Toxicology and Applied Pharmacology 209 (2005) 95 – 104
www.elsevier.com/locate/ytaap
Effects of bisphenol A-related diphenylalkanes on vitellogenin production
in male carp (Cyprinus carpio) hepatocytes and aromatase (CYP19)
activity in human H295R adrenocortical carcinoma cells
Robert J. Letchera,b,*, J. Thomas Sandersonb, Abraham Bokkersb,
John P. Giesyc, Martin van den Bergb
a
National Wildlife Research Centre, Canadian Wildlife Service, Environment Canada, Ottawa, Ontario, Canada K1A 0H3
b
Institute for Risk Assessment Sciences (IRAS), Utrecht University, PO Box 80.176, 3508 TD Utrecht, The Netherlands
c
Department of Zoology, National Food Safety and Toxicology Center, Institute of Environmental Toxicology, Michigan State University,
East Lansing, MI 48824, USA
Received 24 January 2005; accepted 25 March 2005
Available online 19 May 2005
Abstract
The present study investigated the effects of the known xenoestrogen bisphenol A (BPA) relative to eight BPA-related diphenylalkanes on
estrogen receptor (ER)-mediated vitellogenin (vtg) production in hepatocytes from male carp (Cyprinus carpio), and on aromatase (CYP19)
activity in the human adrenocortical H295R carcinoma cell line. Of the eight diphenylalkanes, only 4,4V-(hexafluoropropylidene)diphenol
(BHF) and 2,2V-bis(4-hydroxy-3-methylphenyl)propane (BPRO) induced vtg, i.e., to a maximum of 3% to 4% (at 100 AM) compared with
8% for BPA relative to the maximum induction by 17h-estradiol (E2, 1 AM). Bisphenol A diglycidyl ether (BADGE) was a potent antagonist
of vtg production with an IC50 of 5.5 AM, virtually 100% inhibition of vtg at 20 AM, and an inhibitive (IC50) potency about one-tenth that of
the known ER antagonist tamoxifen (IC50, 0.6 AM). 2,2V-Diallyl bisphenol A, 4,4V-(1,4-phenylene-diisopropylidene)bisphenol, BPRO, and
BHF were much less inhibitory with IC50 concentrations of 20 – 70 AM, and relative potencies of 0.03 and 0.009 with tamoxifen. Bisphenol
ethoxylate showed no anti-estrogenicity (up to 100 AM), and 4,4V-isopropylidene-diphenol diacetate was only antagonistic at 100 AM. When
comparing the (anti)estrogenic potencies of these bisphenol A analogues/diphenylalkanes, anti-estrogenicity occurred at lower concentrations
than estrogenicity. 4,4V-Isopropylidenebis(2,6-dimethylphenol) (IC50, 2.0 AM) reduced E2-induced (EC50, 100 nM) vtg production due to
concentration-dependent cytotoxicity as indicated by a parallel decrease in MTT activity and vtg, whereas the remaining diphenylalkanes did
not cause any cytotoxicity relative to controls. None of the diphenylalkanes (up to 100 AM) induced EROD activity indicating that
concentration-dependent, CYP1A enzyme-mediated metabolism of E2, or any Ah-receptor-mediated interaction with the ER, was not a likely
explanation for the observed anti-estrogenic effects. At concentrations as great as 100 AM, none of the diphenylalkanes directly inhibited
aromatase (CYP19) activity in H295R cells. Environmental exposure of fish to BPA and related diphenylalkanes, depending on the structure,
may pose anti-estrogenic, and to a lesser extent estrogenic, risks to development and reproduction.
D 2005 Elsevier Inc. All rights reserved.
Keywords: Diphenylalkanes; Bisphenol A; (Anti)estrogenicity; Carp hepatocytes; Vitellogenin; Human H295R cells; Aromatase; CYP19
Introduction
* Corresponding author. National Wildlife Research Centre, Canadian
Wildlife Service, Environment Canada, Raven Road (Carleton University),
Ottawa, Ontario, Canada K1A 0H3. Fax: +1 613 998 0458.
E-mail address: robert.letcher@ec.gc.ca (R.J. Letcher).
0041-008X/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.taap.2005.03.013
Exposure to natural, synthetic, and xenobiotic estrogens
(xenoestrogens) has been implicated in the deleterious
health effects observed on male reproduction and reproductive abnormalities, and hormonal changes in humans and
wildlife species (Brucker-Davis et al., 2001; Colborn et al.,
1993). Non-steroidal, synthetic compounds with estrogenic
96
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
or anti-estrogenic activity include medicines (e.g., tamoxifen), pesticides (e.g., methoxychlor, lindane, toxaphenes),
and plasticizers such as bisphenol A (BPA), para-nonylphenol (Hoivik et al., 1998; Smeets et al., 1999a, 1999b).
Males of various fish species exposed to xenoestrogens such
as BPA, para-nonylphenol under laboratory conditions, or
via wastewater effluent have shown increased levels of the
estrogen-inducible yolk precursor protein vitellogenin (vtg),
inhibited testicular growth and abnormalities, and formation
of intersex gonads (Nicolas, 1999; Yamanaka et al., 1998).
BPA is one of many diphenylalkanes that are raw
materials for the production of polymers such as polycarbonates, epoxy, and phenolic resins, polyesters, and polyacrylates, and are used commercially in plastics and
coatings in the dental and food industry. For example,
concentrations of BPA diglycidyl ether (BADGE) were
measured in tuna and oil extracts from cans from four
European countries, with inner-coatings containing BADGE
(Berger and Oehme, 2000; Berger et al., 2001). BADGErelated compounds were identified that originated from
reactions of the glycidyl ethers with bisphenols, phenol,
butanol, water, and hydrochloric acid. In tuna up to 3.7 mg/g
were found for the BADGE hydrochlorination product,
bisphenol F diglycidyl ether (BFDGE + 2 HCl) in tuna. The
highest concentrations were 20 mg/g in tuna and 43 mg/g in
the oil phase. BADGE and BPA have also been measured in
composites and sealants used in the dentistry, as well as the
saliva of recently treated patients (Olea et al., 1996). BPA,
as an additive or hydrolyzed BADGE product, was
measured in saliva at 3– 20 Ag/mL.
BPA mimics the activity of estrogens such as 17hestradiol (E2) via interaction with either a- and h-estrogen
receptors (ERs). BPA was found to have a similar relative
binding affinity with human ERaand ERh proteins, which
was ¨1000-fold less than for E2 (Kuiper et al., 1998).
Furthermore, BPA similarly stimulated ERa and/or ERhmediated transcriptional activity in transfected 293 human
embryonal kidney and HepG2 human hepatoma cells, at
treatment concentrations of 100 –1000 nM (Could et al.,
1998; Kuiper et al., 1998). BPA is considered a model
xenoestrogen despite the fact that the in vitro estrogenic
potency is generally 15,000 times less than that of E2. In
addition to BPA, several related diphenylalkanes have
demonstrated estrogenic activity in vitro including increased
proliferation of MCF-7 human breast cancer cells (Perez et
al., 1998). For BPA-related diphenylalkanes, and those with
ester and ether type bonds, MCF7-treated cells could cleave
these bonds and release para-hydroxy groups (and are thus
more BPA-like), which activated their proliferative estrogenic effects. For example, identified in extracts of resins
composites used in dentistry, BADGE treatments of MCF7
cells at high concentrations of 10 AM have been shown to
estrogenic, although with no competitive binding affinity for
the estrogen receptor (Olea et al., 1996). Evidence shows
that the thermal, chemical, and enzymatic degradation of
some BPA-based diphenylalkane polymers represent a
source of biologically active monomers (e.g., BADGE)
(Climie et al., 1981; Pottenger et al., 2000). Because BPArelated diphenylalkane derivatives are in widespread commercial use and their production is increasing, potential
human exposure to estrogenic diphenylalkanes has become
a significant environmental issue.
A growing number of xenobiotics that possess antiestrogenic activities have been identified, i.e., compounds that
antagonize estrogen-dependent processes in target tissues.
Pathways of anti-estrogenicity include competitive binding
E2 with ERa and/or ERh, down-regulation of ER-mediated
responses via aryl hydrocarbon receptor (AhR) cross-talk,
and accelerated metabolism of estrogens due to AhRmediated enzyme induction (e.g., cytochrome P4501A
(CYP1A)) (Safe and Wormke, 2003). Major examples of
synthetic antiestrogens that are active via ER-independent
and AhR-dependent mechanisms are polychlorinated
dibenzo-p-dioxins (PCDDs) and dibenzofurans (PCDFs),
and dioxin-like non- and mono-ortho PCBs.
In vitro assays developed for the routine screening of
potential xenoestrogens also include competitive ER binding
assays and induction of reporter gene expression in transiently or stably infected cell lines (Ankley et al., 1998; Reel
et al., 1996). Vitellogenin (vtg) is a precursor of the yolk
proteins phosvitin and lipovitelline, and is synthesized in the
liver of female, oviparous vertebrates. Vtg synthesis is
induced by estrogen-dependent stimulation of gene expression. Although the vtg gene is present, the endogenous
plasma concentrations of estrogen are normally too small to
induce vitellogenesis in hepatocytes of male and female
oviparous vertebrates. Vtg has been used as a biomarker of
exposure to (anti)estrogenic compounds in a number of in
vivo and in vitro studies with fish, and in particular carp
(Letcher et al., 2002; Rankouhi et al., 2004; Rodgers-Gray et
al., 2000; Rouhani-Rankouhi et al., 2002; Smeets et al.,
1999a, 1999b; Tolar et al., 2001). Smeets et al. (1999b)
concluded that in a carp hepatocyte-vtg production (CARPHEP) assay that hepatocytes of male fish origin, rather than
female origin, be used when assessing estrogenic potencies of
compounds. They found that female hepatocytes produced
relatively large basal levels of vtg, whereas uninduced male
hepatocytes secreted no detectable vtg. Furthermore, the
absolute quantities of vtg produced by E2-treated cells of
female origin were some 25-fold higher than in similarly
treated cells from males. Nevertheless, dose – response
relationships for vtg induction by E2 were virtually identical
for male and female hepatocytes. They concluded that from a
toxicological point of view, vtg production in male carp is
more relevant since there is little basal vtg secretion, and as a
result induction of vtg can more easily be detected, which has
an obvious advantage for an in vitro screening assay.
The objectives of the present study is to determine the
estrogenic or anti-estrogenic activity of BPA and eight BPArelated diphenylalkanes in a CARP-HEP assay by measuring
the effects on the production of vtg and CYP1A-mediated
catalytic activity in exposed hepatocytes from male carp
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
(Cyprinus carpio) (Letcher et al., 2002; Rankouhi et al.,
2004; Rouhani-Rankouhi et al., 2002; Smeets et al., 1999a,
1999b). Xenobiotic compounds may also disturb endocrine
functions through mechanisms other than via interactions
with hormone receptors, including interference with aromatase (CYP19) enzyme, which mediates the biosynthesis of
estrogens (Sanderson et al., 2001, 2002). Therefore, the
present study also investigated the ability of BPA-related
diphenylalkanes to directly inhibit aromatase (CYP19)
activity (androgen to estrogen conversion) in the human
adrenocortical H295R carcinoma cell line (Sanderson et al.,
2000).
Materials and methods
Chemicals. Bisphenol A (BPA), bisphenol ethoxylate
(BEO, >99%), 2,2V-diallyl bisphenol A (DAB, 98%),
4,4V-isopropylidenebis(2,6-dimethylphenol) (BTM, 98%),
97
2,2V-bis(4-hydroxy-3-methylphenyl)propane (BPRO,
>99%), 4,4V-(1,4-phenylene-diisopropylidene)bisphenol
(BDP, >99%), 4,4V-isopropylidene-diphenol diacetate
(BDA, 98%), para-cresol (98%), and ortho-cresol (98%)
were obtained commercially in the greatest purity available
from Sigma-Aldrich (St. Louis, MO, USA). BPA diglycidyl ether (BADGE, >99%) and 4,4V-hexafluoropropylidene)diphenol (BHF, 98%) were purchased from Fluka
(Buchs, Switzerland). The chemical structures of the
diphenylalkanes are shown in Fig. 1. 17h-Estradiol (E2,
>99%), tamoxifen (>99%), and dimethyl sulfoxide
(DMSO, 99.9%, Janssen Chimica, Geel, Belgium) were
purchased from Sigma (St. Louis, MO, USA). ICI 182,780
was a kind gift from Dr. A. Wakeling (Zeneca Pharmaceuticals, UK). All compounds were prepared as 1000-fold
concentrated DMSO stock solutions.
CARP-HEP/vtg bioassay. The common carp (C. carpio)
used in the carp hepatocyte/vitellogenin (CARP-HEP/VTG)
Fig. 1. Chemical structures, names, and abbreviations of BPA and related diphenylalkanes. The hydrogens on the aromatic ring have been omitted for clarity.
BPA = bisphenol A; BADGE = bisphenol A diglycidyl ether; BEO = bisphenol ethoxylate; DAB = 2,2V-diallyl bisphenol A; BTM = 4,4V-isopropylidenebis
(2,6-dimethylphenol); BPRO = 2,2V-bis(4-hydroxy-3-methylphenyl)propane; BHF = 4,4V-hexafluoropropylidene)diphenol; BDP = 4,4V-(1,4-phenylenediisopropylidene)bisphenol; BDA = 4,4V-isopropylidene-diphenol diacetate.
98
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
bioassay were genetically uniform, all male (XY), F1 hybrid
progenies. Further details of the carp, the maintenance of the
fish prior to use in the assay, and the perfusion procedure are
described in detail in Smeets et al. (1999a, 1999b) and
Rouhani-Rankouhi et al. (2002). Briefly, carp hepatocytes
were freshly perfused by a two-step retrograde technique,
isolated and cultured. The liver was first perfused with Ca2+and Mg2+-free buffer containing EDTA (0.145 M NaCl; 5.4
mM KCl; 5 mM EDTA; 1.1 mM KH2PO4; 12 mM NaHCO3;
3 mM NaH2PO4; 100 mM HEPES; pH 7.5) and followed by
a step with the same buffer containing no EDTA and 0.26
mg/ml collagenase D (Boehringer, Mannheim, Germany).
The perfused liver sections were filtered through nylon
mesh. After removal, mincing, sieving, and washing, the
hepatocytes were re-suspended in culture medium. The cell
viability was >90% as assessed with trypan blue staining.
The isolated hepatocytes were cultured in phenol red-free
DMEM/F12 media (D2906, Sigma, St. Louis, MO) supplemented with 14.3 mM NaHCO3, HEPES (final concentration: 20 mM), 50 mg/l gentamycin, 1 AM insulin, 10 AM
hydrocortisone, 2% v/v Ultroser SF (steroid-free) serum
(Jones Chromatography, Mid Glamorgan, UK), and 2 mg/l
of the protease-inhibitor aprotinin (Fluka, Buchs, Switzerland) at pH 7.4. The concentration of the cell suspension was
1.0 106 cells/ml, and 0.18 ml (or 180,000 cells) was added
to each well of 96-well tissue plates (Griener, Alphen a/d
Rijn, the Netherlands). The plates were maintained at 24 -C
for a period of 36 h to acclimatize the cells. The proportion of
erythrocytes did not exceed 10% of the hepatocytes.
After the 36 h acclimatization period the confluence of the
cells was about 70% and were ready for dosing. All DMSO
stock solutions of compounds and concentrations were
diluted 1000-fold in assay medium for bioscreening in the
CARP-HEP/vtg assay. The culture medium (90%) was
replaced by assay media (164 AM) containing 10% greater
compound concentrations to obtain the desired 1000-fold
dilution. After 2 days, the assay medium (90%) was refreshed
with new assay medium containing nominal compound and
E2 concentrations. After a further 2 days, the medium was
transferred to new 96-well plates and frozen at 70 -C until
further use. The remaining hepatocytes were used to
determine EROD activity, cell viability, or protein content.
An indirect competitive ELISA was used to quantify the vtg
present in the assay medium. The ELISA procedure, as well
as calculations to quantify vtg, has been thoroughly described
(Letcher et al., 2002; Rouhani-Rankouhi et al., 2002; Smeets
et al., 1999a, 1999b). For each experiment set, an E2 dose –
response curve plate was included with concentrations
ranging from 0.6 to 6000 nM. An E2-positive control
(EC50, 100 nM) was included on all other plates. All E2
and compound concentrations were tested in two separate
sets of 6-fold replicates. The final DMSO concentration in the
dosing medium did not exceed 0.2% (v/v).
cells (ATCC No. CRL-2128) for use in the H295R/CYP19
bioassays have been described in Sanderson et al. (2000,
2001). Briefly, the H295R cells were cultured in the
presence of 2% of the steroid-free serum replacement
Ultroser SF (Soprachem, France). Cells were grown until
almost confluent and then trypsinized and suspended in 75
mL if culture media. Three 24-well cell culture plates were
seeded (1 mL in each well) with the cell suspension. After
24 h the medium was changed, and the cells were well
attached at about 2 105 cells/well).
The direct effect of the test compounds on the aromatase
(CYP19) activity in the H295R cells was determined
according to previously described in Materials and methods
(Letcher et al., 1999; Sanderson et al., 2000, 2001). Briefly,
the cells were washed twice with PBS and then exposed to
[1h-3H(N)]-androst-4-ene-3,17-dione (28.5 Ci/mmol; specific activity 1054.5 GBq/mmol; Lot No. 3329278; New
England Nuclear; NET-926). CYP19 catalyzes the conversion of [1h-3H(N)]-androst-4-ene-3,17-dione to estrone and
3
H2O. The H295R cells were co-incubated with [1h-3H]androstenedione and BPA-related diphenylalkanes. The
same BPA and BPA-related diphenylalkanes and concentrations were screened in the H295R/CYP19 bioassay. The
CYP19 activity in the H295R assays was determined from
the radioactivity of extracted 3H2O. Correction was made
for the effect of solvent, background radioactivity, dilution
factor, the 3H distribution on the [1h-3H]-androstenedione,
and the specific activity of [1h-3H]-androstenedione.
H295R/CYP19 bioassay. The details of culturing and
maintenance of the human H295R adrenocortical carcinoma
Cell viability. The viability of the carp hepatocytes,
exposed for 4 days to the various compounds, was assessed
Estrogenic and antiestrogenic activity in the CARP-HEP/vtg
assay. To determine the effects on vtg production in the
CARP-HEP assay, 2 sets of n = 6 replicates were performed
for the estrogenicity and anti-estrogenicity screens for all
compounds and concentrations. Antiestrogenic activity of
the test compounds was assessed in the in vitro assays using
the same nominal concentrations and procedures used in the
estrogenicity screening. In the CARP-HEP/vtg assay, the
hepatocytes were co-administered with 100 nM (EC50) of
E2 and the test compounds. Concentrations of 0.1, 1.0, and
10 AM of the known competitive ER antagonist tamoxifen
were included as positive controls in the anti-estrogenicity
experiments.
An anti-estrogenic effect was defined by the capacity an
antagonist to inhibit the vtg production induced by E2.
The percentage of the remaining E2-induced response
(%RR) is calculated according to the following: %RR =
((A compound+E2
A control) / (A E2
A control)) 100%,
where A compound+E2, A control, and A E2 are the average
activity of the test wells, control wells, and wells
incubated with E2 alone, respectively. A compound that
is not antagonistic will elicit a %RR of 100%, i.e., the
response of the co-administered concentration of E2 will
be the same as the E2 concentration administered alone.
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
by determining changes in the mitochondrial succinate
dehydrogenase-mediated metabolism of the substrate 3-(4,5dimethyl-thiazol-2yl)-2,5-diphenyltetrazolium bromide
(MTT), or the MTT activity according to Denizot and Lang
(1986). The MTT activity in the remaining monolayer of
hepatocytes after the vtg-containing solution was harvested
was determined according to Smeets et al. (1999a) and
Letcher et al. (2002) with minor modifications. The viability
of the exposed H295R cells was determined in the same
manner, but including minor modifications according to
Sanderson et al. (2000) and Letcher et al. (1999).
EROD activity and protein content in the carp
hepatocytes. The diphenylalkane compounds were assessed
for the ability to interact with the AhR and induce CYP1A
enzyme activity. After vtg harvesting, the carp hepatocytes
were frozen or used immediately for the determination of
CYP1A enzyme activity by measurement of ethoxyresorufinO-deethylase (EROD) activity (Burke and Mayer, 1974).
2,3,7,8-Tetrachloro-dibenzo-p-dioxin (TCDD) concentrations
of 0.3, 1.0, 10, 100, and 1000 pM were included as positive
controls.
Protein content of the exposed carp hepatocytes and
H295R cells was determined according to the procedure of
Lowry et al. (1951) and Rutten et al. (1987), with some
modifications. For the CARP-HEP assays, after vtg harvesting, the hepatocytes were washed twice with 200 Al of
phosphate-buffered saline (PBS). An aqueous solution of
0.5% Triton X-100 (v/v, 200 Al) was then added. The plates
were frozen at 70 -C for at least 2 h. For both the carp
hepatocytes and H295R cells, after thawing to room
temperature, the protein content in each well was diluted,
if necessary, for protein determination using a BSA standard
curve. EROD activity was not measured in the H295R cells.
Concentration – response curves, calculations, and
statistics. Concentration – response relationships are
described by the sigmoidal function y = a 0 + a 1 / (1 +
exp((a 2 + x) / a 3)) (SlideWrite Plus 4.0, Advanced Graphics
Software, Carlsbad, CA), in which y is the remaining
amount of vtg production (%RR), and x is the logarithm of
the treatment concentration. For estrogenic effects, vtg
levels were normalized to the maximum levels induced by
E2, and expressed as a percent of this maximum level. For
anti-estrogenic effects, the levels were normalized to the vtg
level induced by the co-administered E2 concentration
alone. The statistical significance of the differences between
compound-treated and control hepatocytes was determined
using a two-way ANOVA (P < 0.05).
Results
The LOEC (4.1 nM) and EC50 (¨100 nM) concentrations of E2 for vtg production in the CARP-HEP assay
were similar to the findings by Smeets et al. (1999a, 1999b).
99
E2-induced vtg production (EC50) was inhibited essentially
100% at tamoxifen concentrations >10 AM (Table 1).
Tamoxifen alone did not induce vtg production up to 10 AM.
Initial screening of the diphenylalkanes and the cresol
compounds (at 10 AM and 100 AM) in the CARP-HEP/vtg
assay are shown in Fig. 2. At the 100 AM treatment level,
BPA, BPRO, and BHF were the only diphenylalkanes that
demonstrated estrogenic activity, i.e., about 3%, 5%, and
8% maximum induction, respectively, in comparison to the
maximum vtg production induced by E2 (1000 nM). For
BHF, at the 10 AM treatment level there was also a
significant induction of vtg.
Initial screening of the diphenylalkanes (at 10 AM and
100 AM) co-incubated with E2 (EC50, 100 nM) in the
CARP-HEP/vtg assay indicated that BADGE, BTM, and
para-cresol were the most potent anti-estrogens (<10 AM),
followed by DAB, BPRO, BHF, BDP (>10 AM), BDA, and
ortho-cresol (<100 AM), and finally by BEO, which had no
significant (P < 0.05) antagonistic effect at 100 AM (Fig. 3).
The IC50 values for the inhibition of E2-induced vtg
production by BADGE, BTM, DAB, BPRO, BHF, BDP,
para-cresol, and ortho-cresol are summarized in Table 1.
The IC50 for BADGE was the lowest, making BADGE
about one-tenth as potent as tamoxifen (Table 1, Fig. 4).
With the exception of BTM, at the 10 AM or 100 AM
treatment level none of the diphenylalkanes or cresol
compounds significantly (P < 0.05) altered the MTT
activity relative to DMSO-dosed hepatocytes (negative
controls). BTM concentration-dependently decreased MTT
activity (Fig. 5A), which was parallel to the concentration-dependent antagonism of E2-induced vtg production (Fig. 5B).
Carp hepatocytes possess inducible CYP1A activity. All
concentrations of the diphenylalkane and cresol compounds,
with the exception of BEO, were evaluated for potential
CYP1A enzyme induction via interaction with the Ah
Table 1
Inhibitory potencies of BPA-related diphenylalkanes (Fig. 1) and tamoxifen
on E2-induced (100 nM) vitellogenin (vtg) production in the CARP-HEP
assay
Diphenylalkane
IC50
(AM)
Relative inhibition
potency (IC50
tamoxifen/IC50)
Maximum Vtg
inhibition (at
100 AM) as a %
of E2-mediated
induction
BADGE
BEO
DAB
BTM
BPRO
BHF
BDP
BDA
para-cresol
ortho-cresol
Tamoxifen
5.5
N/A
22.4
2.0
64.8
68.4
63.7
N/A
8.1
72.7
0.6
0.1
N/A
0.03
0.3
0.009
0.009
0.009
N/A
0.07
0.008
1.0
100
N/A
99
100
95
99
83
47
99
64
94*
*Maximum inhibition reached at 10 AM.
100
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
Fig. 2. Percent vtg induction of diphenylalkanes relative to maximum E2 (1000 nM) induction in the CARP-HEP assay. Vtg induction by 0.0001, 0.001, and
0.01 nM E2 (lower part of the E2 induction curve) is shown for comparison. The asterisk indicates a significant (P < 0.05) increase from zero induction. Error
bars are the TSD (n = 6 replicates).
receptor. None of the compounds or concentrations tested
(up to 100 AM) resulted in the induction of EROD activity
above the detection limit of 1.7 pmol/min/mg protein) (not
shown).
The diphenylalkanes and cresols were screened in the
H295R cells to determine any inhibitory or inductive
effects on aromatase activity. The aromatase activity in
the DMSO controls was not significantly different (P <
0.05) from the activity of non-exposed cells. Therefore,
activities in DMSO and non-exposed cells were set to
100% activity. Inhibition of aromatase activity by the
positive control 4-HA was in all cases greater than 95%.
With the exception of DAB, none of the diphenylalkanes
and cresols significantly inhibited aromatase activity in
H295R cells at concentrations up to 20 AM. The DABmediated antagonism of aromatase activity (IC50 of 70
AM, top a maximum inhibition of about 75%) was
parallel to a concentration-dependent decrease in MTT
activity, suggesting that the decline in aromatase activity
was secondary to cell death.
Discussion
Based on the observed (anti)estrogenic effects for the
BPA analogues tested in our study some preliminary
conclusions can be drawn about the structural requirements
that are necessary to elicit (anti)estrogenic effects in the
Fig. 3. Percent inhibition of E2-induced (EC50, 100 nM) vtg production by the diphenylalkanes in the CARP-HEP assay. The asterisk indicates a significant
(P < 0.05) decrease from the E2 control (only 100 nM). Error bars are the TSD (n = 6 replicates). The dotted line is the IC50 concentration.
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
Fig. 4. A representative concentration – response curves for the percent
inhibition of E2-induced (EC50, 100 nM) vtg production by BADGE
(Fig. 1) versus the known ER-antagonist tamoxifen. Error bars are the TSD
(n = 6 replicates).
CARP/HEP assay. The phenolic moiety appears to play an
essential role in stimulating the estrogen-dependent vtg
synthesis as was shown by the significant potency of BPA
within this group of structural analogues. This prerequisite
is confirmed by the observed estrogenicity of BHF, which
like BPA possesses two unsubstituted phenolic groups.
Substitution adjacent to the hydroxy groups decreased the
potency for vtg synthesis but it depended on the nature and
number of the substituents. In our experiments, induction
properties for vtg synthesis decreased in the order one
adjacent methyl group > two adjacent methyl groups > an
allyl group. In addition, if the phenol moiety in the
diphenylalkane molecule is fully replaced by, e.g., a
diglycidyl ether or acetate group, the inductive properties
for vtg production are lost. Since vtg synthesis is an ERmediated process, these different potencies likely represent
various molecular interactions that involve activation via the
101
ER. Differences in estrogenic potencies related to the
structure of diphenylalkanes have been observed earlier in
the MCF-7 cell line (Perez et al., 1998). Besides the
phenolic structure of these compounds, the length and
nature of the substituents on the bridging carbon play an
essential role in determining interactions with the ERs
(Perez et al., 1998).
Another aspect that may have played a role in the
differential potencies of these diphenylalkanes is bioactivation. The estrogenic potency of BPA has been observed to
be significantly increased when incubated with a rat S9
fraction and this bioactivation depended on microsomal and
cytosolic constituents (Yoshihara et al., 2001). The role of
specific human hepatic CYP enzymes in the metabolism of
BPA has been studied and it was concluded that especially
CYP2C enzymes played a significant role (Niwa et al.,
2001). However, it has been shown for carp hepatocytes that
the major CYP activity in due to CYP1A isoforms relative
to, e.g., CBY2B-like mediated activities (Smeets et al.,
1999a, 1999b). The involvement of CYP-mediated activity
in the possible bioactivation of diphenylalkanes to more
estrogenic compounds requires further investigation. However, earlier studies with these carp hepatocytes with
methoxychlor and a potent CYP1A inducer, TCDD, could
not find evidence for bioactivation of this environmental
estrogen (Smeets et al., 1999a). This may be due to factors
such as relatively short exposure times to the hepatocytes.
With the exception of BTM, the diphenylalkanes
inhibited E2-induced vtg production in the CARP/HEP
assay. Among these congeners, BADGE was the most
potent inhibitor of vtg synthesis with a potency that is
about ten-fold less than that of tamoxifen. When observing
the anti-estrogenic effects of these diphenylalkane analogues, a specific role of the hydroxy group of phenolic
moiety or specific substituents appears less pronounced in
our experiments.
Fig. 5. Concentration – response relationships for (A) MTT activity (cytotoxicity) and (B) inhibition of E2-induced vitellogenin (vtg) production in carp
hepatocytes exposed to the diphenylalkane 4,4V-isopropylidenebis(2,6-dimethylphenol) (BTM, Fig. 1). Error bars are the TSD (n = 6 replicates). The
significance of the BTM effect relative to the (A) DMSO control and (B) E2 (100 nM) controls is indicated (*P < 0.05).
102
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
Basically three different pathways could explain the
anti-estrogenicity of the diphenylalkanes and these include
the following: (1) accelerated metabolism of estrogens due
to enzyme induction, (2) down-regulation of ER-mediated
responses via the AhR cross-talk, and (3) competitive
binding to the ER. Earlier studies with carp or rainbow
trout hepatocytes have also addressed the possibility of
anti-estrogenicity through enhanced metabolism due to
enzyme induction (Navas and Segner, 2000; Smeets et
al., 1999b). In the case of (3), in spite of the fact that both
studies examined potent CYP inducers such polychlorinated dibenzo-p-dioxins (PCDDs), PCBs, or PAHs, no
evidence was found that E2 metabolism explained the
observed anti-estrogenicity.
Another possibility for the anti-estrogenicity comes from
possible down-regulation of estrogen-mediated processes
through cross talk with the AhR (Anderson et al., 1996).
Several studies have shown that Ah receptor agonists such
as PCDDs and non-ortho-chlorinated PCB congeners can
down-regulate estrogen-mediated processes (Safe, 1995;
Safe and Wormke, 2003). The down-regulation of vtg
synthesis due to dioxin-like compounds has been studied
earlier in the CARP/HEP assay (Smeets et al., 1999b). For
these compounds a similar rank order potency for CYP1A
induction and anti-estrogenicity was observed, strongly
indicating the involvement of an AhR-mediated process.
Based on good correlations between PAH-mediated CYP1A
induction and anti-estrogenicity in trout hepatocytes a
similar mechanism via cross talk with the AhR was
suggested to occur (Navas and Segner, 2000). For PAHs
and dioxin-like compounds a similar crosstalk between the
AhR and ER was observed in a human endometrial cell line
and this resulted in a reduction of ER-mediated responses
(Wormke et al. 2000). However, in the MCF-7 breast tumor
cell line it was also observed that metabolites of PAHs could
act as anti-estrogens through interactions with the ER, but
also possibly by increased metabolism and depletion of E2
(Arcaro et al., 1999). In our experiments with carp
hepatocytes and diphenylalkane analogues it seems unlikely
that these compounds are anti-estrogenicity via cross talk
with the AhR, as no CYP1A induction could be observed.
Thus, a direct interaction between the diphenylalkanes and
the ER appears far more likely. In this case these
compounds do have to compete with E2 for binding to
the ER.
None of the diphenylalkanes and cresols significantly
inhibited aromatase activity in H295R cells at concentrations up to 20 AM. This may suggest that hepatic
steroidogenic activity in humans is not significantly affected
by BPA-related diphenylalkanes. However, this may not be
necessarily true in carp liver for other target tissues in carp
where aromatase is significant steroidogenic process.
Ovarian aromatase activity was recently reported in common carp from the Ebro River in Spain (Lavado et al.,
2004). In this study, it was concluded that sewage effluent in
the area was a major causal factor leading to the detected
estrogenic effects in carp, including low ovarian aromatase
activity, and reduced testosterone and estradiol in males in
Zaragoza and Canal Imperial de Aragon (an agricultural
area), which suggest decreased estrogen synthesis, and
possibly reduced sex hormone excretion in carp.
In conclusion, our experiments showed that some BPArelated diphenylalkanes have strong anti-estrogenic properties and are rather weakly estrogenic activity in vitro in a
CARP-HEP assay. For one compound, BADGE, the antiestrogenic properties are within one order of a magnitude
of that of tamoxifen. These findings would suggest that
carp, and perhaps other fish species and populations,
exposed to BPA-related diphenylalkanes, depending on
the structure, may pose a risk for development and
reproduction. This may be especially true with respect to
antiestrogenic effects via ER-dependent processes. However, at the population level for carp, the overall estrogenic
versus anti-estrogenic compound mixture impact is more
complex. Sole et al. (2002) reported the feral female and
male carp living in warm waters of Southern Europe, two
tributaries (the Anoia and the Cardener) of the Llobregat
River in northeast Spain, and known to be polluted by
estrogenic compounds, had elevated levels of vtg in the
plasma and liver and elevated EROD activity in the liver.
This would suggest that a complexity of estrogen-disrupting compounds in human effluent discharges in this
Spanish river system, which perhaps includes BPA-related
diphenylalkanes, and that at a population level for carp the
overall exposure affect is estrogenic rather than antiestrogenic. However, contrary results for feral carp were
reported in male feral carp from the lower course of another
Spanish River, the Ebro, and downstream from sewage
treatment plants (Lavado et al., 2004). In these male carp
low plasmatic vtg levels in male carp were found. Perez et
al. (1998) generally concluded that the extent of environmental exposures to BPA-related diphenylalkanes is presently not well defined, i.e., quantitative daily exposure and
absorption, and toxicokinetics and health impacts. Regardless, these compounds should be regarded as (anti)estrogenic xenobiotics in humans, which may also be likely for
aquatic wildlife in close proximity to discharges and
effluents from substantially urbanized areas.
Acknowledgments
This study was finally supported by an internal research
grant from Utrecht University. We thank Ineke van Holsteijn
(IRAS) for her invaluable laboratory assistance.
References
Anderson, M.J., Miller, M.R., Hinton, D.E., 1996. In vitro modulation of
17h-estradiol-induced vitellogenin synthesis: effects of cytochrome
P4501A1 inducing compounds on rainbow trout (Oncorhynchus
mykiss) liver cells. Aquat. Toxicol. 34, 327 – 350.
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
Ankley, G., Mihaich, E., Stahl, R., et al., 1998. Overview of a workshop on
screening methods for detecting potential (anti-) estrogenic/androgenic
chemicals in wildlife. Environ. Toxicol. Chem. 17, 68 – 87.
Arcaro, K.F., O’Keefe, P.W., Yang, Y., Clayton, W., Gierthy, J.F., 1999.
Anti-estrogenicity of environmental polycyclic aromatic hydrocarbons
in human breast cancer cells. Toxicol. Sci. 133 (2 – 3), 115 – 127.
Berger, U., Oehme, M., 2000. Identification of derivatives of bisphenol
A diglycydyl ether and Novolac glycidyl ether in can coatings by
liquid chromatography/ion trap mass spectrometry. J. AOAC Int. 83,
1367 – 1376.
Berger, U., Oehme, M., Girardin, L., 2001. Quantification of derivatives of
bisphenol A diglycydyl ether (BADGE) and novolac glycidyl ether
(NOGE) migrated from can coatings into tuna by HPLC/fluorescence
and MS detection. Fresenius’ J. Anal. Chem. 369, 115 – 123.
Brucker-Davis, F., Thayer, K., Colborn, T., 2001. Significant effects of
endogenous hormonal changes in humans: considerations for low-dose
testing. Environ. Health Perspect. 109, 21 – 26.
Burke, M.D., Mayer, R.T., 1974. Ethoxyresorufin: direct fluorometric assay
of a microsomal-O-dealkylation which is preferably inducible by
3-methylchloanthrene. Drug Metab. Dispos. 2, 583 – 588.
Climie, I.J.G., Hutson, D.H., Stoydin, G., 1981. Metabolism of the epoxy
resin component 2,2-bis[4-(2,3-eopxypropoxy)phenyl]propane, the
diglycidyl ether of bisphenol A (DGEBPA) in the mouse: Part I. A
comparison of the fate of a single dermal application and of a single oral
dose of 14C-DGEBPA. Xenobiotica 11, 391 – 399.
Colborn, T., vom Saal, F.S., Soto, A.M., 1993. Developmental effects of
endocrine-disrupting chemicals in wildlife and humans. Environ. Health
Perspect. 101, 378 – 384.
Could, J.C., Leonard, L.S., Maness, S.C., Wagner, B.L., Conner, K.,
Zacharewski, T., Safe, S., McDonnell, D.P., Gaido, K.W., 1998.
Bisphenol A interacts with the estrogen receptor a in a distinct manner
from estradiol. Mol. Cell. Endocrinol. 142, 203 – 214.
Denizot, F., Lang, R., 1986. Rapid colorimetric assay fro cell growth and
survival. Modifications to the tetrazolium dye procedure giving
improved sensitivity and reliability. J. Immunol. Methods 89, 271 – 277.
Hoivik, D.J., Safe, S.H., Gaido, K.W., 1998. Effects of xenobiotics on
hormone receptors. In: Denison, M.S., Helferich, W.G. (Eds.),
Toxicant – Receptor Interactions. Taylor and Francis, Philadelphia,
pp. 53 – 68.
Kuiper, G.G.J.M., Lemmen, J.G., Carlsson, B., Corton, J.C., Safe, S.H., van
der Saag, P.T., van der Burg, B., Gustafsson, J.-Å., 1998. Interaction of
estrogenic chemicals and phytoestrogens with estrogen receptor h.
Endocrinology 139, 4252 – 4263.
Lavado, R., Thibaut, R., Raldua, D., Martin, R., Porte, C., 2004. First
evidence of endocrine disruption in feral carp from the Ebro River.
Toxicol. Appl. Pharmacol. 196, 247 – 257.
Letcher, R.J., van Holsteijn, I., Safe, S., Bergman, Å., Norstrom, R.J.,
Pieters, R., van den Berg, M., 1999. Cytotoxicity and aromatase
(CYP19) activity modulation by organochlorines in human placental
JEG-3 and JAR choriocarcinoma cells. Toxicol. Appl. Pharmacol. 160,
10 – 20.
Letcher, R.J., van der Burg, B., Brouwer, A., Lemmen, J., Bergman, Å., van
den Berg, M., 2002. In vitro anti-estrogenic effects of aryl methyl
sulfone metabolites of polychlorinated biphenyls and 2,2-bis(4-chlorophenyl)-1,1-dichloroethene on 17h-estradiol-induced gene expression
in several bioassay systems. Toxicol. Sci. 69, 362 – 372.
Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein
measurement with the Folin phenol reagent. J. Biol. Chem. 193,
265 – 275.
Navas, J.M., Segner, H., 2000. Anti-estrogenicity of h-naphthoflavone and
PAHs in cultured rainbow trout hepatocytes: evidence for a role of the
aryl hydrocarbon receptor. Aquat. Toxicol. 51 (1), 79 – 92.
Nicolas, J.-M., 1999. Vitellogenesis in fish and the effects of
polycyclic aromatic hydrocarbon contaminants. Aquat. Toxicol. 45,
77 – 90.
Niwa, T., Fujimoto, M., Kishimoto, K., Yabusaki, Y., Ishibashi, F., Katagiri,
M., 2001. Metabolism and interaction of bisphenol A in human hepatic
103
cytochrome p450 and steroidogenic CYP17. Biol. Pharm. Bull. 24,
1064 – 1067.
Olea, N., Pulgar, R., Pérez, P., Olea-Serrano, F., Rivas, A., Novillo-Fertrell,
A., Pedraza, V., Soto, A.M., Sonnenschein, C., 1996. Estrogenicity of
resin-based composites and sealants used in dentistry. Environ. Health
Perspect. 104, 298 – 305.
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, 167 – 174.
Pottenger, L.H., et al., 2000. The relative bioavailability and metabolism of
bisphenol A in rats is dependent upon the route of administration.
Toxicol. Sci. 54, 3 – 18 (and references therein).
Rankouhi, T., Sanderson, J.T., van Holsteijn, I., van Leeuwen, C., Vethaak,
A.D., van den Berg, M., 2004. Effects of natural and synthetic estrogens
and various environmental contaminants on vitellogenesis in fish
primary hepatocytes: comparison of bream (Abramis brama) and carp
(Cyprinus carpio). Toxicol. Sci. 81, 90 – 102.
Reel, J., Lamb IV, J., Neal, B., 1996. Survey and assessment of mammalian
estrogen biological assays for hazard characterization. Fundam. Appl.
Toxicol. 33, 288 – 305.
Rodgers-Gray, T.P., Jobling, S., Morris, S., Kelly, C., Kirby, S., Janbakash,
J., Harries, J.E., Waldock, M.J., Sumpter, J.P., Tyler, C.R., 2000. Longterm temporal changes in the estrogen composition of treated sewage
effluent and its biological effects on fish. Envrion. Sci. Technol. 34,
1521 – 1528.
Rouhani-Rankouhi, T., van Holsteijn, I., Letcher, R.J., Giesy, J.P., van den
Berg, M., 2002. The effects of pre-exposure with environmental and
natural estrogen on vitellogenin production in carp (Cyprinus carpio)
hepatocytes. Toxicol. Sci. 67, 75 – 80.
Rutten, A.A.J.J.L., Falke, H.E., Catsburg, J.F., Topp, R., Blaauboer, B.J.,
van Holsteijn, I., Doorn, L., van Leeuwen, R., 1987. Interlaboratory
comparison of total cytochrome P450 and protein determinations in rat
liver microsomes. Arch. Toxicol. 61, 27 – 33.
Safe, S., 1995. Modulation of gene expression and endocrine response
pathways by 2,3,7,8,-tetrachloro-p-dioxin and related compounds.
Pharmacol. Ther. 67, 247 – 281.
Safe, S., Wormke, M., 2003. Inhibitory aryl hydrocarbon receptor-estrogen
receptor a cross-talk and mechanisms of action. Chem. Res. Toxicol. 16
(7), 807 – 816.
Sanderson, J.T., Seinen, W., Giesy, J.P., van den Berg, M., 2000. 2-Chloros-triazine herbicides induce aromatase (CYP19) activity in H295R
human adrenocortical carcinoma cells: a novel mechanism for estrogenicity? Toxicol. Sci. 54, 121 – 127.
Sanderson, J.T., Letcher, R.J., Heneweer, M., Giesy, J.P., van den Berg, M.,
2001. Effects of chloro-s-triazine herbicides and metabolites on
aromatase activity in various human cell lines and on vitellogenin
production in male carp hepatocytes. Environ. Health Perspect. 109,
1027 – 1032.
Sanderson, J.T., Boerma, J., Lansbergen, G.W.A., van den Berg, M., 2002.
Induction and inhibition of aromatase (CYP19) activity by various
classes of pesticides in H295R human adrenocortical carcinoma cells.
Toxicol. Appl. Pharmacol. 182, 44 – 54.
Smeets, J.M.W., Rouhani-Rankouhi, T., Nichols, K.M., Komen, H.,
Kaminsky, N.E., Giesy, J.P., van den Berg, M., 1999. In vitro
vitellogenin production by carp (Cyprinus carpio) hepatocytes as a
screening method for determining (anti)estrogenic activity of xenobiotics. Toxicol. Appl. Pharmacol. 157, 68 – 76.
Smeets, J.M.W., van Holsteijn, I., Giesy, J.P., van den Berg, M., 1999. The
anti-estrogenicity of Ah receptor agonists in carp (Cyprinus carpio)
hepatocytes. Toxicol. Sci. 52, 178 – 188.
Sole, M., Barcelo, D., Porte, C., 2002. Seasonal variation of plasmatic and
hepatic vitellogenin and EROD activity in carp, Cyprinus carpio, in
relation to sewage treatment plants. Aquat. Toxicol. 60, 233 – 248.
Tolar, J.F., Mehollin, A.R., Douglas-Watson, R., Angus, R.A., 2001.
Mosquitofish (Gambusia affinis) vitellogenin: identification, puri-
104
R.J. Letcher et al. / Toxicology and Applied Pharmacology 209 (2005) 95 – 104
fication, and immunoassay. Comp. Biochem. Physiol. Part C 128,
237 – 245.
Wormke, M., Castro-Rivera, E., Chen, I., Safe, S., 2000. Estrogen and
aryl hydrocarbon receptor expression and crosstalk in human
Ishikawa endometrial cancer cells. J. Steroid Biochem. Mol. Biol.
72 (5), 197 – 207.
Yamanaka, S., Arizono, K., Matsuda, Y., Soyano, K., Urushitani, H.,
Iguchi, T., Sakakibara, R., 1998. Development and application of
an effective detection method for fish plasma vitellogenin induced
by environmental estrogens. Biosci. Biotechnol. Biochem. 62,
1196 – 1200.
Yoshihara, S., Makishima, M., Suzuki, N., Ohta, S., 2001. Metabolic
activation of bisphenol A by rat liver S9 fraction. Toxoicol. Sci. 62,
221 – 227.