS, D-8, BPF and BPA lay between 0.380 and 0.476, showing a moderate risk of binding the proteins. Only Pergafast® 201 has a low risk of binding, with a TP value of 0.269 (Vedani et al., 2014).
The main target for BPA, BPF, BPS and D-8 was the estrogen receptor ß with binding affinity values ranging from 84.4 nM to
| μΜ | BPA | BPF | BPS | Pergafast® 201 | D-8 |
|---|---|---|---|---|---|
| 0.1 | 101 ± 2.5 | 106 ± 3.2 | 106 ± 2.2 | 104 ± 3.4 | 105 ± 4.6 |
| 0.3 | 105 ± 0.2 | 111 ± 3.9 | 109 ± 2.7 | 103 ± 3.5 | 105 ± 2.8 |
| 1 | 104 ±2.2 | 110± 1.0 | 106 ± 0.9 | 103 ± 5.7 | 106 ± 6.5 |
| 3 | 104 ± 2.4 | 111 ± 3.4 | 110 ±2.2 | 103 ± 1.7 | 105 ± 6.8 |
| 10 | 104±1.2 | 110 ± 5.0 | 107 ± 6.6 | 101 ± 5.0 | 103 ± 7.3 |
| 30 | 102 ± 2.7 | 108 ± 7.9 | 104 ±3.3 | 99 ± 3.0 | 102 ± 2.7 |
| 100 | 77 ± 8.3 | 105 ± 6.8 | 94 ± 6.7 | 57 ± 6.1 | 74 ±5.0 |
3
Average Relative Change
2
*
*
1
T
T
0
0
0.1 µM
0.3 LM
1 µM
3 µM
10 µM
30 μΜ
100 µM
Bisphenol A Concentration
3
*
3
Average Relative Change
Bisphenol A
Average Relative Change
Bisphenol A
L
Bisphenol F
Bisphenol S
*
2
a
2
A
2 1
1
T
1
1
T
T
T
A
T
1
T
L
L
S
L
1
1
0
0
0 µM
0.1 µM
0.3 µM
1 µM
3 µM
10 μM
30 μM
100 μΜ
0 µM
0.1 µM
0.3 µM
1 µM
3 µM
10 μM
30 µM
100 µM
Substance Concentration
Substance Concentration
3
Bisphenol A
3
Average Relative Change
Bisphenol A
Pergafast 201
Average Relative Change
D-8
2
2
2
2
LL
1
I
T
.
T
L
H
1
S
T
=
E
1
*
T
T
¥
E
L
T
L
0
0
0 μΜ
0.1 µM
0.3 µM
1 µM
3 µM
10 μΜ
30 µM
100 µM
0 μM
0.1 µM
0.3 µM
1 μM
3 µM
10 μΜ
30 μM
100 μM
Substance Concentration
Substance Concentration
1.33 µM. PPARy was the main target for Pergafast® 201 with a binding affinity of 22 uM.
4. Discussion
4.1. Market analysis in Switzerland
Currently BPA is the most commonly used color developer in thermal paper. Many studies investigated the concentration of BPA in such paper (Biedermann et al., 2010; Environmental
Working Group (EWG), 2010; Geens et al., 2012a; KEMI, 2012; Lassen et al., 2011; Liao and Kannan, 2011; Lu et al., 2013; Mendum et al., 2011; US EPA, 2014). In these studies BPA was found with a detection frequency of 44-100% with a concentration of up to 28 mg/g. Our BPA results (frequency of 81% with a range of 6-30 mg/g) are in line with the detection frequencies and concen- trations found worldwide.
In recent years, BPA has increasingly fallen into disrepute and as a consequence, alternative substances have been developed for thermal paper. BPA was not found in 19% of the thermal papers col-
| 17ß-Estradiol | Free testosterone | |||||
|---|---|---|---|---|---|---|
| Overall response | LOEC (µM) | Max. change | Overall response | LOEC (µM) | Max. change | |
| Bisphenol A | Inducer® | 30 | 1.85 | Inhibitor | 1 | 0.29 |
| Bisphenol F | Inducer | 30 | 2.80 | None | – | – |
| Bisphenol S | None | – | – | Inhibitor | 30 | 0.33 |
| Pergafast® 201 | None | – | – | None | – | – |
| D-8 | None | – | – | None | – | – |
* The 100 µM BPA treatment is taken into account as a statistically significant increase in 17B-estradiol was observed despite the low viability of the treated cells.
lected in our study. Recently some Swiss retailers announced stop- ping the use of thermal papers containing BPA. Therefore, it was not astonishing to find thermal papers with alternatives. BPS has been detected in 3% of the collected papers, in a range of 8- 13 mg/g, which is comparable to the concentrations found in the study of Liao et al. (2012).
Our study is one of the first to find substitutes of BPA other than bisphenols in thermal paper, i.e. D-8 and Pergafast® 201. Concen- trations of D-8 and BPS were in a similar range, reflecting the struc- tural similarity of these 2 substances. The average concentration of Pergafast® 201 is a factor of 2.5 lower than the average BPA con- centration. The concentration of this substance has been deter- mined by 2 different methods and the stability of the substance in thermal paper has been confirmed by repetition of the analysis after 3 months, obtaining similar results (data not shown). This corresponds to the information in the NICNAS report (NICNAS, 2004), indicating that the concentration should be less than 10 mg/g in the end product.
Traces of BPS have been detected in all thermal papers contain- ing D-8. As this substance is the isopropylether of BPS, the traces found could be an impurity or decomposition product of technical D-8.
Thermal papers are recycled (Terasaki et al., 2007). Therefore, as it has been shown for BPA, these substances can also find their way back to our daily life in the form of other papers such as journals or toilet papers (Liao and Kannan, 2011). Accordingly, not only ther- mal papers but also other kinds of papers should be in the scope of further investigations. Furthermore the probability that these substances end up in the aquatic environment is high. Due to the suspected risks of most BPA alternatives concerning environmental endpoints, including for D-8 and Pergafast® 201 (US EPA, 2014), the consequences of thermal paper recycling for aquatic organisms should be evaluated.
4.2. Effects on steroidogenesis in vitro
BPA and BPF led to an increase in 17-estradiol concentration. Furthermore all tested bisphenols showed a decrease in free testosterone level. This decrease was significant for BPA and BPS, but not for BPF, probably due to the high standard deviation found for this last substance. Mostly one out of three replicates was not in line with the other two therefore leading to a higher standard deviation than expected. This observation however was more related to the assay performance and the study design than to the material and is not regarded as biologically relevant. The effects on steroidogenesis of BPA have previously also been inves- tigated in the H295R assay (Rosenmai et al., 2014; Zhang et al., 2011) showing the same tendencies as we found in our study. In addition, Rosenmai et al. (2014) also investigated the effects of BPF and BPS on steroidogenesis. Our results are in line with theirs. They have also looked at the effect of BPA and its analogs on the other hormones intermediates of the steroidogenesis pathway.
Interestingly, they observed a significant increase of 17a-OH pro- gesterone level with BPS, whereas BPA did not affect this hormone (Rosenmai et al., 2014). In the present study, we only focused on testosterone and 17ß-estradiol levels, as the assay is only validated for these two hormones (OECD, 2011).
The metabolic capability of H295R cell line is unknown, but it is probably quite limited. Accordingly, substances that need to be metabolically activated to show endocrine activity could be missed in this assay (OECD, 2011). BPS was shown to be negative for estro- genic activity in the E-screen assay without metabolic activation (Hashimoto et al., 2001). However, after metabolic activation estrogenic activity could be seen. Therefore, it is possible that BPS first needs metabolic activation to elicit estrogenic activity in the H295R steroidogenesis assay. This issue has to be further inves- tigated as metabolic activation can take place in the human body.
Up to now there is only limited data available on the endocrine activity of D-8 and Pergafast® 201. It was for the first time that the H295R steroidogenesis assay was conducted with these two substances.
So far, there is only one in vitro study available showing that Pergafast® 201 is non-estrogenic (US EPA, 2014), and it is support- ed by our analysis. However, we observed a significant decrease of free testosterone level for the concentrations of 1-10 uM, but not at 30 µM. Although these values were significant, they were weak- ened by uncertainties, since they are in the range of variations of the conducted assay. This interpretation is also supported by the fact that no toxic potential is found with VirtualToxLab™. There- fore, there is no indication that Pergafast® 201 does exhibit hor- monal activity. However, steroidogenesis could be affected by several mechanisms such as binding to pathway enzymes or modulation of metabolism. Thus further tests would be required in order to make a final decision on this issue.
Although D-8 is structurally related to BPS, neither effect on the concentration of 17ß-estradiol nor free testosterone was found, suggesting that this substance does not influence steroidogenesis. D-8 was found to be negative for estrogenic activity in a study con- ducted by Terasaki et al. (2007). They also showed that D-8 is anti- estrogenic. However, this cannot be supported by our results.
As mentioned above, the metabolic activity is very limited in this test system. Therefore, it could not be excluded that a metabolic activation of Pergafast® 201 and D-8 would lead to estrogenic activity.
4.3. VirtualToxLab™M
This part of the study was conducted to support our findings in the in vitro H295R steroidogenesis assay. Our prediction showed that the main target for all compounds except Pergafast® 201 was the estrogen receptor ß (ERB). These findings are supported by studies found in the literature, where the affinity of BPA to the ERß is stronger than for ERa (Kolsek et al., 2014). In addition, it was also shown that both BPF and BPS are positive for ER binding
2.0
Average Relative Change
1.5
1.0
*
王
I
*
0.5
=
*
E
*
I
*
0.0
0
0.1 µM
0.3 µM
1 µM
3 µM
10 μΜ
30 µM
100 μΜ
Bisphenol A Concentration
2.0
Bisphenol A
Bisphenol A
Average Relative Change
Bisphenol F
2.0
Average Relative Change
Bisphenol S
1.5
1.5
1.0
1.0
I
1
5
*
-
0.5
0
.5
*
L
I
I
I
0.0
0.0
0 μ.Μ
0.1 μ.Μ
0.3 µM
1 μ.Μ
3 µM
10 μ.Μ
30 μ.Μ
100 μΜ
0 µM
0.1 µM
0.3 µM
1 μ.Μ
3 µM
10 μΜ
30 µM
100 μM
Substance Concentration
Substance Concentration
2.0
Bisphenol A
Bisphenol A
Pergafast 201
2.0
Average Relative Change
Average Relative Change
D-8
1.5
1.5
1.0
*
*
1.0
5
A
1
L
1
王
-
E
*
*
0.5
5
I
0.0
0.0
0 μΜ
0.1 µM
0.3 µM
1 µM
3 µM
10 µM
30 µM
100 µM
0 μ.Μ
0.1 µM
0.3 μΜ
1 µM
3 µM
10 μ.Μ
30 μM
100 μΜ
Substance Concentration
Substance Concentration
using two other QSAR models, i.e. MultiCASE and Leadscope (Rosenmai et al., 2014).
The binding affinity to the ERß for BPA is about two times stron- ger than for BPF and nearly 100 times stronger than for BPS. How- ever, compared to 17ß-estradiol, the binding affinity of BPA is still one order of magnitude weaker (Table 5). Due to the binding affini- ties of BPF to the other receptors, this substance has a TP almost
similar to BPA. D-8 shows weak binding affinities for several recep- tors. This indicates that BPA and BPF have a higher potential than BPS and D-8 to show hormonal activity, but at significantly higher concentrations than the endogenous hormone 17ß-estradiol. Per- gafast® 201 shows less concern.
These in silico results have been confirmed in in vitro assays for the bisphenols (Rosenmai et al., 2014). However further in vitro
Table 5 VirtualToxLab™. Binding affinity profile (binding constant K) and estimated toxic potential (TP) of BPA, BPF, BPS, Pergafast® 201 and D-8. The lower the concentration, the stronger the binding affinity to the target protein. Binding affinity >100 uM are considered not binding. Toxic potential is ranked in 3 categories: TP < 0.3 (low), 0.3 < TP < 0.6 (moderate), and TP > 0.6 (high). Strongest binding affinity and toxic potential for each compound are highlighted in bold.
| Bisphenol A | Bisphenol F | Bisphenol S | Pergafast® 201 | D-8 | 17ß-Estradiol | |
|---|---|---|---|---|---|---|
| Androgen receptor | 0.577 μΜ | 1.96 μΜ | 21.1 μΜ | Not binding | 7.93 μΜ | 0.047 μΜ |
| Aryl hydrocarbon receptor | 9.90 μΜ | Not binding | Not binding | Not binding | 77.3 μΜ | 16.1 μΜ |
| CYP450 1A2 | 29.0 μ.Μ | 32.7 μΜ | Not binding | Not binding | Not binding | 4.38 μΜ |
| CYP450 2C9 | 39.9 μΜ | Not binding | Not binding | Not binding | Not binding | 21.5 μΜ |
| CYP450 2D6 | 31.0 μΜ | Not binding | Not binding | 92.7 μΜ | Not binding | 6.16 μΜ |
| CYP450 3A4 | Not binding | Not binding | Not binding | 97.1 μΜ | not binding | 39.0 μΜ |
| Estrogen receptor a | 5.76 μΜ | 3.10 μΜ | 25.7 μΜ | Not binding | 2.01 μΜ | 0.038 μΜ |
| Estrogen receptor B | 0.084 μΜ | 0.161 μΜ | 0.742 μΜ | Not binding | 1.33 μΜ | 0.004 μΜ |
| Glucocorticoid receptor | 0.203 μ.Μ | 1.57 uM | 10.2 µM | 44.3 μΜ | 9.93 μΜ | 0.172 uM |
| hERG | 14.1 uM | 28.9 μΜ | 49.8 μΜ | Not binding | 3.18 μΜ | 3.45 μΜ |
| Liver X receptor | 64.7 μ.Μ | 56.1 μ.Μ | Not binding | 40.7 μΜ | Not binding | 11.6 μΜ |
| Mineralocorticoid receptor | 0.430 μΜ | 3.78 uM | 6.64 uM | 31.0 μΜ | 3.58 µM | 0.034 μΜ |
| PPAR Y | 9.80 μΜ | 1.22 μΜ | 35.1 μΜ | 22.0 μΜ | 12.9 uM | 13.1 μΜ |
| Progesterone | 2.29 μΜ | 0.925 μΜ | 13.5 μ.Μ | 23.0 μΜ | 20.3 μΜ | 0.206 μ.Μ |
| Thyroid receptor a | 37.5 µM | 2.55 µM | 78.2 μΜ | 99.2 μ.Μ | 8.46 uM | 4.77 μΜ |
| Thyroid receptor B | 7.02 uM | 22.4 μΜ | 70.2 μ.Μ | 37.5 μΜ | 3.83 uM | 2.71 μΜ |
| Toxic potential | 0.476 | 0.447 | 0.380 | 0.269 | 0.386 | 0.574 |
D.M. Goldinger et al./Regulatory Toxicology and Pharmacology 71 (2015) 453-462
and/or in vivo analyses are required in order to make a concluding statement on the binding activity of D-8 and Pergafast® 201 to hor- mone receptors.
5. Conclusion
Substitution of BPA by its structural analogs BPF and BPS should be considered with caution, since those bisphenols exhibit almost a similar endocrine activity as BPA in the tests applied in this study. Although our study showed that D-8 and Pergafast® 201 could be good alternatives for the replacement of BPA with regards to their in vitro effects on steroidogenesis, further studies are required to show that there are no adverse effects on the hormonal system. Indeed substances which influence the steroidogenesis through the HPG axis (hypothalamic-pituitary-gonadal axis) are not recog- nized by the H295R steroidogenesis assay (OECD, 2011). Effects on non-sexual hormones, such as thyroid hormones are also not cov- ered. Endocrine disruptors can also act through other pathways than receptor binding (Yoon et al., 2014). Further investigation on the effect of these substances, particularly of Pergafast® 201 on enzymes involved in steroidogenesis could give clues for the understanding of the mechanism of toxicity. Finally, there are actu- ally no data available on metabolic activation of D-8 and Pergafast® 201. Therefore, further tests which cover also these and other aspects of the hormonal system have to be conducted to perform a final assessment on the safety of the substitutes.
Conflicts of interest
We declare that the authors have no conflicts of interest.
Acknowledgments
We would like to acknowledge Connect Chemicals GmbH and BASF for kindly providing us D-8 and Pergafast® 201, respectively.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.yrtph.2015.01. 002.
References
Akahori, Y., Nakai, M., Yamasaki, K., Takatsuki, M., Shimohigashi, Y., Ohtaki, M., 2008. Relationship between the results of in vitro receptor binding assay to human estrogen receptor A and in vivo uterotrophic assay: comparative study with 65 selected chemicals. Toxicol. In Vitro 22 (1), 225-231.
Biedermann, S., Tschudin, P., Grob, K., 2010. Transfer of bisphenol A from thermal printer paper to the skin. Anal. Bioanal. Chem. 398 (1), 571-576.
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., 2000. The estrogen receptor relative binding affinities of 188 natural and xenochemicals: structural diversity of ligands. Toxicol. Sci. 54 (1), 138-153.
Cabaton, N., Dumont, C., Severin, I., Perdu, E., Zalko, D., Cherkaoui-Malki, M., Chagnon, M.C., 2009. Genotoxic and endocrine activities of bis (hydroxyphenyl) methane (bisphenol F) and its derivatives in the HepG2 cell line. Toxicology 255 (1), 15-24.
Chen, M .- Y., Ike, M., Fujita, M., 2002. Acute toxicity, mutagenicity, and estrogenicity of bisphenol-A and other bisphenols. Environ. Toxicol. 17 (1), 80-86.
Coleman, K.P., Toscano, W.A., Wiese, T.E., 2003. QSAR models of the in vitro estrogen activity of bisphenol A analogs. QSAR Comb. Sci. 22 (1), 78-88.
ECHA, 2014. Annex XV Restriction Report, Proposal for Restriction, Substance Name: 4.4’-isopropylidenediphenol (bisphenol A; BPA). Available at <http:// echa.europa.eu/documents/10162/c6a8003c-81f3-4df6-b7e8-15a3a36baf76> (accessed October 2014).
EFSA, 2013. Draft scientific opinion on the risks to public health related to the presence of bisphenol A (BPA) in foodstuff. Part: Exposure assessment. EFSA J., 1-314, EFSA-Q-2012-00423
EFSA, 2014. Draft scientific opinion on the risks to public health related to the presence of bisphenol A (BPA) in foodstuff. Part: Assessment of human health risks. EFSA J., 1-532, EFSA-Q-2012-00423
Environmental Working Group (EWG), 2010. BPA in store receipts. Available at http://www.ewg.org/bpa-in-store-receipts (accessed October 2014).
Geens, T., Aerts, D., Berthot, C., Bourguignon, J.P., Goeyens, L., Lecomte, P., Maghuin- Rogister, G., Pironnet, A.M., Pussemier, L., Scippo, M.L., Van Loco, J., Covaci, A., 2012a. A review of dietary and non-dietary exposure to bisphenol-A. Food Chem. Toxicol. 50 (10), 3725-3740.
Geens, T., Goeyens, L., Kannan, K., Neels, H., Covaci, A., 2012b. Levels of bisphenol-A in thermal paper receipts from Belgium and estimation of human exposure. Sci. Total Environ. 435-436, 30-33.
Hashimoto, Y., Moriguchi, Y., Oshima, H., Kawaguchi, M., Miyazaki, K., Nakamura, M., 2001. Measurement of estrogenic activity of chemicals for the development of new dental polymers. Toxicol. In Vitro 15 (4), 421-425.
Hashimoto, Y., Nakamura, M., 2000. Estrogenic activity of dental materials and bisphenol-A related chemicals in vitro. Dent. Mater. J. 19 (3), 245-262.
Hecker, M., Hollert, H., Cooper, R., Vinggaard, A.M., Akahori, Y., Murphy, M., Nellemann, C., Higley, E., Newsted, J., Laskey, J., 2011. The OECD validation program of the H295R steroidogenesis assay: phase 3. Final inter-laboratory validation study. Environ. Sci. Pollut. Res. 18 (3), 503-515.
KEMI, 2012. Bisfenol A i kassakvitton- rapport från ett regeringsuppdrag. KEMI4/12, 1-58.
Kitamura, S., Suzuki, T., Sanoh, S., Kohta, R., Jinno, N., Sugihara, K., Yoshihara, S., Fujimoto, N., Watanabe, H., Ohta, S., 2005. Comparative study of the endocrine- disrupting activity of bisphenol A and 19 related compounds. Toxicol. Sci. 84 (2), 249-259.
Kolsek, K., Mavri, J., Sollner Dolenc, M., Gobec, S., Turk, S., 2014. Endocrine disruptome - an open source prediction tool for assessing endocrine disruption potential through nuclear receptor binding. J. Chem. Inf. Model. 54 (4), 1254-1267.
Kuruto-Niwa, R., Nozawa, R., Miyakoshi, T., Shiozawa, T., Terao, Y., 2005. Estrogenic activity of alkylphenols, bisphenol S, and their chlorinated derivatives using a GFP expression system. Environ. Toxicol. Pharmacol. 19 (1), 121-130.
Lassen, C., Mikkelsen, S.H., Brandt, U.K., 2011. Migration of Bisphenol A from Cash Register Receipts and Baby Dummies. Environmental Protection Agency, 1-67.
Laws, S.C., Yavanhxay, S., Cooper, R.L., Eldridge, J.C., 2006. Nature of the binding interaction for 50 structurally diverse chemicals with rat estrogen receptors. Toxicol. Sci. 94 (1), 46-56.
Liao, C., Kannan, K., 2011. Widespread occurrence of bisphenol A in paper and paper products: implications for human exposure. Environ. Sci. Technol. 45 (21), 9372-9379.
Liao, C., Liu, F., Kannan, K., 2012. Bisphenol S, a new bisphenol analogue, in paper products and currency bills and its association with bisphenol a residues. Environ. Sci. Technol. 46 (12), 6515-6522.
Lu, S.Y., Chang, W.J., Sojinu, S.O., Ni, H.G., 2013. Bisphenol A in supermarket receipts and its exposure to human in Shenzhen, China. Chemosphere 92 (9), 1190-1194.
Mendum, T., Stoler, E., VanBenschoten, H., Warner, J.C., 2011. Concentration of bisphenol A in thermal paper. Green Chem. Lett. Rev. 4 (1), 81-86.
Miller, D., Wheals, B.B., Beresford, N., Sumpter, J.P., 2001. Estrogenic activity of phenolic additives determined by an in vitro yeast bioassay. Environ. Health Perspect. 109 (2), 133.
NICNAS, 2004. Pergafast 201, full public report, file No .: STD/1058. Available at <http://www.nicnas.gov.au/_data/assets/pdf_file/0017/10394/STD1058FR. pdf> (accessed October 2014).
Nishihara, T., Nishikawa, J.I., Kanayama, T., Dakeyama, F., Saito, K., Imagawa, M., Takatori, S., Kitagawa, Y., Hori, S., Utsumi, H., 2000. Estrogenic activities of 517 chemicals by yeast two-hybrid assay. J. Health Sci. 46 (4), 282-298.
OECD, 2011. OECD Guidelines for the Testing of Chemicals, Section 4: Test No. 456: H295R Steroidogenesis Assay. Available at <http://www.oecd-ilibrary. org/docserver/download/9745601e.pdf?expires=1410345422&id=id&accname= guest&checksum=AF893851A46C6FB1A9F795BD800E41F7> (accessed October 2014).
Perez, P., Pulgar, R., Olea-Serrano, F., Villalobos, M., Rivas, A., Metzler, M., Pedraza, V., Olea, N., 1998. The estrogenicity of bisphenol A-related diphenylalkanes with various substituents at the central carbon and the hydroxy groups. Environ. Health Perspect. 106 (3), 167.
Rosenmai, A.K., Dybdahl, M., Pedersen, M., van Vugt-Lussenburg, B.M.A., Wedebye, E.B., Taxvig, C., Vinggaard, A.M., 2014. Are structural analogues to bisphenol A safe alternatives? Toxicol. Sci. 139 (1), 35-47.
Stroheker, T., Picard, K., Lhuguenot, J.C., Canivenc-Lavier, M.C., Chagnon, M.C., 2004. Steroid activities comparison of natural and food wrap compounds in human breast cancer cell lines. Food Chem. Toxicol. 42 (6), 887-897.
Terasaki, M., Shiraishi, F., Fukazawa, H., Makino, M., 2007. Occurrence and estrogenicity of phenolics in paper GCÉ recycling process water: pollutants originating from thermal paper in waste paper. Environ. Toxicol. Chem. 26 (11), 2356-2366.
US EPA, 2014. Bisphenol A alternatives in thermal paper, Final Version. Available at http://www.epa.gov/dfe/pubs/projects/bpa/bpa-report-complete.pdf (accessed October 2014).
Vandenberg, L.N., Chahoud, I., Heindel, J.J., Padmanabhan, V., Paumgartten, F.J., Schoenfelder, G., 2010. Urinary, circulating, and tissue biomonitoring studies indicate widespread exposure to bisphenol A. Environ. Health Perspect., 1055-1070
Vandenberg, L.N., Hunt, P.A., Myers, J.P., vom Saal, F.S., 2013. Human exposures to bisphenol A: mismatches between data and assumptions. Rev. Environ. Health 28 (1), 37-58.
Vedani, A., Dobler, M., Hu, Z., Smiesko, M., 2014. OpenVirtualToxLab - A platform for generating and exchanging in silico toxicity data. Toxicol. Lett.
Yamasaki, K., Noda, S., Imatanaka, N., Yakabe, Y., 2004. Comparative study of the uterotrophic potency of 14 chemicals in a uterotrophic assay and their receptor- binding affinity. Toxicol. Lett. 146 (2), 111-120.
Yoon, K., Kwack, S.J., Kim, H.S., Lee, B.M., 2014. Estrogenic endocrine-disrupting chemicals: molecular mechanisms of actions on putative human diseases. J. Toxicol. Environ. Health Part B 17 (3), 127-174.
Zhang, X., Chang, H., Wiseman, S., He, Y., Higley, E., Jones, P., Wong, C.K., Al- Khedhairy, A., Giesy, J., Hecker, M., 2011. Bisphenol A disrupts steroidogenesis in human H295R cells. Toxicol. Sci. 121, 320-327.